Beruflich Dokumente
Kultur Dokumente
COGNITIVE NEUROSCIENCE
Developmental Cognitive Neuroscience
Mark Johnson and Bruce Pennington, editors
Edited by
Charles A. Nelson
Monica Luciana
A BRADFORD BOOK
THE MIT PRESS
CAMBRIDGE, MASSACHUSETTS
LONDON, ENGLAND
© 2001 Massachusetts Institute of Technology
QP363.5.H365 2001
612.8'2—dc21 00-046567
CONTENTS
Foreword ix
JOHN T. BRUER
Introduction xi
I. FUNDAMENTALS OF DEVELOPMENTAL
NEUROBIOLOGY 1
CONTENTS V
II. METHODOLOGICAL PARADIGMS 107
VI CONTENTS
V. LANGUAGE 267
23. The Functional Development and Integration of the Dorsal and Ventral Visual
Pathways: A Neurocomputational Approach 339
MARK H.JOHNSON, DENIS MARESCHAL, AND GERGELY CSIBRA
29. A Model System for Studying the Role of Dopamine in the Prefrontal Cortex
during Early Development in Humans: Early and Continuously Treated
Phenylketonuria 433
ADELE DIAMOND
CONTENTS Vll
VII. NEURODEVELOPMENTAL ASPECTS OF CLINICAL
DISORDERS 489
35. Tics: When Habit-Forming Neural Systems Form Habits of Their Own 549
JAMES F. LECKMAN, BRADLEY S. PETERSON, ROBERT T. SCHULTZ,
AND DONALD J. COHEN
Contributors 663
Index 665
Vlll CONTENTS
FOREWORD
FOREWORD IX
This handbook also illustrates how the brain sciences contribute to cognitive
neuroscience. Neuropsychology has a long history of the study of how brain trauma
affects behavior. The power of studying normal versus impaired brains, both mature
and developing, is evident in the handbook's section on neurodevelopmental as-
pects of clinical disorders.
Basic developmental neurobiology has elucidated developmental phenomena
such as corticogenesis, neurogenesis, and synaptogenesis, as well as the role of
critical and sensitive periods in development. There have also been attempts to
discern how such phenomena relate to human behavior. Combining basic neuro-
biology with systems neuroscience, imaging, and cognitive science will no doubt
expedite our understanding of how changes at the cellular and subcellular levels
affect and are affected by experience. Some of the most recent, and exciting, find-
ings in basic neurobiology establish that the human brain remains malleable and
plastic throughout our lifetimes. One of the great opportunities in developmental
cognitive neuroscience, as this handbook indicates, is to explore what this
new understanding of lifelong, neural plasticity implies for brain and child
development.
Recently, claims about what brain science might mean for parenting and early
childhood have captured the public's imagination. We should encourage the pub-
lic's interest in developmental brain science and applaud attempts to base early
childhood policy and practice on a scientific basis. However, in some instances
public enthusiasm far outstrips our scientific understanding. Too often the mes-
sages broadcast by advocates and the media do not accurately reflect what scientists
currently know about synapses, critical periods, neural plasticity, and how experi-
ence affects the brain. If we wish to base policy and practice on brain science or
cognitive neuroscience, it is imperative that scientists knowledgeable in the relevant
disciplines engage in careful discussion of what their disciplines, alone and collec-
tively, can contribute to better policy and practice and articulate their conclusions
to the interested public. This handbook provides an excellent starting point for
that discussion.
The power, promise, and importance of the attempt to understand the biological
foundations of human cognition, are amply illustrated throughout the handbook's
chapters.
John T. Bruer
X FOREWORD
INTRODUCTION
INTRODUCTION XI
Through the late 1980s into the 1990s, the field of cognitive neuroscience took
hold. The Summer Institutes continued and flourished, the Cognitive Neuroscience
Society and the Journal of Cognitive Neuroscience were launched, and psychology de-
partments throughout the United States began to advertise for faculty positions in
this field. However, interest in the neural underpinnings of development occurred
very slowly. There are many reasons for this trend (for discussion, see Nelson and
Bloom, 1997), but suffice to say that change began to occur on three fronts. One
was that through the 1980s the field of developmental neuroscience began to yield
discoveries of profound importance (e.g., the pattern of synaptogenesis; the role
of experience in influencing brain development). A second was that a number of
prominent cognitive neuroscientists who had historically studied mature function
acknowledged the importance of studying development (e.g., in the mid 1990s a
meeting of the Cognitive Neuroscience Society devoted a symposium to develop-
mental cognitive neuroscience). Finally, a new generation of developmental psy-
chologists began to recognize the need for considering the neurobiological
underpinnings of behavioral development (again, see Nelson and Bloom, 1997 for
discussion). Collectively, then, the time was right for the field of developmental cog-
nitive neuroscience to be born (approximately 16 years after Michael Gazzaniga's
first Handbook of Cognitive Neuroscience was published).
This Handbook of Developmental Cognitive Neuroscience represents the distillation of
the best this new field has to offer; it also reflects a number of strong biases by the
editors. One such bias is that the field of developmental cognitive neuroscience
must be grounded in basic neurodevelopmental science, particularly developmen-
tal neurobiology. To this end, the first part of this volume (Fundamentals of De-
velopmental Neurobiology) is devoted to basic studies and principles of neural
development. Here the reader learns about pre- and postnatal neurogenesis, syn-
aptogenesis, and myelination; the effects of sex hormones on brain development;
and about development of the hippocampus and prefrontal cortex in particular
(given the importance of these regions for cognitive development). A second bias
is our emphasis on the importance of methodological advances. Thus, the second
part of the volume (Methodological Paradigms) is devoted to describing methods
that have proved so important in elucidating brain-behavior relations in the con-
text of cognitive development. These methods include behavioral "marker" tasks,
along with event-related potentials (ERPs), functional Magnetic Resonance Imag-
ing (fMRI), and genetic and computational (neural network) modeling.
Over the past several years the area of neural plasticity has received tremendous
attention by both neuroscientists and behavioral scientists. Indeed, the forces that
shape and mold the brain may well represent the "new" developmental psychology,
albeit a more mechanistic and reductionistic version than offered by previous gen-
erations of developmental psychologists and one that emphasizes development
within a lifespan context. That is, the forces that mold the brain's structure and
physiology are now recognized to operate well into adulthood (see Tanapat, Has-
tings, and Gould, chapter 7, this volume). To this end, the third part (Neural Plas-
ticity of Development) is devoted to a discussion of this area, emphasizing both
normative and atypical aspects of development.
Because the development of sensory and sensorimotor systems and language has
played such a prominent role in both neuroscience and in developmental psy-
chology, the fourth (Sensory and Sensorimotor Systems) and fifth (Language) parts
of this volume are devoted to this area. Here we are enlightened about the devel-
opment of the visual and auditory systems, the development of skilled motor move-
ments, and several aspects of language development. Embedded within these
chapters is yet another bias of the editors: the need to juxtapose the study of nor-
xii INTRODUCTION
mative development against that of atypical development, as each mutually informs
the other.
The sixth part of the book (Cognition) reflects the substance of the volume, as
befits a book on cognitive neuroscience. Here we provide chapters on the develop-
ment of attention, memory, face/object recognition, spatial cognition, number
comprehension, and executive functions. Naturally, the work captured by the dis-
tinguished authors of these chapters is framed in a neuroscience context.
The seventh part of this volume (Neurodevelopmental Aspects of Clinical Dis-
orders) returns the reader to the theme of how studies of atypical development can
inform the study of typical development. However, these chapters directly target
studies of atypical populations. We introduce the reader to the importance of nu-
trition on brain-behavioral development, on the risks of prenatal alcohol and co-
caine exposure, on the development of autism, Tourette's syndrome, and
schizophrenia, and on disorders of attention.
The eighth and final part of the book (Emotion and Cognition Interactions)
anticipates what may well represent the next cutting edge area, that of develop-
mental affective neuroscience and the manner in which affective experience may
act to shape cognitive processes. This part begins with a tutorial on the neurobiology
of attachment, and progresses to a discourse on the effects of deprivation on emo-
tional development, on the neurobiology of temperament, and on reward-seeking
behavior.
Overall, this volume possesses the breadth and depth necessary to do justice to
this exciting new scientific field. It does so by drawing on internationally known
experts who have graciously contributed their time and energy to bring the reader
the first complete Handbook of Developmental Cognitive Neuroscience. We hope that you
are as pleased with the result as we are.
Charles A. Nelson
Monica Luciana
REFERENCES
Hebb, D. O., 1949. The Organization of Behavior. New York: Wiley Press.
Nelson, C. A., and F. E. Bloom, 1997. Child development and neuroscience. Child Devel.
68:970-987.
Posner, M. I., S. E. Petersen, P. T. Fox, and M. E. Raichle, 1988. Localization of cognitive
operations in the human brain. Science 240(4859) :1627-1631.
Lectures will be in Room 1 of William James Hall, which is located at 33 Kirkland Street (at
the corner of Kirkland and Divinity Avenue) in Cambridge. Laboratories and discussion
groups will meet in smaller rooms (to be announced) within William James Hall. Unless
otherwise noted, lunch and dinner will be open; a booklet describing the many and varied
Harvard Square restaurants will be distributed at the Institute. Coffee and donuts will be
available beginning at 8:15 A.M. in Room 105 of William James Hall (off the lobby), and soft
drinks and cookies will be provided in the laboratory rooms in the afternoon.
Prior reading:
Cognition (2nd ed.). A. L. Glass and K. Holyoak (New York. Random House).
Chapters 1, 2, 4, and 9.
INTRODUCTION Xlll
Parallel Distributed Processing. D. E. Rumelhart andj. McClelland
(Cambridge, MA: MIT Press).
Chapters 1-4; 16 and 17 optional.
Principles of Behavioral Neurology. M.-M. Mesulam (Philadelphia: F. A. Davis).
Chapters 3, 4, and 7.
Neurobiology (2nd ed.). G. M. Shepherd (New York: Oxford University Press).
Chapters 16, 29, and 30.
WEEK 1: MEMORY
Monday, 13 June
Tuesday, 14 June
Wednesday, 15 June
xiv INTRODUCTION
Dinner
Thursday, 16 June
Friday, 17 June
INTRODUCTION XV
Tuesday, 21 June
9:00 The anatomy and physiology of the posterior parietal lobe
(Richard A. Andersen)
10:30 The functions of the visual-spatial attention system
(Anne M. Treisman)
Lunch
2:00 Demo A connectionist/control architecture for attention and skill
acquisition: From potential physiology to behavior
(Walter Schneider)
3:00 Lab Studying visual attention: Pop-out, illusory conjunctions
(William Prinzmetal)
4:00 Hands-on demonstrations of visual attention phenomena; small
group discussion
5:00 General review and discussion
Dinner
8:00 Anatomical, physiological, and psychophysical evidence for separate
pathways for form, color, movement and stereo (David F. L. Hubel)
Wednesday, 22 June
9:00 Neurological patients as an approach to studying spatial attention
(M. Marsel Mesulam)
10:30 Cognitive studies of neurological deficits (Michael I. Posner)
Lunch
2:00 Demo Neuropsychological examination of patients (William P. Milberg)
3:00 Demo Anatomical methods for brain studies (M. Marsel Mesulam)
4:00 Lab Hands-on demonstrations: Cerebral laterality experiments (Stephen
M. Kosslyn and John D. E. Gabrieli)
5:00 General review and discussion
Thursday, 23 June
9:00 The visual agnosias (Elizabeth K. Warrington)
10:30 The anomias (Harold Goodglass)
Lunch
3:00 Demo Neural basis of cortical maps (Gordon M. Shepherd and John S.
Kauer)
4:00 Small group discussion
5:00 General review and discussion
Friday, 24 June
9:00 Visual object recognition (Irving Biederman)
10:30 Visual imagery and visual perception (Stephen M. Kosslyn)
Lunch
2:00 Lab A computer simulation of neurological disorders of high-level visual
recognition processes (Stephen M. Kosslyn and Gretchen Wang)
4:00 Hands-on experimentation with computer simulation of high-level
visual recognition syndromes
5:00 General review and discussion
8:00 Banquet
After dinner: Brief summation by Michael I. Posner, Stephen M. Kosslyn, and Gordon M.
Shepherd
xvi INTRODUCTION
I
FUNDAMENTALS OF
DEVELOPMENTAL
NEUROBIOLOGY
This page intentionally left blank
1 Neocortical Neuronogenesis:
Regulation, Control Points, and
a Strategy of Structural Variation
T. TAKAHASHI, R. S. NOWAKOWSKL
AND V. S. CAVINESS, JR.
ABSTRACT Cognitive and adaptive behaviors both across and ture across and within species (Hahn, Jensen, and
within mammalian species vary in association with variation in Dudek, 1979). That is, the study of the evolutionary and
neocortical structure. Here we present an algorithm of neo- ontogenetic origins of these structural variations is a
cortical histogenetic operation, based upon experiments in
the mouse with an emphasis upon neuronogenesis, that de- cornerstone of cognitive neuroscience.
fines control points at which large-scale revisions may be as- The general structural features of the neocortex are
sociated with speciation and small-scale revisions with the greatly regular across mammalian species. In all species,
structural distinctions within species. We suggest that the mas- neurons with long axons, which are excitatory, group
ter control mechanism for the neuronogenetic sequence oc- broadly into five classes which are arrayed tangentially
curs at the Gl restriction point and that it is at this control
point that both large- and small-scale variations in numeration by class into regular laminar order (figure 1.1 A; table
are coordinated with those of specification of neuronal class 1.1; Cajal, 1952; Lorente de No, 1938; Peters and Jones,
and regional specification. Lower amplitude and regionally 1984). These general classes are, by layer, the small pyr-
diverse modulations, typical of within-species differences, will amids of layer II, the medium-sized pyramids of layer
also arise from modulation of cortical development down- III, the granule cells of layer TV, the large pyramids of
stream from neuronogenesis.
layer V, and the polymorphic neurons of layer VI.
Inhibitory interneurons, a minority population, distrib-
Introduction ute without laminar grouping more or less uniformly
throughout the height of the cortex (Lavdas, Mione,
The neocortex is the central and indispensable "organ" and Parnavelas, 1996; Peters and Jones, 1984). The gen-
of cognitive process. It is by far the largest structure eral laminar structure varies regionally in terms of these
of the brain and is bilaterally represented in the cere- cytologic details as well as the relative numbers and the
bral hemispheres of the mammalian forebrain. It is packing density of the separate neuronal classes. These
bounded at the medial and ventral margin of the hemi- variations have represented the principal criteria for the
sphere by the hippocampal formation and amygdaloid regional parcellation of the neocortex into architec-
nucleus and rostrally and ventrally by piriform cortex, tonic fields that form the basis for understanding the
the diagonal band, and septal structures. To the extent multimodal and hierarchic neocortical representations
that "cognition follows form," the range and quality of of neural and cognitive systems (Brodmann, 1909; Cav-
cognitive processes, as these vary across and within spe- iness, 1975; Zilles, 1990). Thus, the generic neocortical
cies, are correlates of the variations in neocortical struc- map places perceptual (visual, acoustic, somatosensory,
and gustatory representations) in constant positions
relative to each other in the posterior and lateral
T. TAKAHASHI Department of Neurology, Massachusetts
General Hospital, Harvard Medical School, Boston, Massachu-
regions of the hemisphere and motor representations
setts; Department of Pediatrics, Keio University School of anteriorly (figure LIB).
Medicine, Tokyo, Japan. The general similarity in organization of the neocor-
R. s. NOWAKOWSKI Department of Neuroscience and Cell Bi- tex of all mammals stands in marked contrast to the
ology, UMDNJ-Robert Wood Johnson Medical School, Piscat- variable features that differentiate the neocortex from
away, New Jersey.
v. s. CAVINESS, JR. Department of Neurology, Massachusetts species to species. Most notable among these variable
General Hospital, Harvard Medical School, Boston, features is neocortical volume or, more specifically,
Massachusetts. surface area (table 1.2; Haug, 1987; Holloway, 1974;
quence is initiated with the generation of neurons, or 1990, 1992; Finlay and Pallas, 1989; Finlay and Slattery,
neuronogenesis, which occurs in a pseudostratified ven- 1983; Spreafico et al., 1995; Thomaidou et al., 1997;
tricular epithelium (PVE; for abbreviations, see table Verney et al., 1999). Surviving neurons form synapses
1.3) at the margins of the forebrain ventricle. Prior to with other neurons, grow, and further differentiate as
the formation of the earliest neurons, the epithelium they become integrated into neural circuitry (Bour-
proliferates exponentially (Rakic, 1988) and is already geois, Jastreboff, and Rakic, 1989; Bourgeois and Rakic,
partitioned regionally with reference to expression do- 1993; Rakic, 1995; Rakic, Bourgeois, and Goldman-
mains of a diversity of transcription factors (Puelles Rakic, 1994). The progression of the sequence, ex-
and Rubenstein, 1993; Shimamura et al., 1995). In all pressed in terms of the origin of successive neocortical
species, once neuronogenesis is initiated, postmitotic layers, is scaled identically to the overall time course of
neurons migrate across the width of the cerebral wall, neuronogenesis in each species. As indicated in figure
where they become sorted by class into layers (Rakic, 1.3, progression through this temporal interval is ex-
1972; Takahashi et al., 1999b). A substantial number are pressed as a percentile of the duration of the interval
then eliminated by histogenetic cell death (Ferrar et al., (Caviness, Takahashi, and Nowakowski, 1995).
These commonalities of sequence and scaling of se- tures that distinguish species. Yet, in order to assure
quence progression suggest that histogenesis is regu- characteristic constancy of structure among individuals
lated by similar mechanisms across species (Caviness, of a species, the execution of these molecular and cel-
Takahashi, and Nowakowski, 1995). Therefore, it can lular biological events must be greatly regular and stably
be inferred from comparative study that regulatory buffered in the face of modest variations in fetal geno-
mechanisms operate at three levels: (1) a general mam- type and developmental conditions that must exist be-
malian program, which assures the generic events of tween individuals (Kennedy et al., 1998). With respect
specification and histogenesis; (2) a species-specific to neocortical development, the nature of variations in
variation of the generic histogenetic program, which histogenetic process associated with species-specific dif-
assures specific structural adaptations; and (3) modu- ferences is not readily tackled directly by experiment,
latory mechanisms within species, which determine On the other hand, the workings of histogenetic pro-
structural variations distinctive to the individual. cess may be elucidated through experiment as it pro-
The execution of the histogenetic sequence must de- ceeds in a given species. Through experiment in a
pend upon a great number of interdependent molec- species, one may construct a specific program or algo-
ular and cellular biological events (Gerhart and rithm of the generic sequence in its detailed operations
Kirschner, 1997). These must vary in systematic ways and with respect to control points that regulate the
across species to provide for the distinct structural fea- whole sequence. In principle, such an algorithm and,
Ill of the neocortex (Takahashi et al., 1996, 1999b). It opmentally primitive PVE in the course of the
is significant that the neighborhood relationships of the proliferative process (Miyama et al., 1997; Rakic, 1988).
proliferative population generally determine the neigh- In summary, the advance of the neuronogenetic
borhood relationships of the postmigratory population sequence within all regions of the PVE is ordered tem-
within the cortex (figure 1.4), and this appears to be a porally according to the TNG and spatially by the to-
particularly stringent rule for the cells destined for lay- pographic organization of the PVE. This integrated
ers V and VI (Kornack and Rakic, 1995; Misson et al., operation is the principle through which the two-
1991; Walsh and Cepko, 1992a,b, 1993). This observa- dimensional (rostrocaudal and mediolateral) structure
tion is only one component of the body of evidence of the PVE is "translated" upon what becomes the lam-
suggesting that a protomap of the neocortical architec- inate—i.e., three-dimensional—structure of the neo-
tonic map is in some way represented within the devel- cortex. Our investigations, conducted in the mouse,
information is encoded as they arise with cycles 1-7. We Alternatively, they might modulate the encoding
envision the following operating properties of such a process within the territory of overlap. The process
mechanism. At any given time there are multiple CC would be repeated as each TE leaves the origin so that
domains spanning the TNG in overlapping fashion (fig- successive waves of parcellation would lead to an in-
ure 1.7). The critical mechanism complementary to do- creasingly atomized mosaic of map units within the
main propagation, we suggest, would be an encoding PVE. We suggest that this mechanism of parcellation
signal that emanates from origin as each LE passes would extend the relatively coarse-grained parcellation
through the origin and then is propagated rapidly (i.e., implicit in the domains of distribution of transcription
traveling across the PVE in a time much less than a factors, already instantiated in the PVE by the outset of
single cell cycle) down the TNG. One possible mecha- the neuronogenetic interval (Puelles and Rubenstein,
nism for the propagation of synchronizing signal is via 1993; Shimamura et al., 1995). To the extent that such
gap junctions between cells of the PVE (Bittman et al., a mechanism might serve to "translate" a CC domain
1997). The borders of a "mapping unit" encoded by a into an architectonic area, there will be linkage between
given CC domain could correspond to its LE and trail- the temporospatial domain (the set of Tcs plus the
ing edge (TE) where they abut the TE and LE of do- TNG) and the cytoarchitectonic subdivisions of the ma-
mains of CC V _ 2 and CCN+2t respectively (figure 1.7). ture neocortex. Such specification of a unit of the
tions 1.1, 1.2, and 1.3). An increase in k greater than internal organs, including the brain. It is pertinent to
0.009 would, via the same logic, produce a cortex the argument that the enlargement occurs as a conse-
smaller than that of mouse. These changes in size would quence of increase in cell number (Fero et al., 1996;
apply to the entire neocortex and do so uniformly. This Kiyokawa et al., 1996; Nakayama et al., 1996). As far as
is because the number of neurons produced from is known, p27 operates only within the cell cycle at the
equivalent founder populations will be identical as the restriction point, and the inference from the knockout
neuronogenetic sequence is enacted sequentially along is that the "computation" governed by p27 is to deter-
the TNG and under the regulation of Q. mine the gating of probability that cells will pass the Gl
Thus, a single change in k, induced at the Gl restric- restriction point. Importantly, the p27-governed com-
tion point, could produce the step changes in neocor- putation can be only partially rate-limiting; otherwise,
tical magnitude that are necessary for speciation. This p27 knockout animals would be infinitely large. Other
is illustrated by the consequences of knockout of the cycle progression inhibitors acting at the restriction
p27 gene in mice, which results in an animal that is point, for example p21, must complement the action of
some 30-40% larger than animals with the wild-type p27. These may include inhibitors that act globally, such
gene. The enlargement in this animal is a more-or-less as p27, but also others which are strictly regional in their
uniformly scaled increase in body size and in (most) jurisdictions. Note that the actions of either restriction
FIGURE 1.9 Schema correlating variation in the growth of the (Dl, D2), the PVE would be partitioned into a smaller number
PVE and progression of cell cycle domains. Depending upon of CC domains. With greater pre-NI expansion of the PVE
the number of cycles antedating the onset of neuronogenesis, (A2) giving rise to a larger PVE (B2), the same cell cycle length
the PVE might enlarge moderately (Al) or greatly (A2) prior as that in mouse would result in more CC domains (C2). With
to the onset of neuronogenesis (pre-NI expansion of PVE). greater pre-NI expansion but with longer cell cycle length the
Once neuronogenesis is initiated (Bl, B2) in the far rostro- PVE would be larger PVE but partitioned in the same number
lateral region of the PVE with cell cycle 1 ("CCj begins"), the of CC domains (D2). (The theoretical consequences of other
PVE will become partitioned into cell cycle domains (Cl, C2, possible patterns of variation in Tc, i.e., shorter cell cycle
Dl, D2). In mouse (Cl) it is estimated that five cycle domains length, and preneuronogenetic size of PVE are not illus-
will be established before the domain of CCj invades and re- trated.) Mechanisms postulated here might lead to an in-
places completely the last vestige of the preneuronogenetic creased number of mapping units or, alternatively, assure a
PVE at the caudomedial extremity of the PVE (see also figure common mapping pattern across the range of small to large
1.7). In the specific instances where cell cycle length is longer rodents, for example, from mouse to beaver.
ABSTRACT We have identified five distinct phases of synap- The formation of synaptic contacts can be described
togenesis in the cerebral cortex of primates from conception at two distinct levels. On the one hand, one can describe
to death. The first two phases, with a low density of synapses, the assembly of the numerous molecular components
occur during early embryonic life. They are followed by a third
phase of very rapid accumulation of synapses around birth. building up the pre- and postsynaptic domains of the
During the fourth phase the density of synapses is maintained neuron. On the other hand, one can describe the de-
at a very high level from the early postnatal period through velopment of ensembles of synaptic contacts in the
puberty. This is followed by a true loss of synapses during pu- cortical neuropil. Putting these two aspects of synapto-
berty. The fifth phase begins after sexual maturity and extends
into old age. It is characterized by a slow decline in the density
genesis in parallel is not trivial because the synaptic con-
of synapses. Experimental observations indicate that the early tact constitutes a crucial point of articulation between
phases of synaptogenesis are robust neurodevelopmental pro- the cellular properties of the single cortical neuron and
cesses determined by mechanisms intrinsic and common to the neurophysiological functions associated with large
the whole cortical mantle. The various postnatal phases coin- ensembles of these cortical neurons, which are under
cide with final adjustments of many aspects of synaptoarchi-
tectony, depending on normal functional inputs. They overlap distinct categories of constraints:
with critical periods for diverse cortical functions until pu- 1. The constraints related to the multiple intracellu-
berty. Inside each of these phases, distinct classes of synapses
appear in successive waves of synaptogenesis, providing the lar mechanisms control the type, amount, and distri-
first examples of a long list of discrete synaptogenetic events bution of presynaptic active zones, as well as
we are just beginning to explore. postsynaptic densities differentiated on the cell surface
of a cortical neuron. Are these constraints all under
The development of the neocortex in mammals is a strict genetic control?
highly orchestrated cascade of histological events in- 2. The constraints related to intercellular mecha-
cluding, successively, the generation and differentiation nisms are linked to the neurophysiological functions of
of neurons (Rakic, 1995), the navigation and organi- large ensembles of these cortical neurons and control
zation of the axonal projections between ensembles of
the development of the topological distribution of the
neurons (Barone et al., 1995), then the formation and
synaptic contacts in the neuropil. Are these constraints
maturation of the synaptic contacts that constitute the
major final steps of corticogenesis (Bourgeois, Goldman- all under strict epigenetic control?
Rakic, and Rakic, 1994, 2000; Bourgeois and Rakic, The interactions between genetically defined factors
1993; de Felipe et al., 1997; Huttenlocher and Dabhol- and experientially modifying factors are crucial in the
kar, 1997). All these biological events ultimately lead to development and morphofunctional maturation of the
individuation, the process by which individual mam- neocortex, in which synaptogenesis is a key event.
mals become differentiated in their societies. Knowing the exact timing of synaptogenesis in the cas-
As reviewed by Zoli and colleagues (1998), intercel- cade of histological events will provide a neuroanatom-
lular communications in the neocortex can occur ei- ical basis for the experimental identification of these
ther via volume transmission or by wiring transmission. interactions through the time course of development
This chapter describes the development of the wiring and maturation. The description of formation of en-
transmission via synaptic contacts. sembles of synaptic contacts is currently less explored
J.-P. BOURGEOIS Laboratoire de Recepteurs et Cognition,
than the description of the mechanisms of assembling
Departement des Biotechnologies, Institut Pasteur, Paris, of synaptic macromolecules. We tried to re-equilibrate
France. this. The neocortex of the rhesus monkey provides an
Total
Phase 3 Number of Total
Puberty Duration Density of Synapses per Cortical Number of
Gestation Onset (onset-end) synapses1 Neuron Areas Visual Areas
Rat 21 days 82 DAC (2 14 days 320-946 12,500-13,500 21 4-6
months (P2-P16)
after birth)
Cat 65 days 248 DAC (6 30 days 276-406 5800-9300 30-50? 12-17
months (P9-P39)
after birth)
Macaque 165 days 1260 DAC (3 136 days 276-620 2000-5600 72 25
years after (E90-P61)
birth)
Human 280 days 4660 DAC 470 days 350 6800-10,000 200? 50?
(12 years (E120-P310)
after birth)
1
Density of synapses is expressed in millions per cubic millimeter of cortical tissue.
DAC = days after conception; E = embryonic day; P = postnatal day.
Reproduced from Bourgeois (1997) with permission from Scandinavian University Press.
ger, and Greenough, 1992) and diffusible or signaling natal developmental events, while complete maturation
molecules (Edlund and Jessel, 1999). This heteroge- of cortical functions is protracted until puberty. Al-
neity may also be functional: In different cortical areas though kinetics is not the whole story of synaptogenesis,
neurons fire with different temporal patterns (Ferster our identification of distinct phases of synaptogenesis
and Spruston, 1995) and different conduction times provides a new developmental frame for the description
(Salin and Bullier, 1995). In the morphofunctional hy- of the establishment and successive adjustments of syn-
pothesis all these heterogeneities increase the number aptoarchitectony in neocortex of primates. The density
and the complexity of cellular interactions between of synapses is the same at birth and after puberty; how-
nerve endings during the formation of synapses, their ever, the developmental constraints on synaptogenesis
selective stabilization, their elimination, or all these are expected to be totally different. We observed that
steps. This, along with an increased number of waves of the onset and time course of phase 3 of synaptogenesis
synaptogenesis, might increase the duration of phases are very robust developmental events, while influences
3 and 4 of synaptogenesis. from environment increase thereafter. Among multiple
The morphofunctional hypothesis leads to another mechanisms, this plasticity is related to the high vari-
falsifiable experimental prediction: The anatomical or ability in the density, localization, assembly, and nature
functional suppression of a specified number of pre- of many types of molecules in the pre- and postsynaptic
striate cortical areas should shorten the duration of domains of neurons. To cite only a few examples, one
phase 3 of synaptogenesis observed in the primary vi- should refer to the large diversity and plasticity of the
sual cortex. Conversely, the addition of cortical areas subunits of classical pharmacological receptors (Chan-
should prolong phase 3. geux et al., 1997; Craig 1998; Vannier and Triller, 1997)
These two hypothesis are not mutually exclusive: The as well as to new families of molecules such as class I
two classes of mechanisms might be at work at different MHC (Corriveau, Huh, and Shatz, 1998) and synaptic
moments of phase 3 or/and for distinct ensembles of cadherins (Hagler and Goda, 1998; Tanaka et al., 2000),
synapses. These mechanisms are expected to be iden- which have been observed during formation of synap-
tical in the whole cortical mantle. tic contacts. These numerous structural and signal-
transducing molecules assemble and deassemble in
New perspectives on genetics of synaptogenesis, highly dynamic pre- and postsynaptic domains of mor-
evolution of epigenesis, and individuation phologically permanent synaptic contacts.
We hypothesize that, upstream of all these molecular
In neocortices of nonhuman and human primates, neu- events, new families of genes control the onset and
rogenesis and synaptogenesis are very precocious pre- duration of the different phases of synaptogenesis.
ABSTRACT Myelination in the human central nervous system weeks of gestation in the spinal cord and continues well
(CNS) is a complex but orderly process, occurring in predict- into the third and fourth decades of life in the intra-
able topographical and chronological sequences. CNS myeli- cortical fibers of the cerebral cortex; but the most
nation begins as early as 12-14 weeks of gestation in the spinal
cord and continues well into the fourth decade of life. The important and dramatic changes occur between mid-
most significant period of CNS myelination, however, occurs gestation and the end of the second postnatal year, with
between midgestation and the second postnatal year. Myeli- myelination accounting in large part for the large gain
nation can be studied by various methods, including myelin- in brain weight, which more than triples during this
stained histopathological brain sections and, more recently, period. The process slows markedly after 2 years of age
magnetic resonance (MR) imaging. The latter constitutes the
focus of this chapter. (Yakovlev and Lecours, 1967).
Fluid, on the other hand, is ideally imaged on T2- myelination provides an excellent parameter of brain
weighted images: The ventricles are bright. maturation.
On MR imaging, brain maturation during the first 2 The changes in signal intensity seen on Tl- and T2-
years of life consists primarily of signal changes second- weighted images closely parallel the known pattern of
ary to the process of myelination and occurs at different myelination as determined by histochemical studies. At
rates and at different times on different types of images. birth, the dorsal pons, portions of the superior and in-
Changes of brain signal due to myelination will ap- ferior cerebellar peduncles, the decussation of the su-
pear first on Tl-weighted images as increased signal perior cerebellar peduncles, the posterior limb of the
relative to gray matter. Eventually, a corresponding internal capsule, the central corona radiata, the optic
drop of T2 signal will be seen. In general, changes in chiasm, the optic tract, the ansa lenticularis, and the
white matter maturation are best seen on Tl-weighted fibers originating in the external segment and travers-
images during the first 6-8 months of life and on the ing the internal segment of the globus pallidus are par-
T2-weighted images between the ages of 6 and 18 tially myelinated. These are seen as areas of increased
months (Barkovich et al., 1988). Most recently, a new signal intensity on the Tl-weighted images and as de-
MR imaging technique, diffusion-weighted imaging, creased signal intensity on the T2-weighted images of
has been used for assessment of brain maturation in newborns (figures 3.1, 3.2, and 3.3).
humans (Huppi et al., 1998; Takeda et al., 1997). On Tl-weighted images, unmyelinated white matter
Diffusion-weighted imaging is sensitive to brownian mo- is hypointense relative to gray matter. With myelination,
tion of water. Acute injury to the brain, such as edema phospholipids occupy the space otherwise occupied by
and/or cytotoxicity, results in restricted brownian mo- interstitial water and the Tl signal increases, from hy-
tion, and hence altered diffusion, of extracellular water. pointense to hyperintense, relative to gray matter. On
Similarly, on diffusion-weighted images, it is possible to T2-weighted images, the reverse picture can be seen,
map brain maturation, as unbound water tends to fol- with unmyelinated white matter being hyperintense
low pathways of least resistance. With myelination, water relative to gray matter. As the white matter matures, its
is more likely to follow an axonal bundle than to cross T2 signal decreases from hyperintense to hypointense
it. This so-called diffusion anisotropy becomes apparent relative to gray matter, passing through a phase of iso-
within white matter tracts. In fact, the signal changes intensity to gray matter.
on diffusion-weighted images may precede correspond- The causes of the Tl shortening (increase in Tl sig-
ing changes on conventional Tl-weighted images. Ac- nal intensity) associated with myelination are related to
tive myelination from the perinatal period onward can but, at least in part, independent from the causes of T2
be identified by imaging, and thus the progression of shortening (decrease in T2 signal intensity). As already
mentioned, brain maturation occurs at different rates myelination with even a temporary decline in brain
and times on Tl-weighted and T2-weighted images. The weight. It is, however, probable that the changes in Tl
discrepancy is mild in certain areas (2 months in the and T2 values are due not only to a decreasing water
corpus callosum) but marked in others (6-10 months content of the brain, but also to a change in biochem-
in the centrum semiovale) (Barkovich et al., 1988). ical composition caused by the arrival of precursors of
From estimations of Tl and T2 relaxation times, it is myelin constituents and materials necessary in the pro-
evident that a significant and global decrease in Tl and cess of myelin formation (van der Knaap and Valk,
T2 relaxation times occurs during the first few months 1990). When myelin is deposited, another important
of life, preceding the myelination of large parts of the decrease in Tl and T2 values occurs. This decrease is
brain (van der Knaap and Valk, 1990). This observation presumed to be related to the degree of myelination.
is in accord with animal experimental work, which The further Tl shortening correlates temporally with
shows that an impressive decrease in water content of the increase in cholesterol and glycolipids that accom-
the brain occurs just prior to the onset of a wave of pany the formation of myelin from oligodendrocytes
and probably reflects shortening of the Tl relaxation nation, whereas T2-weighted images seem better suited
time of the water in the white matter resulting from a for evaluating the associated changes in water content
very large interaction with myelin lipid (Koenig et al., (Barkovich et al., 1988). Diffusion-weighted images are
1990; Nakagawa et al., 1998; Poduslo and Jang, 1984). able to detect the process of myelination owing to their
The further decrease in T2 signal intensity of myelinat- sensitivity to changes of water diffusion induced by the
ing white matter correlates temporally with the tight- myelin membranes (Huppi et al., 1998; Takeda et al.,
ening of the spiral of myelin around the axon, which is 1997). They may be more sensitive than Tl-weighted
accompanied by loss of white matter water content. As images to white matter myelination, but their use has
myelin matures, it becomes increasingly hydrophobic been limited essentially to experimental work.
because of the development of the inner layer of phos- Neonatal posterior fossa structures that exhibit high
pholipids. This hydrophobic layer results in the pres- Tl and low T2 signal intensity at birth include the dor-
ence of fewer aqueous protons and a diminution in sal brainstem and the inferior and superior cerebellar
signal intensity on the T2-weighted images secondary to peduncles (Barkovich et al., 1988). An increase in Tl
a combination of shortened T2 relaxation time and a signal intensity denoting early myelination of the deep
decrease in water content. Tl-weighted images are a cerebellar white matter appears near the end of the first
better way of detecting the primary process of myeli- month of life and steadily increases, with high signal
intensity developing in the subcortical white matter of In the supratentorial brain, the decussation of the
the cerebellar folia by the third month. At 3 months of superior cerebellar peduncles, the ventral lateral region
age the cerebellum has a Tl appearance similar to that of the thalamus, the dorsal putamen, and the posterior
seen in the adult. On T2-weighted images, the middle limb of the internal capsule exhibit high Tl signal in-
cerebellar peduncle begin to decrease in signal inten- tensity at birth. To a lesser extent, these structures also
sity during the second month of life. Low T2 signal in- demonstrate areas of low T2 signal intensity at birth.
tensity in the subcortical white matter of the cerebellar The development of Tl and T2 signal changes proceeds
folia (arborization) begins to develop at approximately rostrally from the pons along the corticospinal tracts
the eighth month (5 months later than on Tl-weighted into the cerebral peduncles, the posterior limb of the
images), and the cerebellum reaches an adult appear- internal capsule, and the central portion of the cen-
ance at approximately 18 months (15 months later than trum semiovale. The white matter of the pre- and post-
on Tl-weighted images). Signal intensity in the ventral central gyri is of high Tl and low T2 signal intensity,
pons increases less rapidly, occurring during the third respectively, compared to the surrounding cortex by
through the sixth months. about 1 month of age. The change to high Tl signal
ABSTRACT Cytoarchitectonic layers of the human hippo- prematurely born infants survive. The aim of this chap-
campal formation are formed by the 24-25th fetal weeks. Con- ter is to provide age-specific morphological data against
sequently, cell formation in the hippocampal ventricular zone which behavioral scientists might correlate their rele-
is occasional after the 24th week. Cell formation in the hilar
region is continuous perinatally, but neuronal cell formation vant behavioral observations. The results of animal ex-
and cell death occur at a very low rate. Cell migration from periments are only briefly reviewed.
the proliferative zones lasts until the 32-34th fetal weeks in
Ammon's horn. Immature cells accumulate in the hilus of the Neuronal cell formation
dentate gyrus throughout the first year, when the subgranular
hilar zone appears cell-free, suggesting that immature cells from in the hippocampal formation
the hilus migrate to the granule cell layer and differentiate into
granule cells. Dendritic development and synapse formation In animal experiments, multiplying neurons can be
last for several years. Light microscopic changes of dendrites marked by isotope-labeled thymidine or the uridine an-
of the hippocampal neurons have been reported until the alogue bromodeoxyuridine (BrdU). It has been shown,
fifth postnatal year. Although the hippocampal formation of
newborn infants has the necessary synaptic connections for both in rodents and primates, that a majority of neurons
memory formation, the number of postnatal morphological in the entorhinal cortex, subicular complex, and Am-
changes suggests a significant modification of hippocampal mon's horn are formed before birth (Angevine, 1975;
circuits between the neonatal period and adulthood. Rakic and Nowakowski, 1981). In rodents, a few hip-
pocampal neurons continue to form in the last days of
The hippocampal formation plays an important role pregnancy, but no neurons are formed postnatally (An-
in the mechanisms of memory and learning. Experi- gevine, 1975). In primates, all neurons are formed in
mental data in rodents and primates as well as clinical the first half of pregnancy (Rakic and Nowakowski,
data in humans suggest that an individual with an im- 1981). There is only one region of the hippocampal
mature or lesioned hippocampus is unable to perform formation where postnatal neuronal formation occurs,
memory and learning tasks as well as a normal adult. and that is the dentate gyrus. The major difference be-
Brain development does not stop at birth, and recent tween rodents and primates in the formation of granule
data suggest that development may last until adulthood. cells is that in rodents approximately 85% of these neu-
In harmony with that notion, accumulating data indi- rons are formed postnatally, whereas similar numbers
cate a prolonged development of the human hippo-
of cells are formed prenatally in primates (Bayer, 1980;
campal formation that includes cell formation, cell
Rakic and Nowakowski, 1981). In rodents, the process
migration, and synapse formation. This chapter sum-
of granule cell formation lasts for 3 weeks, whereas in
marizes the morphological changes that occur from the
monkeys newly formed granule cells have been visual-
24th week of gestation to the period when further
ized through the first postnatal month.
changes in the hippocampal formation are not detect-
In humans, data about neuronal formation are based
able under the light microscope and the structure ap-
on the first appearance of stained neurons at their final
pears to be adult-like. The 24th week has been chosen,
cytoarchitectonic position; this is because direct label-
as that week is the approximate time after which most
ing of dividing cells is not possible. In the human hip-
pocampus, pyramidal neurons already form distinct cell
layers by the 15th gestational week. During the 23-25th
LASZLO SERESS Central Electron Microscopic Laboratory, gestational weeks the pyramidal cell layer of Ammon's
Faculty of Medicine, University of Pecs, Pecs, Szigeti, Hungary. horn and the cytoarchitectonics of the subiculum and
granule cell and molecular layers of the dentate gyms the labeling index is above 35%, while the hippocampal
in the 11.5-month-old child (figure 4.5D,E; see color formation shows only a few labeled cells (Abraham et
plate 4) than in either the 3- or the 5-month-old chil- al., 1999). There were only a few cases when material
dren. The large number of small, round, darkly stained could be obtained from children older than 1 year, and
labeled cells in the upper molecular layer suggests an those specimens were not suitable for immunostaining.
active cell multiplication of oligodendroglia cells that Therefore, we have no direct information about cell for-
might be related to active myelin formation during this mation in children older than 1 year.
age span. Considering the total number of neurons of
the granule cell layer of a newborn child, the labeling Evidence of cell death
index per section is in the range of 1:1000. In later age
groups this index is lower. The cerebellum of the same Cell formation is always accompanied by cell death.
children has always been processed for immunostaining Therefore, we have examined the frequency of occur-
together with the hippocampus. Sparsity of immuno- rence of pycnotic nuclei, which indicate that cell death
reactive cells in the hippocampal formation is not a has occurred in the hippocampal formation. In those
methodological failure: In the external germinal layer cases where hypoxic damage was not significant we have
of the cerebellum of the 3- and 5-month-old children, found only a few pycnotic nuclei inside the granule cell
layer and the hilus of the dentate gyrus. Those are the 4.6A; see color plate 5). The young cells have no cyto-
areas where the frequency of labeled cells is the highest plasm and their nuclei are darkly stained (figure 4.6A).
in the hippocampal formation. In order to avoid mis- Pycnosis has been found rarely in the dentate granular
calculation caused by postmortem autolysis or hypoxic layer and the pycnotic cells are not exclusively at the
damage, we have examined the frequency of pycnotic hilar border, where newly generated cells are supposed
nuclei in the hippocampi of newborn perfused mon- to be located, but are distributed in the entire width of
keys. Monkeys have been used to study the postnatal the granule cell layer (figure 4.6C; see color plate 5).
development of hippocampal connections and the In a few cases, apoptosis may be assumed based on the
maturation of the ultrastructure (Seress and Ribak, nuclear changes of the neurons that were probably
1995a,b). It is known that granule cell formation occurs granule cells (figure 4.6D; see color plate 5). Such apop-
in the monkey hippocampus until the 32nd postnatal totic cells were evident both at the molecular layer (fig-
day (Rakic and Nowakowski, 1981). In the perinatal pe- ure 4.6.D) and hilar borders (figure 4.6B; see color
riod a large number of young granule cells locate them- plate 5) of the granule cell layer, suggesting that cell
selves at the hilar border of the granule cell layer. death does not necessarily occur among the newly
Therefore, we assume that in newborn and 4-day-old generated cells. The frequency rate of cell death was
monkeys cell death may accompany cell formation. In extremely low in the monkey dentate gyrus, approxi-
newborn monkeys, cresyl violet-stained semithin sec- mately 2-3 cells per 1000 granule cells.
tions reveal the developed granule cells owing to their In conclusion, the overwhelming majority of neurons
purple-stained cytoplasm which sharply contrasts with of the human hippocampal formation are formed in
the thin, pale cytoplasmic rim of young cells (figure the first half of pregnancy, before the 24th week. There
is limited cell formation within the dentate gyrus in the torhinal cortex, bypass previously generated neurons
perinatal period, but the number of newly formed cells on their way to the superficial limits of the developing
is very low relative to the total number of cells. Similarly, cortical plate (Nowakowski and Rakic, 1981). There-
there is no significant cell death in the perinatal period, fore, the inside-out migration pattern is similar to that
neither in the human nor in the monkey hippocampal in the neocortex. The exception is again the dentate
formation. This suggests that formation of neuronal gyrus, where the granule cell layer is formed in an
connections is not accompanied by cell death in the outside-in pattern (Bayer, 1980; Nowakowski and Rakic,
primate hippocampus and that most of the newly gen- 1981). The dentate gyrus receives neurons both from
erated neurons survive. the ventricular zone and from the hilar proliferative
zone, which may be a hippocampal equivalent of the
Cell migration in the hippocampal formation subventricular zone, because the dynamics of cell for-
mation appear to be similar in the subventricular zone
Experimental studies in rodents and primates have and in the hilus (Bayer, 1980; Seress, 1977). Cell migra-
shown that the ventricular zone underlying the hippo- tion to Ammon's horn, the subiculum, and the ento-
campal formation is the source of neurons (Bayer, 1980; rhinal cortex terminates early in pregnancy, whereas
Nowakowski and Rakic, 1981). Migrating neurons in migration of granule cells continues for a long period
Ammon's horn, in the subiculum as well as in the en- of time postnatally. In rodents, granule cell migration
ABSTRACT Sex steroid hormones play an important role in cies, however, it is now clear that there are sex
regulating plastic changes in neuronal structure and function differences in areas of the brain that are not associated
throughout development and adulthood. There is a diversity with the control of reproductive function, including the
of cellular mechanisms by which steroid hormones influence
neural function, including metabolism to more active forms, hippocampus, striatum, cerebellum, amygdala, and ce-
modulation of function through traditional nuclear receptors, rebral cortex, and/or are associated with nonreproduc-
binding to membrane receptors, and activation of a variety of tive behaviors, such as spatial problem solving abilities,
second messenger pathways. A wide variety of neural processes verbal abilities, aggression, defensive behaviors, motor
are influenced by sex steroid hormones, including neuro- activity, and various aspects of learning and memory.
genesis, cell migration, growth of the neuronal cell body, den-
dritic growth, differentiation and synapse formation, synapse
elimination, neuronal atrophy and apoptosis, neuropeptide Sex steroid receptors within the brain
expression, the expression of neurotransmitter receptors, and
neuronal excitability. In many cases steroid hormones must ESTROGEN RECEPTORS Early studies of the effects of
be present both developmentally and in adulthood to elicit sex steroid hormones on the brain showed that hypo-
maximal effects on behavior.
thalamic regions, including the preoptic area and ven-
When they have been looked for, differences tromedial hypothalamus, played a central role in
throughout the brains of males and females have been mediating the action of estrogen on reproductive be-
clearly demonstrated in nearly all species, both with re- haviors and control of the hypothalamic-pituitary-
gard to the structural organization of many brain nuclei ovarian axis (Madeira and Lieberman, 1995; Pfaff,
and neural circuits, and in the functionality and re- 1980). Complementary studies identifying regions of
sponsiveness of neural systems to internal and external the brain with bound labeled estrogens showed that
stimuli. Moreover, many of the sex differences in brain these hypothalamic regions had the highest concentra-
structure and function result from differential exposure tions of estrogen receptors (ER) in the brain (Pfaff and
of males and females to the sex steroid hormones (tes- Reiner, 1973; Stumpf, Sar, and Keefer, 1975). Further
tosterone in males, estradiol and progesterone in fe- localization studies, identifying estrogen receptors by
males) . Sex steroid hormones play a role early in brain immunocytochemistry or in situ hybridization, con-
development in the "organization" of neural circuits, firmed the strong presence of estrogen receptors in the
and have critical roles throughout adulthood in "acti- hypothalamus and in brain areas with strong connec-
vating" specific behaviors. Although some generaliza- tions to the hypothalamus, including the amygdala, sep-
tions can be made, the specific neural circuits that are tal nuclei, the bed nucleus of the stria terminalis, the
influenced by sex steroid hormones and the time pe- medial part of the nucleus of the solitary tract, and the
riods of sensitivity to sex steroid hormone influence lateral portion of the parabrachial nucleus. However,
show dramatic species specificity. these studies showed relatively few estrogen receptors
The earliest studies of sex hormone influences on in the hippocampus, cerebellum, or the cerebral cortex
brain structure and function examined the effects of (Cintra et al., 1986; DonCarlos, Monroy, and Morrell,
testosterone and estradiol on sexual functions, includ- 1991; Simerly et al., 1990). No major differences were
ing reproductive capacity and behavior. In many spe- found in the distribution of estrogen receptors in males
versus females (Simerly et al., 1990).
In 1996 a new member of the steroid hormone re-
ceptor superfamily, one with a high sequence homology
JUDY L. CAMERON Departments of Psychiatry, Neuroscience,
and Cell Biology & Physiology, University of Pittsburgh, Pitts- to the classical estrogen receptor (now referred to as
burgh, Pennsylvania; Oregon Regional Primate Research Cen- ER-a), was isolated from rat prostate; this receptor
ter, Oregon Health Sciences University, Beaverton, Oregon. was named ER-b (Kuiper et al., 1996, 1998). This novel
shown to increase both sustained and transient calcium Sexual dimorphisms in the brain
currents—actions that may well be mediated by ge-
nomic actions of this steroid hormone (Joels and Karst, PROCESSES AFFECTED BY SEX STEROIDS Sex steroid
1995). hormones lead to sexual differentiation of behavior by
There is also evidence that estrogen may act as a neu- affecting a wide variety of cellular mechanisms within
ral growth factor within the brain, influencing neuronal the nervous system (Cooke et al., 1998). Processes that
development, survival, plasticity, and regeneration have been shown to be modulated by sex steroid hor-
(Toran-Allerand, Singh, and Setalo, 1999). Neurons in mones include neurogenesis (Gurney and Konishi,
the forebrain coexpress estrogen and neurotrophin re- 1979; Jacobson and Gorski, 1981), cell migration (Sen-
ceptors, and these neurons are also sites of both estro- gelaub and Arnold, 1986), growth of the neuronal soma
gen and neurotrophin synthesis. Moreover, estrogen (Breedlove and Arnold, 1981), dendritic growth, differ-
and nerve growth factor can reciprocally regulate the entiation and synapse formation (Frankfurt et al., 1990;
expression of each other's receptors, and accumulating Goldman and Nottebohm, 1983; Gomez and Newman,
evidence suggests that there is cross-coupling of their
1991; Gould, Woolley, and McEwen, 1990; Juraska,
signaling pathways (Toran-Allerand, Singh, and Setalo,
1991; Raisman and Field, 1973; Woolley and McEwen,
1999).
1992), synapse elimination (Leedy, Beattie, and Bres-
Further evidence of complex interactions between
steroid hormones and neural signaling systems comes nahan, 1987), neuronal atrophy and apoptosis (Davis,
from studies showing that dopamine can modulate sex- Shryne, and Gorski, 1996; McEwen, 1996; Woolley,
ual behavior in rodents via a mechanism that requires Gould, and McEwen, 1990), neuropeptide expression
the presence of intracellular progesterone receptors (De Vries, 1990; Malsbury and McKay, 1994), the ex-
within hypothalamic neurons (Mani, Blaustein, and pression of neurotransmitter receptors (Levesque, Gag-
O'Malley, 1997). This action of dopamine involves a non, and Di Paolo, 1989; McEwen, 1991; Turner and
ligand-independent activation of progesterone recep- Weaver, 1985), and neuronal excitability (Teyler et al.,
tors (Mani et al., 1994). 1980; Wong and Moss, 1992).
70 FUNDAMENTALS OF DEVELOPMENTAL N E U R O B I O L O G Y
greater dendritic arborizations in Wernicke's area in fe- shown that neonatal exposure to androgens can mas-
males compared to males have been reported (Jacobs, culinize some sexually dimorphic neuroanatomical dif-
Schall, and Scheibel, 1993). ferences found in the cortex, such as cortical thickness
A third sexual dimorphism that has been proposed (Lewis and Diamond, 1995), little is known about the
to underlie the sexual dimorphism in verbal abilities is effects of sex steroid hormones on the morphology of
the size of the corpus callosum. The corpus callosum is brain areas specifically involved in language processing
a nerve fiber tract that connects the two cerebral hemi- or the size or shape of the corpus callosum. However,
spheres, allowing transfer of information between the one principle is clear from studies in other parts of the
hemispheres for complex cognitive processing, includ- brain, including the hypothalamus and the hippocam-
ing language (Hines et al., 1992). In 1982, de Lacoste pus: There is profound cellular and circuit specificity
and Holloway hypothesized that the decrease in func- for the actions of sex steroid hormones. Thus, in mak-
tional neural asymmetry in women was due to increased ing conclusions about the role of steroid hormones on
interhemispheric exchange via the corpus callosum. cortical functions, it will be important to be very specific
Since that time, sexually dimorphic structural asym- with regard to both function and neuroanatomical
metries of the corpus callosum have been reported for substrate.
its area, shape, and fiber composition (Wisniewski,
1998). However, the issue of whether there is a consis- Effects of sex steroid hormones on behavior
tent sexual dimorphism of this fiber tract has been over the lifespan
highly controversial. The splenium is the posterior por-
tion of the corpus callosum that contains fibers pro- Most of the data reviewed in this chapter concern ef-
jecting between the main auditory and visual cortical fects of sex steroid hormones on adult behavior, result-
areas of the two hemispheres. In humans, a number of ing either from organizational effects of early hormonal
postmortem studies and MRI investigations provide exposure during the fetal or neonatal periods or from
support for the conclusion that females have a more activational effects of sex steroid hormones in adult-
bulbous splenium, although it is important to control hood. However, sex steroid hormone levels change dra-
for the overall corpus callosum shape, chronological matically both in children as they mature and enter
age of the individuals, cerebral volume, and handed- puberty, and in women when the activity of the repro-
ness when making such comparisons (Allen et al., 1991; ductive axis declines at menopause; thus, one would
Burke and Yeo, 1994; Johnson et al., 1996; Wisniewski, expect to see accompanying changes in behaviors mod-
1998). The isthmus of the corpus callosum connects the ulated by sex steroid hormones at these times. One
parietal and temporal lobes, and has been more consis- would also predict that sexually dimorphic areas of the
tently reported to have a larger area in women com- brain that are dependent on adult levels of sex steroid
pared to men, although for such comparisons it is again hormones for maintenance of the dimorphism would
important to control for the same factors as in studies change in morphology over puberty and menopause.
of the splenium (Denenberg, Kertesz, and Cowell, 1991; Although we know a good deal about the develop-
Steinmetz et al., 1992; Witelson, 1989). Sexual dimor- ment of sex-related behaviors at puberty (Baum, 1979),
phism in the corpus callosum of several rodent species surprisingly few studies have examined pubertal or
has also been reported (Wisniewski, 1998), and an im- menopausal changes in other behaviors that appear to
portant finding in these studies is that environmental be influenced by sex steroid hormones. In part, this
rearing conditions can have modulatory effects on the paucity may reflect the fact that much of the work ex-
sexual dimorphism of corpus callosum (Juraska and amining effects of sex steroid hormones on such be-
Kopcik, 1988). Another interesting finding of these lat- haviors as learning and memory and other cognitive
ter studies is that there are sexual dimorphisms in the functions is relatively recent. It may also reflect the dif-
ultrastructure of the rat corpus callosum, with sex dif- ficulty in determining what portion of the developmen-
ferences in the number and diameter of axons project- tal and aging changes in these behaviors results from
ing across the corpus callosum (Juraska and Kopcik, changes in sex steroid hormones and what portion is
1988; Kopcik et al., 1992). attributable to hormone-independent developmental
Despite the great interest in and numerous studies and aging processes. Finally, methodological prob-
devoted to the neural underpinnings of sexual dimor- lems—poor characterization of the activity of the re-
phisms in verbal abilities, little information is available productive axis; the use of widely varying tests of
regarding the role of sex steroid hormones in mediat- learning, memory, and other cognitive processes; and
ing the development of these functional differences. Al- extreme variation in hormone treatment regimens in
though some investigators using rodent models have postmenopausal studies—also contribute to the lack of
ABSTRACT This chapter provides an overview of the devel- other components of this system with which it is inti-
opment of the prefrontal cortex that begins with the differ- mately linked. The discussion that follows provides an
entiation of its intrinsic projection and local circuit neurons overview of corticolimbic development at various stages
during embryogenesis, but extends well beyond. Despite the
early appearance of laminar patterns that reflect those seen of the life cycle, beginning with gyral development dur-
in the adult brain, the architecture of synaptic connections ing embryogenesis and extending to the postnatal pe-
and the maturation of intrinsic and extrinsic neurotransmitter riod. Consideration is given to the maturation of both
systems continue to change in the prefrontal cortex during intrinsic and extrinsic neurons and their respective neu-
the postnatal period. It is particularly noteworthy that a variety rotransmitter systems, and the chapter concludes with
of studies have demonstrated that the monoaminergic neu-
rons, especially those employing dopamine as a neuromodu- a discussion of how the latter interact with one another
lator, continue to infiltrate the cortical mantle until the early during postnatal development.
adult period. More recent evidence, however, even suggests
that dopaminergic and serotonergic fibers form appositions
with the same projection and local circuit neurons of the pre-
Gyral development
frontal cortex. Because of this convergence, a disturbance in
the development of serotonergic projections in this region is The development of the cerebral cortex occurs at dif-
associated with an increase of its dopaminergic innervation. ferent rates that, to some extent, reflect phylogeny. For
Taken together, these findings suggest that these two mono- example, at 16-19 weeks of embryogenesis, the cingu-
aminergic systems may compete with each other for the estab- late gyrus can be distinguished, but this occurs long be-
lishment of functional territory on intrinsic neurons of the
cortex, although it is not clear whether this interaction is me- fore the parahippocampal gyrus and hippocampal
diated within the cortex itself or possibly at midbrain levels formation have become discernible in the medial tem-
where the respective nuclei are located. Overall, the results poral lobe (Gilles, Shankle, and Dooling, 1983). Corti-
described in this chapter suggest that the ontogenesis of the cal regions undergo ontogenesis at varying rates that,
prefrontal cortex is a protracted process in which a marked to some degree, reflect phylogeny (table 6.1). For ex-
degree of plasticity, even in the early stages of adulthood, con-
tributes to the establishment of mature patterns of connectivity. ample, limbic cortical areas tend to differentiate early
in gestation. At 16-19 weeks in humans, the cingulate
In recent years, it has become apparent that the de- region can be distinguished as a gyrus, but this is well
velopment of the corticolimbic system continues well before the parahippocampal gyrus and the hippocam-
beyond the neonatal period, perhaps even as late as the pal formation have begun to take shape in the medial
fifth and sixth decades (Benes, 1995, 1997b; Benes et temporal area at weeks 20-23 (Gilles, Shankle, and
al., 1994). Of particular relevance to our understanding Dooling, 1983). The superior and medial frontal gyri
of the maturation of cognitive and emotional behaviors (prefrontal regions) do not take on a clear gyral con-
is the development of the prefrontal cortex and the figuration until 24-27 weeks of gestational age, and the
angular and supramarginal gyri (inferior parietal area)
are not distinguished until 28-31 weeks (Gilles, Shan-
FRANCINE M. BENES Laboratory for Structural Neuroscience,
McLean Hospital, Belmont, Massachusetts; Program in Neu- kle, and Dooling, 1983). The orbital frontal gyri, which
roscience and Department of Psychiatry, Harvard Medical (like the cingulate gyri) are well represented in earlier
School, Boston, Massachusetts. mammalian forms such as rodents, are among the last
FIGURE 6.2 Light photomicrographs of rat anterior cingulate between PI and P20, there is a gradual increase in the thick-
cortex (medial prefrontal are) at postnatal days PI, P5, P10, ness of the cortical mantle as the packing density of neuronal
P20, and P41. At PI, the immature cortical plate (CP) lies cell bodies decreases. At postweanling day P41, there is no
between layer I and deeper layers V and VI. By P5, however, further change in the density of neurons. WM = white matter.
the cortical plate has differentiated into layers II and III, but (Reprinted with permission from Vincent, S. L., L. Pabreza,
the density of cells in these superficial laminae is greater than and F. M. Benes, 1995. Postnatal maturation of GABA-
that observed in the deeper layers. Between P10 and P20, the immunoreactive neurons of rat medial prefrontal cortex. /.
increased thickness of the cortical mantle is primarily related Comp. Neurol. 355:84.)
to the expansion of neuropil in layers II and III. Overall,
fivefold increase during the perinatal period (Brooks- Norepinephrine Noradrenergic cell bodies found in the
bank, Atkinson, and Balasz, 1981; Diebler, Farkas- locus coeruleus (Levitt and Moore, 1978; Lindvall,
Bargeton, and Wehrle, 1979) and an additional 100% Bjorklund, and Divac, 1978) undergo their last mitotic
increase for several weeks thereafter (Brooksbank, division at gestational days 12-14 in rat brain (Lauder
Atkinson, and Balasz, 1981). Functional inputs are and Bloom, 1974) and at gestational days 27-36 in
thought to regulate the maturation of the GABA system. monkeys (Levitt and Rakic, 1982). By birth, there is a
For example, in cat visual cortex, dark-rearing from rather extensive network of noradrenergic fibers in rat
regarding human brain, available evidence suggests transmitter systems is extensive. At this juncture, a
that postnatal maturation of both the cholinergic and reasonable question to ask is whether these progressive
serotonergic systems may continue throughout child- postnatal changes may involve an increase in the degree
hood and adolescence, and may, in some cases, persist to which some or all of these transmitter systems are
throughout the lifespan of the normal individual. Based interacting with one another.
on observations in rats, however, it seems likely that
changes in the dopamine system are also occurring un- DOPAMINE-GABA INTERACTIONS As shown in figure
6.6 (see also color plate 9), a recent study has suggested
til the early adult period. that frequent appositions occur between dopaminergic
afferents and both pyramidal neurons and GABAergic
Developmental interactions between intrinsic interneurons in the medial prefrontal cortex of adult
and extrinsic cortical transmitters rat (Benes, Vincent, and Molloy, 1993). When the fre-
quency of these contacts was compared for preweanling
It is evident from the earlier discussion that the post- and postweanling rats, it was found that there is a
natal maturation of both intrinsic and extrinsic neuro- curvilinear increase in the number of dopamine-
immunoreactive varicosities found in contact with non- studies, because messenger RNA for both the Dl and
pyramidal neurons in layer VI of rat medial prefrontal D2 subtypes has been localized in both pyramidal and
cortex (figure 6.7). In layer II of this region, there was nonpyramidal neurons (Huntley et al., 1992). Consis-
a sevenfold increase on nonpyramidal neurons, but no tent with this, a large proportion of interneurons in rat
change on pyramidal cells (Benes, 1997b). These find- mPFCx have been found to express Dl and D2 binding
ings suggest that (1) postnatal increases in the inter- activities (Vincent, Khan, and Benes, 1992, 1995), and
action of one transmitter system with another probably studies in which in vitro microdialysis was employed
do occur to a significant degree, (2) such changes may have demonstrated that agonists for both the Dl and
vary in magnitude according to cortical layer, and (3) D2 subtypes are associated with a reduction in the re-
changes of this type may occur differentially for one lease of 3H-GABA (Penit-Soria, Audinat, and Crepel,
neuronal cell type versus another within a given cortical 1987; Retaux, Besson, and Penit-Soria, 1991a,b). Thus,
layer. In other words, postnatal maturational changes in as suggested by these results, cortical dopamine projec-
cortical neurotransmitter systems may be far more com- tions can potentially exert a significant nonsynaptic
plex than heretofore suspected, and to fully understand modulatory influence on GABAergic cells in rat
the implications of postnatal ontogeny, a clear delin- mPFCx.
eation of how changes in neurochemical markers may
be reflected in alterations of specific synaptic connec- Convergence of dopamine and serotonin fibers
tivity is needed. on prefrontal neurons
A variety of studies have demonstrated that both py-
ramidal and nonpyramidal neurons express receptors Serotonin fibers are believed to converge on the same
that mediate the effect of dopamine. For example, in pyramidal cells and GABA neurons that receive inputs
one study dopamine Dl receptor binding activity has from catecholaminergic neurons (Gellman and Agha-
been localized to nonpyramidal cells (Vincent, Khan, janian, 1993), and a triple immunofluorescence study
and Benes, 1993), while in another it was associated was able to confirm that this arrangement occurs in the
with pyramidal cells (Bergson et al., 1995) in prefrontal medial prefrontal cortex of rats to an extensive degree
cortex. This apparent discrepancy can be explained by (Taylor and Benes, 1996). It is noteworthy that pyra-
methodological and species differences in these two midal neurons also received inputs to their somata from
ABSTRACT The dentate gyrus of the hippocampal formation Adult neurogenesis has been demonstrated to occur
continues to produce granule neurons throughout adulthood in the brains of all vertebrate species investigated thus
in many species, including humans. Although the functional far (Altman and Das, 1965; Goldman and Nottebohm,
significance of adult hippocampal neurogenesis is unknown,
its magnitude and conservation across taxa suggest that this 1983; Gould et al., 1997, 1998, 1999b; Kornack and
process may be fundamental to normal hippocampal func- Rakic, 1999; Perez-Canellas and Garcia-Verdugo, 1996;
tion. Previous studies have demonstrated that the production Polenov and Chetverukhin, 1993; Zupanc, 1999), in-
and survival of neurons in the adult hippocampal formation cluding humans (Eriksson et al., 1998), suggesting that
are regulated by both experiential and neuroendocrine fac- it is both a primitive and a highly conserved process.
tors. Stress reduces the proliferation of granule cell precur-
sors, in part, by increasing circulating levels of adrenal
Quantitative estimates indicate that a significant
glucocorticoids. Conversely, estrogen increases the prolifera- amount of hippocampal neurogenesis even occurs in
tion of granule cell precursors and is responsible for a sex the adult primate brain (Eriksson et al., 1998; Gould et
difference favoring females in the production of immature al., 1998, 1999b). The conservation of adult hippocam-
granule neurons. Additionally, certain types of learning en- pal neurogenesis across vertebrate taxa suggests a fun-
hance the number of new neurons in the dentate gyrus of
damental biological significance that is only beginning
adult rats. Collectively, these findings suggest that adult-
generated neurons are a substrate by which experiential and to be explored experimentally. Although the functional
hormonal cues influence normal hippocampal function. significance of this process is not yet known, the discov-
ery of adult neurogenesis in the hippocampal forma-
Most neurons in the adult mammalian brain are pro- tion has significantly altered the way in which the adult
duced during a discrete part of the embryonic period. brain is viewed in terms of its potential for structural
However, as indicated by a group of now classic studies, change. Perhaps of even greater significance is the pos-
a substantial number of granule cells are produced in sibility that adult neurogenesis can be harnessed to im-
the cerebellum, olfactory bulb, and hippocampal for- prove the clinical prognosis of neurodegenerative
mation during the postnatal period (Altman, 1969; Alt- disease and traumatic brain injury. However, before this
man and Das, 1966). Furthermore, in the olfactory bulb goal can be realized, investigators must first elucidate
and dentate gyrus of the hippocampal formation, neu- the role that adult-generated neurons play in normal
rogenesis continues throughout adulthood (Altman functional circuitry, as well as the factors that regulate
and Das, 1965). The concept that structural modifica- the production, differentiation, survival, and integra-
tions, such as changes in dendrite length or synapse tion of newly generated neurons into the adult neu-
number, are a normal physiological occurrence in the roanatomic context.
adult brain was (and remains) fairly controversial. Thus, In this chapter, we first consider the key experimental
the observation that new neurons are produced in the findings that led to the discovery that the adult brain
mammalian olfactory bulb and dentate gyrus through- continues to produce hippocampal neurons. In so do-
out adulthood was surprising, and, in fact, has only re- ing, we discuss techniques important for analyses of
cently gained acceptance. neurogenesis in the adult brain. Second, we consider
the factors and conditions, including hormones and ex-
perience, that regulate neurogenesis in the hippocam-
PATIMA TANAPAT, NICHOLAS B. HASTINGS, AND ELIZABETH
pal formation. Third, we consider the possible role of
GOULD Department of Psychology, Princeton University, late-generated cells in the normal function of this brain
Princeton, New Jersey. region.
It is not clear what specific aspects of these tasks are ated during the postnatal developmental period, such
necessary for the enhanced survival of adult-generated as those in the dentate gyrus, might be important for
cells. Indeed, electrophysiological studies have shown forming associations during development (Altman,
that both hippocampal-dependent and hippocampal- Brunner, and Bayer, 1973; Altman and Das, 1965,1967),
independent tasks activate the hippocampal formation suggesting that microneurons are morphologically well
(Weisz, Clark, and Thompson, 1998). One possible dif- suited for the maturation of learning processes. More
ference between the two sets of tasks is their difficulty. recently, Nottebohm suggested that new hippocampal
Both place learning and trace eyeblink conditioning are neurons may be a potential cellular substrate for learn-
more difficult to acquire than are their hippocampal- ing in the adult (Nottebohm, 1989). Indeed, a contin-
independent counterparts (cue learning and delay eye- ually rejuvenating population of new neurons seems
blink conditioning). However, the observation that well suited for the proposed transient role of the hip-
place learning (a task that is normally learned in fewer pocampal formation in information storage (Squire,
than 10 trials) is sufficient to rescue new cells, whereas 1992; Squire and Zola, 1998). Although no direct evi-
delay eyeblink conditioning (a task that typically re- dence supports the view that adult-generated hippo-
quires 100-200 trials to reach criterion) suggests that campal cells are involved in learning, several studies
task difficulty is not the only factor. These data clearly have demonstrated a positive correlation between the
demonstrate that certain types of learning are sufficient number of adult-generated cells and performance on
to enhance the number of new neurons in the dentate certain types of learning tasks (for review, see Gould
gyrus of adult rats and present the possibility that these et al., 1999c). Moreover, many parallels between the
new cells may play a role in learning. number of new neurons and hippocampal-dependent
learning can be observed when considering the neuro-
Functional significance of adult-generated neurons endocrine and experiential factors discussed previously.
When considering the functional implications of
Currently, the functional significance of neurons gen- changes in the postnatal production of new neurons, it
erated in the dentate gyrus of adult animals is unknown. is important to bear in mind that new granule cells re-
Given their incorporation into the hippocampal for- quire time to differentiate and become integrated into
mation, a brain region involved in certain types of learn- functional circuitry. Thus, changes in cell proliferation
ing and memory, and the effect of learning on the are not likely to result in immediate functional conse-
survival of these new cells, it seems reasonable to con- quences. However, that does not imply that adult-
sider a potential role for these new neurons in learning. generated cells are not capable of exerting an impact
Altman was the first to propose that neurons gener- prior to differentiation into mature granule neurons.
FIGURE 7.9 Conditioning of the trace eyeblink response does jected with BrdU on the last day of training and perfused 24
not enhance the proliferation of adult-generated cells in the hours later. (B) Stereological estimates of the total numbers
dentate gyrus of adult male rats. (A) Acquisition of the trace of BrdU-labeled cells in the dentate gyrus of these animals
eyeblink conditioned response (trace paired) and the un- after training. Bars represent mean ± SEM obtained from 5-
paired condition (trace unpaired) in animals that were in- 6 animals. Asterisk indicates a significant difference, p < 0.05.
In fact, we have shown that adult-generated neurons Because of this, conditions under which cell prolifera-
extend axons into the CA3 region as early as 4-10 days tion and survival are chronically enhanced or dimin-
following mitosis (figure 7.1), considerably prior to the ished are of particular interest because they are most
time believed to be required for neuronal maturation likely to elucidate the functional consequences of
(Hastings and Gould, 1999). changes in adult neuron production and survival.
In addition, it is important to note that acute changes In general, chronic increases in the factors that
in cell proliferation and survival may not be of sufficient negatively regulate adult granule neuron production
magnitude to produce an observable functional impact. are associated with poor performance on hippocampal-
dependent tasks, whereas chronic increases in the fac- levels of circulating estrogen. As in the case of acute
tors that positively regulate adult granule neuron stress studies, it is important to recognize that the time
production are associated with enhanced cognitive frame examined in many studies is one that is likely to
function. Several studies have demonstrated that be too early to implicate changes in neuron production,
chronic stress results in an impairment of hippocampal- Interestingly, in a recent study (Luine et al.,1998), it has
dependent learning. For example, it has been reported been shown that repeated treatment with estrogen for
that in rats, chronic stress leads to impaired perfor- more, but not fewer, than 4 days results in enhanced
mance on a spatial navigation learning task (Bodnoff et performance on a hippocampal-dependent task. This
al., 1995; Krugers et al., 1997; Luine et al., 1994, 1996). finding is consistent with the observation that adult-
The stress-induced impairment that is observed is not generated cells do not extend axons prior to 4 days after
permanent; animals that were tested on a spatial task at their production. It is likely that sufficient time for
a later time point following the termination of the stress adult-generated cells to extend axons is required for
period did not perform significantly differently from estrogen-induced increases in cell production to have a
unstressed animals (Luine et al., 1994, 1996). The ob- functional effect.
servation that stress-induced impairment of perfor- Consistent with studies that have demonstrated a
mance on these tasks is not permanent is consistent positive effect of estrogen on the performance of hip-
with a possible role for adult-generated cells. It may be pocampal-dependent tasks, a recent study has reported
that the impairment persists only as long as the pro- a sex difference in rats favoring females in spatial learn-
duction of these cells has been altered. It should be ing in a Morris water maze (Perrot-Sinal et al., 1996).
noted that previous work has reported that brief stress Previous studies of sex differences in spatial navigation
enhances hippocampal-dependent learning (Shors, learning in rodents on this task have yielded conflicting
Weiss, and Thompson, 1992). However, these behav- data (Bucci, Chiba, and Gallagher, 1995; Galea, Kava-
ioral changes were observed shortly after stress and are liers, and Ossenkopp, 1996). However, the results of this
likely to involve other cellular mechanisms, such as more recent study demonstrate that although male rats
changes in synaptic plasticity. initially perform better than female rats on the task,
Several studies have demonstrated that estrogen has females learn better than males once the animals have
a positive effect on the acquisition of hippocampal- been acclimated to the testing apparatus—that is, once
dependent tasks. Although estrogen treatment has gen- stress has been reduced. Thus, a sex difference in re-
erally been found to enhance learning, some studies action to the novelty of this task prevents females from
report decreases in performance during times of high performing well. Once this performance confound is
ABSTRACT The application of animal tests of cognition to valuable knowledge about age-related sequences of cog-
children of various ages can provide inferential evidence of nitive development, but they provide little information
the time course of functional maturation of restricted neural about the functional development of underlying neural
substrates. This chapter describes the methods of administer-
ing to children various tasks in the Wisconsin General Testing circuits.
Apparatus, radial-arm maze, and Morris search apparatus. This chapter describes methodologies of applying
When compared with results from animals, the results from animal tests to children, summarizes the results of these
children allow inferences about the functional development tests, discusses inferences about the functional devel-
of a number of neural systems and structures including orbital opment of restricted brain systems, and presents some
prefrontal cortex, entorhinal and perirhinal cortices, hippo-
campus, and the inferior temporal corticostriatal system. of the limitations and caveats inherent in this experi-
mental approach.
A primary goal of developmental cognitive neurosci-
ence is to trace the functional maturation of neural sys- Testing children with tasks developed for monkeys:
tems that underlie various aspects of cognition. Toward Use of the Wisconsin General Testing Apparatus
this goal, we have tested children on a variety of cog-
nitive tasks that were originally developed for animals. APPARATUS For more than 40 years, the Wisconsin
The logic of this approach consists of three steps. First, General Testing Apparatus (WGTA) has been used to
a number of cognitive tests have been documented to administer cognitive tests to animals. The WGTA has
measure the function of specific neural circuits in ani- been used extensively with nonhuman primates be-
mals. Second, many of these animal tests can be applied cause it takes advantage of reaching and investigative
to children with little procedural change. Third, the re- behaviors that occur very early in life (Harlow, 1951,
sulting performance by children of different ages allows 1958). We have tested children in modified versions of
us to make logical inferences about the functional de- the WGTA on a variety of tasks including object discrim-
velopment of underlying neural substrates under the inations (Overman, 1990; Overman et al., 1992), trial-
assumption that the substrates of animals and humans unique delayed matching and nonmatching to sample
are similar. (Overman, 1990; Overman et al., 1992), single pair and
This mixture of animal neuroscience and develop- concurrent discriminations (Overman et al., 1992), ob-
mental psychology allows inferences that differ from ject reversal (Overman et al., 1996b), visual preferential
those generated by the procedures of conventional viewing (Overman et al., 1993), trial-unique oddity
child development. Traditional tests for children yield (Overman et al., 1996a), and landmark discriminations
(Coleman and Overman, 1999). Child testing proce-
dures in the WGTA have been changed from animal
WILLIAM H. OVERMAN Psychology Department, University of testing procedures as little as possible; however, a few
North Carolina at Wilmington, Wilmington, North Carolina.
JOCELYNE BACHEVALIER Department of Neurobiology and differences do exist. The human WGTA is of slightly
Anatomy, University of Texas Health Science Center, Houston, different dimensions from that used for monkeys, and
Texas. it has a movable tray instead of a stationary tray (figure
groups for whom the population standard does not validated with further experimentation, they may affect
hold. This would be expected for any sex-biased phe- not only psychological and social theory but educa-
nomenon (Halpern, 1992). The extremely close simi- tional practice as well. In our current research we are
larity of the tasks, behavioral data, and endocrine investigating the implications of individual differences
functions of infant humans and monkeys provides in- and the fact that, in both species, the sex differences
ferential evidence that, in both species, there is more seem to disappear in older subjects (over the age of 3
rapid maturation of orbital prefrontal circuits in males years for children). Preliminary results with cognitively
and more rapid maturation of inferior temporal circuits more challenging tests and older subjects suggest that
in females. Presumably, this phenomenon is related to some of these cognitive sex differences, e.g., those that
the perinatal testosterone surge that occurs in males of reflect differences in orbital prefrontal functioning,
both species (Bachevalier, Hagger, and Bercu, 1989; may exist across the lifespan (Reavis et al., 1998). This
Corbier, Edwards, and Roffi, 1992; Stahl et al., 1978). finding is in agreement with positron emission tomog-
This demonstration of cognitive sex differences in raphy evidence showing significant differences in sev-
children is one of the best examples of the kind of re- eral brain areas including the orbital prefrontal cortex
suits that can be derived from testing children with tests and temporal lobes of adult men and women (Andrea-
developed for animals. If our inferences are persistently son et al., 1993; Gur et al., 1995).
reinforced with sweetened cereal for the children and Forced-choice testing The forced-choice RAM procedures
M&M's for the adults and with social rewards (one ex- prevent the use of algorithmic solutions, thereby ensur-
perimenter said "Good" or "Good job" each time the ing the use of relational cognitive processes. In this pro-
subject retrieved a food reward. cedure, four nonadjacent arms are initially opened for
There were two testing versions in the RAM, free exploration while the remaining four are blocked. On
choice and forced choice, either of which could be cued trials 1-4 the subject retrieves rewards from the four
or noncued. The two versions allow measurement of open arms. Performance on trials 1-4 provides a mea-
different cognitive processes. sure of working memory. Next, the subject waits in the
center for a period of time during which all arms are
Free-choice testing In this version of the task a subject
blocked. During the retention interval, the subjects are
freely explores the eight open, baited arms of the maze.
free to visually inspect any or all arms of the maze. Then
The optimal strategy is to enter each arm only once.
all eight arms are opened and the subject searches for
The free-choice task can be solved either by a response-
based process or by a spatial-relational process. In the the remaining rewards in the four unvisited arms (trials
first case, the subject uses an algorithmic solution such 5-8). Performance on trials 5-8 yields the critical mea-
as "turn right" after exiting each arm. Here the subject sure of relational place learning ability, provided there
does not have to use spatial memory of which arm has are only extramaze cues.
been visited. In contrast, when using a spatial-relational
solution, the subject chooses nonadjacent arms; in this Cued versus uncued choice testing Both free- and forced-
case the subject must remember the visited versus the choice versions can be cued or uncued. In the cued
unvisted arms by knowing the arms' positions in rela- version each arm is rendered physically distinct by sa-
tionship to cues inside or outside the maze. lient visual stimuli (posters with colorful icons) at the
facts seem clear: (1) In rodents, an intact and anatom- least the fifth year of life, namely, development of spine
ically mature hippocampal system is required for rela- structure of mossy cells. This prolonged dendritic de-
tional solutions but not for cue-based solutions in the velopment suggests that synaptic connections between
RAM and Morris task (Jarrard, 1993; Keith, 1990; Rudy granule cells and their postsynaptic targets may last as
et al., 1987; Sutherland, Kolb, and Whishaw, 1982; Suth- long as 5 years postnatally. But, paradoxically, some hip-
erland and Rudy, 1989); and (2) performance on spa- pocampally mediated behaviors appear very early in life
tial-relational tasks does not reach maturity in children in primates. For example, visual recognition memory as
until after age 5 and perhaps as late as age 7-15 years. assessed by visual paired comparison procedures (also
With these two findings in mind, we speculated in called preferential viewing for novelty) depends on an
earlier reports (Overman et al., 1996c) that the late intact hippocampus (Zola et al., 2000; Pascalis and
maturation of spatial-relational behaviors in children Bachevalier, 1999), and this behavior is present at least
simply reflected late anatomical maturation of the as early as 1 month of age in monkeys (Bachevalier,
hippocampal and related systems. It is now clear, how- 1990) and at birth or earlier (in premature infants) in
ever, that the true nature of the matter is more com- humans (see Fagan, 1990, for review). Thus, the ques-
plex. First, as shown by Seress (1992, 2001 [chapter 4, tion is: Why do some hippocampally mediated behav-
this volume]), whereas much of the hippocampal for- iors (VPC-elicited visual recognition memory) appear
mation development occurs very early in postnatal life early in life while others (spatial-relational behavior)
of humans, some anatomical details are delayed until at appear much later? There are several possible
TABLE 8.1
Results of testing children with cognitive tasks developed for animals
ABSTRACT Event-related potentials (ERPs) can be recorded velopment; indeed, this volume is a recognition of this
noninvasively at the scalp surface in a matter of minutes and change. We now a have a variety of tools that permit us
do not require a motor or verbal response; thus, they repre- to study the living brain, although not all are amenable
sent an ideal method by which to study the neural correlates
of cognitive function across the entire lifespan. In this chapter
to studying the developing brain. This chapter focuses
we begin by reviewing the physiological basis of ERPs, and specifically on the use of one such tool, the recording
then describe the various ERP components that have been of event-related potentials (ERPs).
used to chart the development of the neural correlates of dif- Our goal is to provide a tutorial on how ERPs have
ferent cognitive functions. We conclude by describing how been used to study the ontogeny of cognition (for other
ERPs can be used to study brain development across the first
1-2 decades of life.
examples using this method, see de Haan, Johnson, and
Richards, this volume; Fabiani and Wee, this volume).
To accomplish our goal, we begin with an overview of
Our understanding of the behavioral correlates of the physiological bases of ERPs. We then move on to
cognitive development has increased exponentially dur- review the literature on how particular components of
ing the past three decades, and remains a rich vein for ERPs have been used to study different cognitive phe-
mining by the behavioral scientist. Unfortunately, our nomena. Because comprehensive reviews of this litera-
knowledge of the neurobiological forces that shape— ture exist elsewhere (see Nelson and Luciana, 1998),
and are shaped by—changes in cognition remains we shall provide illustrative examples rather than an ex-
primitive. As has been described elsewhere (Nelson, haustive review.
1995; Nelson and Bloom, 1997), this paucity of infor-
mation about the neural bases of cognition can be The genesis of the ERP signal
attributed in part to the fact that developmental
psychology and the neurosciences typically seek answers In order to measure changes in the brain's electrical
to questions that lie at different points on a knowledge activity, scalp electrodes require the summation of large
continuum (e.g., physiological basis of attention vs. be- numbers of neurons. Action potentials provide the larg-
havioral description of attention). Furthermore, each est electrical change in the nervous system, but the time
science requires different levels of analysis, the tools of course of this change is too brief to permit the neces-
the neurosciences being far more reductionistic than sary summation across neurons. Therefore, excitatory
those of the behavioral scientist. Gradually, however, the and inhibitory postsynaptic potentials provide the cur-
fields of neuroscience and behavioral science have be- rent that is detected by ERPs (Allison, Woods, and Mc-
gun to merge. Nowhere is this more evident than in the Carthy, 1986). These potentials occur over a relatively
field of functional brain imaging and cognition. Here long period of time, so that many neurons generating
we have seen a revolution in our ability to image the electrical current simultaneously are allowed to sum-
living brain at work. Although the majority of research mate, then propagate to the scalp surface.
in this area remains confined to the mature organism, The ERP signal is thought to be primarily derived
increasing effort is being devoted to understanding de- from pyramidal cells in the cerebral cortex and hippo-
campus (Allison, 1984). These cells tend to be parallel
to one another and many are perpendicular to the
CHARLES A. NELSON Institute of Child Development and De- scalp. Because many of these cells are aligned in par-
partment of Pediatrics, University of Minnesota, Minneapolis,
Minnesota. allel, their synchronous activation leads to the summa-
CHRISTOPHER s. MONK Institute of Child Development, Uni- tion of current in the same direction. Therefore, an
versity of Minnesota, Minneapolis, Minnesota. open field is created (reviewed in Allison, 1984). Open
What develops?
that of the adult (although undoubtedly task conditions nation, and the like—also contribute to the morpho-
will influence this similarity). Still, at least one late com- logical and topographic differences. Because recording
ponent—the P300—is broader peaked and longer in ERPs is completely noninvasive, can be done across the
latency, and the response to novel events is quite differ- lifespan while employing exactly the same task, and is
ent. These observations suggest that much develop- relatively inexpensive, this method would seem to rep-
ment remains after the age of 8. resent an ideal tool with which to explore a variety of
aspects of brain-cognitive relations. Thus far, its use has
Conclusions been confined predominantly to infants and young chil-
dren, and the majority of studies are cross-sectional in
The changes in cognition that occur from infancy to nature. It is our hope that, in the future, long-term Ion-
adulthood clearly manifest themselves in different gitudinal studies will be launched, and that this method
ERP signatures. These electrophysiological "footprints" will be cross-registered to other imaging methods that
probably reflect differences in cognitive ability across lend themselves to children, such as fMRI. In so doing,
this span of time, coupled with a host of differences in the superb temporal resolution of ERPs can be com-
the neural substrate that underlies cognition (e.g., bined with the excellent spatial resolution of fMRI; and
synaptic connections). In addition, less interesting when grounded in behavior, considerable more insight
changes—skull thickness, cell packing density, myeli- will have been gained into brain-behavior relations.
ABSTRACT Recent methodological advances in magnetic the neuroimaging methodology will be provided within
resonance imaging (MRI) have revolutionized our ability to this context.
study the developing human brain. This chapter examines the
use and promise of MRI in addressing key developmental
questions, including how the healthy normal brain develops Magnetic resonance imaging
and how such development is related to behavior. This meth-
odology can also help us understand the biological substrates MRI has had a dramatic impact in the diagnosis of a
of childhood disorders. Examples of studies that examine the variety of diseases and the in vivo study of the develop-
biological progression of developmental disorders following ing brain. This technique provides high spatial resolu-
treatment and remediation are provided. Used effectively, this
methodology could shed light on an array of developmental tion images of the brain based on the nuclear magnetic
questions with respect to both healthy and pathological de- resonance (NMR) properties of water protons and
velopment. other nuclei found in brain tissue (Young, 1988). NMR
consists of applying a radio frequency (RF) pulse (with
Magnetic resonance imaging (MRI), with its lack of an excitation frequency coinciding with the natural fre-
ionizing radiation and capacity to provide exquisite an- quency of the system, known as the Larmor frequency)
atomical detail, has revolutionized the study of human to the tissue. The RF pulse flips the net magnetization
brain development. Other imaging modalities, such as perpendicular to the main field. The MR image is gen-
conventional radiography, computerized tomography erated by differences in the concentration of nuclei and
(CT), positron emission tomography (PET), and single their nuclear magnetic relaxation times (T: and T2) in
photon emission computerized tomography (SPECT), the different tissue environments (Bloch, 1946; Hahn,
use ionizing radiation. Although these latter techniques 1950). Tj relaxation refers to the return of precessing
may be used with pediatric patient populations when nuclei to alignment with the main field after excitation
clinically warranted, the ethics of exposing children to (i.e., longitudinal relaxation, or spin-lattice). This re-
radioactive isotopes for the advancement of science are laxation is related to surrounding tissue composition in
less clear (Casey and Cohen, 1996; Morton, 1996; Za- that small water molecules relax faster than larger lipid
metkin, 1996). The advent of functional MRI (fMRI) molecules, thereby providing information on gray and
has further extended the utility of MRI to explore the white matter differences. T2 relaxation refers to the fall
developing human brain in ways not previously of the transverse magnetization leading to decay in the
possible. signal even though nuclei remain excited. This relaxa-
This chapter addresses applications of structural and tion time is associated with interactions among nuclear
functional MRI to the study of development. Emphasis spins and local inhomogeneities in the applied field.
is placed on the utility of MRI in understanding This interaction causes nuclei to precess, or spin, at dif-
(1) brain maturation and its relation to behavioral ferent rates and deviate from the uniform motion of
development, (2) the effects of learning on brain de- the initial excitation (Bottomley etal., 1984, 1987). The
velopment, and (3) the effects of behavioral and phar- rate of dephasing (loss of uniform motion) depends on
macological intervention on brain development (figure resonance within the environment (i.e., spins of neigh-
10.1). Examples of behavioral paradigms and details of boring nuclei). For example, T2 in free water is much
longer than that in bound water, and so prolonged T2
observed in lesions results from an increase in free/
B. J. CASEY, KATHLEEN M. THOMAS, AND BRUCE MCCAND-
LISS Sackler Institute for Developmental Psychobiology, bound water ratio. Localized inhomogeneities in the
Joan and Sanford I. Weill Medical College of Cornell Univer- applied field lead to local differences in the Larmor
sity, New York, New York. frequencies, causing a decrease in T2, which is then des-
ignated T2*. T2*-weighted images are important for un- creases with age for males (Giedd et al., 1996c). This
derstanding the basis of functional MRI, as will be latter finding may be consistent with the distribution of
described later. sex hormone receptors for these structures, with the
amygdala having a predominance of androgen recep-
BRAIN DEVELOPMENT MRI-based anatomical studies tors (Clark et al., 1988; Sholl and Kim, 1989) and the
have revealed some interesting maturational changes in hippocampus having a predominance of estrogen re-
brain structure. The most informative studies to date ceptors (Morse et al., 1986).
are those based on carefully quantified volumetric mea-
sures and large sample sizes of 50 or more subjects (e.g., BEHAVIORAL DEVELOPMENT One way of linking mor-
Giedd et al., 1996b,c; Reiss et al., 1996). The most con- phometric changes in the brain with behavior is to cor-
sistent findings across these studies include (1) a lack relate MRI-based anatomical measures with behavioral
of any significant change in cerebral volume after 5 measures. One of the first examples of such a study was
years of age (Giedd et al., 1996b,c; Reiss et al., 1996); reported by Casey and colleagues (1997a). The study
(2) a significant decrease in cortical gray matter after examined the role of the anterior cingulate cortex in
12 years (Giedd et al., 1999); and (3) an increase in the development of attention. Attention was assessed
cerebral white matter throughout childhood and young with a forced-choice visual discrimination paradigm in
adulthood (Jernigan et al., 1991; Pfefferbaum et al., 26 normal children between the ages of 5 and 16 years.
1994; Caviness et al., 1996; Rajapakse et al., 1996; Reiss Performance during attention tasks characterized as
et al., 1996). Specifically, subcortical gray matter regions predominantly automatic versus effortful was assessed
(e.g., basal ganglia) decrease in volume during child- in parallel with MRI-based morphometric measures of
hood, particularly in males (Giedd et al., 1996b; Raja- the anterior cingulate cortex.
pakse et al., 1996; Reiss et al., 1996), while cortical gray The behavioral paradigm consisted of presenting
matter in the frontal and parietal cortices does not ap- three stimuli that varied in shape and/or color in a row
pear to decrease until roughly puberty (Giedd et al., on a computer screen. The subject's task was to indicate
1999). White matter volume appears to increase which of the three stimuli was different from the other
throughout childhood and well into adulthood (Cavi- two in a forced-choice task. Subjects were not informed
ness et al., 1996; Rajapakse et al., 1996). These increases as to which feature would be salient in making the dis-
appear to be regional in nature. For example, white criminations. There were two conditions: one requiring
matter volume increases in dorsal prefrontal cortex, but predominantly automatic processing and one requiring
not in more ventral prefrontal regions (i.e., orbitofron- controlled processing. In the automatic condition the
tal cortex) (Reiss et al., 1996). Total temporal lobe vol- stimuli differed on a single attribute (e.g., color).
ume appears relatively stable across the age range of Forced-choice detection based on a single feature has
4-18 years, while hippocampal formation volume in- been suggested to be relatively automatic (Treisman,
creases with age for females and amygdala volume in- 1986). In the controlled processing condition, the
child who commanded a large sight vocabulary (able to string blocks, suggesting the involvement of phonolog-
read accurately 88% of the 100 most frequent words ical processing regions as a result of the cognitive
contained in children's books) yet fell below the 30th intervention.
percentile for his age in grapheme-phoneme decoding Although the conclusions that one can draw from the
ability as well as in measures of phonological awareness. previously described adult and child case studies are
Further behavioral testing revealed a general inability limited until they can be replicated in larger samples,
to read all but the most rudimentary pseudowords. these case studies mark a first step in addressing pro-
Functional MR images were acquired before and after cesses and cortical functions related to the initial ac-
a 12-week (24 sessions) intervention program that fo- quisition of literacy.
cused on grapheme-phoneme decoding skills via real- Pediatric fMRI research allows developmental disor-
word exemplars (McCandliss et al., 1999a). Behavioral ders such as developmental dyslexia to be studied
testing before and after the intervention revealed a dra- within the age range in which they typically first appear,
matic increase in grapheme-phoneme decoding skills, and thus provides opportunities both for early preven-
with gains exceeding 1.5 grade levels in less than a third tion/intervention and dynamic tracking of cortical
of a year. Results of the fMRI scans acquired before the changes associated with gains in cognitive skills.
intervention demonstrated little activation in the pos-
terior region of the superior temporal gyrus (figure
10.7; see color plate 12). However, following the inter- Pharmacologic MRI
vention, significant increases in this target region were
reported. This pattern was observed for both the word Having illustrated how the technique of fMRI has led
and pseudoword blocks, yet was absent in consonant to a revolution in mapping changes in the human brain
as a function of development and learning, we now turn leagues (1998), who examined patterns of brain activity
to an illustration of how we can push the methodology in children with and without ADHD on ritalin during
a step further and investigate changes in the human performance of the go/no-go task (Casey et al., 1997b)
brain in response to pharmacologic probes. A number described previously. Overall, Vaidya showed improved
of recent studies with both humans and animals have behavioral performance and an increase in prefrontal
shown the detection of neurotransmitter activity using activity for healthy children on ritalin and for children
pharmacologic MRI (phMRI) and demonstrated how with ADHD on ritalin. Significant changes in basal gan-
results from this technique are correlated with PET, mi- glia activity were observed for the ADHD children when
crodialysis, and behavioral data (e.g., Chen et al., 1997). on ritalin. Specifically, the pattern of activity in this re-
These studies have focused largely on animal work, but gion for children with ADHD on ritalin was almost iden-
more recent work illustrates the utility of the method tical to the pattern observed in the healthy children off
with humans (Braus et al., 1997; Chen et al., 1997; ritalin, suggesting normalization of activity in the basal
Kleinschmidt et al., 1997; Sell et al.,1997; Vaidya et al. ganglia with medication.
1998). Ultimately, pharmacologic MRI studies may be The previous study reports results that are reminis-
sensitive probes in elucidating brain regions involved in cent of two classic PET studies (Baxter et al., 1988;
developmental disorders. These studies may prove ex- Swedo et al., 1989) that demonstrated normalization of
tremely useful in addressing questions relevant to why brain activity in individuals with obsessive compulsive
some individuals are nonresponders to medications and disorder following pharmacologic treatment. Most in-
why some medications are less effective in children than teresting for the Baxter et al. study was that normaliza-
in adults. tion in activity occurred with either pharmacological or
An illustration of this work with a developmental behavioral intervention. Ultimately, it will be interesting
population was recently reported by Vaidya and col- to see whether fMRI studies using behavioral and phar-
ABSTRACT This chapter reviews how genetic methods are that the genes expressed uniquely in the central ner-
relevant for developmental cognitive neuroscience. Since the vous system are less polymorphic than other genes. (In
brain is the most genetically complex organ in the body, there this context, being less polymorphic means having
are important genetic contributions to brain development in
terms of both human universals and individual differences. fewer alternate forms, or alleles, where an allele of a
Methods are now available to trace the complex developmen- gene is a variant DNA sequence of that gene.) In fact,
tal pathway that runs from a DNA sequence in a gene to brain given the relatively recent evolution of some human be-
development to aspects of developing behavior. This chapter haviors, we have good reasons for supposing they are
reviews those methods and considers the implications of re- more polymorphic. Because much of human evolution
sults from such methods for fundamental issues in develop-
mental cognitive neuroscience. has involved brain evolution, we should expect the
genes expressed in the brain to be more polymorphic
Developmental cognitive neuroscience has the am- than those expressed in evolutionarily older organs.
bitious goal of elucidating the neural mechanisms un- Therefore, we expect individual differences in the al-
derlying both universals and individual differences in leles of the genes that influence human brain devel-
cognitive and behavioral development. In this chapter, opment. Consistent with this theoretical prediction,
we review the way in which genetic methods are rele- behavioral geneticists have documented moderate her-
vant, indeed crucial, for achieving that goal. By adding itability (about 0.50) for most human cognitive and per-
the word developmental to the term cognitive neuroscience, sonality traits (Plomin and McClearn, 1993).
we are committing ourselves to the proposition that Thus, individual differences in cognitive and behav-
brain-behavior relations in the mature organism can- ioral development are partly due to allelic differences
not be understood without understanding how the in polymorphic genes that influence brain develop-
brain develops and how the development of the brain ment. And because the heritability of these individual
mediates the development of cognition and behavior. differences averages about 0.50, there are also sub-
And once we take development seriously, we must also stantial environmental influences on such individual
take genetics seriously. differences.
Here 1(1) consider general issues in the relation be- Unfortunately, most studies of environmental influ-
tween genetics and cognitive neuroscience, (2) discuss ences on human development confound genetic and
two broad strategies for studying genetic effects on cog- environmental influences. Such studies do not use ge-
nitive development, (3) briefly survey behavioral and netically sensitive designs; therefore, their findings of a
molecular genetic methods, and (4) consider the im- correlation between an environmental variable and a
plications of these genetic methods for fundamental is- behavior in the developing child do not necessarily
sues in developmental cognitive neuroscience. demonstrate an environmental influence on develop-
ment (Scarr, 1992). Suppose, for instance, we find a
Genetics and cognitive neuroscience correlation between how emotion is regulated in the
mother-infant relationship and the child's later emo-
On general principles, it should not be surprising that tion regulation in peer relations. It can seem straight-
there are genetic influences on cognitive and behav- forward to conclude that the earlier social experience
ioral development. Of the roughly 105 structural genes shaped the later social behavior. However, if the sample
in the human genome, approximately 40% are uniquely consists solely of biological mothers with their infants,
expressed in the central nervous system (Hahn, van the study's design confounds genetic and environmen-
Ness, and Maxwell, 1978), whereas the remaining 60% tal influences: Such mother-infant pairs share early
are expressed both in the central nervous system and social experiences, but they also share half their segre-
elsewhere in the body. We have no reason to suppose gating genes (genes that vary across individual hu-
mans). Consequently, the observed correlation could
BRUCE F. PENNINGTON Department of Psychology, University be partly, or even totally, mediated by genes. Thus we
of Denver, Denver, Colorado. need genetically sensitive designs not only to study
TABLE 11.1
Biometric (indirect) methods
the models tested and their background assumptions. of this regression in relatives varies as a function of their
A good introduction to twin and adoption methods is degree of genetic relation to the proband, we have "dif-
provided by Plomin and colleagues (1997). ferential regression to the mean." A finding of differ-
To study the etiology of developmental psychopa- ential regression to the mean indicates that there is
thology, we need to apply twin and adoption methods genetic influence on the trait, all other things being
to the analysis of extreme variations in behavior rather equal. Thus, in the case of twin pairs selected because
than the analysis of variance across the whole distribu- at least one twin has an extreme score, comparison of
tion. Extreme variations may be treated as categories, MZ and DZ cotwin means provides a test of whether
in which case the appropriate method of analysis is non- there is differential regression to the mean. If the DZ
parametric and involves comparing the rates of con- cotwin mean is significantly closer to the population
cordance for different degrees of relationship. Twin mean than the MZ cotwin mean, then there is evidence
concordance studies have found genetic influences on of genetic influence on the extreme trait.
most psychiatric disorders (Plomin et al., 1997; Plomin, The degree of genetic influence on the extreme trait
Owen, and McGuffin, 1994). is h2g, where the subscript g refers to the extreme group
However, if the trait in question is quantitative (can of probands. So h2g measures the heritability of the
be measured using an ordinal or interval scale), treat-
group deficit (or talent), in contrast to h2, which mea-
ing the trait as categorical does not use all the infor-
sures the heritability of variations across the whole dis-
mation contained in the data and is thus less statistically
tribution. The comparison of h2 and h2g provides one
powerful. If we can measure not only whether someone
has ADHD or reading disability, but also how severely test of whether the etiology of normal variations differs
he or she has it, then our data are quantitative rather from the etiology of extreme ones—a fundamental is-
than qualitative and can be analyzed with parametric sue in developmental psychopathology. However, be-
statistics. A particularly powerful twin method for ana- fore the advent of the DF method and molecular
lyzing extreme variation in quantitative traits was devel- methods, empirical methods to address this issue were
oped by DeFries and Fulker (1988) and is now called mostly lacking.
the DF method in their honor. The DF method is based The advantage of the DF method is that it provides a
on the phenomenon of differential regression to the statistically powerful method of modeling these com-
mean, first described by Francis Gal ton (1869). Let us ponents of the etiology of the scores of an extreme
now examine the logic of the DF method. group. Another advantage of this method is its flexibil-
Suppose we have an index case or proband who has ity; it can readily be extended to (1) the bivariate case,
been selected because he/she is extreme on some di- where we are interested in the etiologic basis of the phe-
mension; this proband is thus at the low or high end of notypic correlation between two extreme traits; (2) an
the distribution for that trait. A relative of this proband evaluation of differential etiology (of subgroups, such
will have, on average, a score on this trait distribution as males and females, or extreme and normal varia-
that is closer to the mean than the proband's score; this tions); or (3) a test for linkage. Let us briefly examine
is an example of regression to the mean. If the degree the bivariate extension.
steps: computing the net input to the unit, then com- ure 12.2), based on a sigmoidal activation function of
puting the unit's activity as a function of the net input. the following form:
The inputs to a unit are weighted by the strength of the
connections from the sending units; the stronger the
connection, the more the sending unit activity contrib- where aj is the activation of the unit, and hj is its net
utes to the net input to the receiving unit. Mathemati- input. This S-shape reflects two important aspects of
cally, the net input to a unit j (hj) is expressed as neural activity regarding the nonlinear response of neu-
rons in relation to their inputs. First, the unit is not
guaranteed to become active just because it is receiving
some amount of input. As indicated by the lower left
where Wij is a weight from unit i to unit j, and ai is the part of the S-shaped curve, this net input must exceed
activity of unit i. a certain threshold for the unit to become very active.
The activation function specifies how the units in a Second, once the unit is active to some degree, it is not
network update their activity as a function of this net guaranteed to become much more active with increas-
input. Activation functions are typically S-shaped (fig- ing amounts of input. As indicated by the upper right
ABSTRACT Brain development progresses through a series timately the behavior, are modulated. Because most of
of steps that are genetically programmed but influenced sig- what we know about the changes in the normal and
nificantly by environmental events, including injury. The mor- injured brain come from studies of the rat brain, our
phological and functional effects of brain perturbations
during development vary qualitatively with the precise embry- discussion focuses on a consideration of the rat. We are
ological events occurring at the time of injury. Recovery is confident, however, that our results will generalize to
especially poor if the cortex is injured immediately following other mammalian species, and especially to humans.
the completion of neurogenesis, and is especially good if in-
jury occurs during the time of maximum astrocyte prolifera-
tion and synapse formation. The young injured brain is Changes in the normal brain
especially sensitive to environmental events, and recovery thus
can be enhanced by various forms of stimulation such as tactile In order to understand how the brain can be changed
stroking in infancy and complex housing as juveniles and ad- to support functional restitution, we must first consider
olescents. Recovery is also modulated by gonadal hormones how the normal brain can be changed. The general
and by neuromodulators (noradrenaline and acetylcholine). logic is that the nervous system is conservative, and that
plastic changes that normally occur during develop-
The goal of this chapter is to illustrate some of the ment are likely to be recruited in an attempt to repair
general principles underlying how injury to the devel- the abnormal brain.
oping brain, and particularly to the cerebral cortex, can
lead to alterations in brain and behavioral development NEURONAL CHANGES DURING DEVELOPMENT There are
and how such alterations can be modulated. In princi- two aspects of neural development that are especially
ple, there are three ways that the brain could show important in the current context. First, neurons of the
plastic changes that might support recovery during neocortex of rats are generated from about embryonic
development. First, there could be changes in the or- day 12 to 21, birth being on day 22. Thus, the genesis
ganization of the remaining, intact, circuits in the brain. of cortical neurons begins about two-thirds of the way
This would likely involve the generation of new synapses through gestation and is complete at about the time of
in extant pathways. Second, there could be a develop- birth. Neurons and glia arise from neural stem cells that
ment of new circuitry that is novel to the injured brain. reside in the ventricular zone of the developing brain,
Third, there could be a generation of neurons and glia as shown in figure 13.1. Neural stem cells can divide
to replace at least some lost neurons and glia. In fact, symmetrically, to produce two stem cells, or asymmet-
the developing brain makes use of all of these changes, rically, to produce a stem cell and a progenitor cell.
although the details vary with the precise developmen- Stem cells can be thought of as multipotent cells that
tal age at the time of injury. Furthermore, each of these have the potential to reproduce themselves continu-
outcomes can be influenced by various modulating fac- ously throughout the lifetime of an organism. In con-
tors, including especially experience, neuromodulators, trast, progenitor cells have a limited capacity for
and gonadal hormones. reproduction and are destined to produce neurons or
We begin by considering how the normal brain glia. Stem cells are located in the subependymal zone
changes in response to experience during development and remain active throughout life, although they may
and then consider each of the three types of changes have a finite number of divisions before they die. It is
that occur in the developing injured brain. Finally, we believed that progenitor cells that can divide to produce
consider the way in which the plastic changes, and ul-
neurons and/or glia can migrate away from the sub-
BRYAN KOLB AND ROBBIN GIBB Departments of Psychology ependymal zone and may lie quiescently in the white or
and Neuroscience, University of Lethbridge, Lethbridge, gray matter. While in these locations they can be acti-
Canada. vated to produce neurons and/or glia. (For a useful
review, see Weiss et al., 1996.) One challenge with stem developing brain, which, as we shall see, can lead to
and progenitor cells is to find the "switch" to turn on enhanced cortical regeneration and functional recov-
controlled cell production when they are needed after ery (Kolbetal., 1998).
an injury. Considerable evidence now suggests that the The second important stage of neural development
mammalian brain, including the primate brain, can is the formation of synapses. As neurons migrate to
generate neurons destined for the olfactory bulb, hip- their final destinations, they begin to develop axons and
pocampal formation, and even the neocortex of the dendrites that will form synapses. Synapses do not form
frontal and temporal lobes (Eriksson et al., 1998; Gould randomly on cortical pyramidal neurons but show im-
et al., 1999; Kempermann and Gage, 1999). The reason portant characteristics. One feature is that most excit-
for this generation is not at all clear at present, although atory inputs (up to 95%) synapse on spines, which are
it may function to enhance brain plasticity, particularly found on the dendrites but rarely on or near the cell
with respect to processes underlying learning and mem- bodies. This means that it is possible to estimate excit-
ory. It turns out, however, that although the generation atory synaptic numbers by estimating the number of
of these neurons occurs continuously in the normal spines, which is typically done by estimating both the
brain, the generation rate is influenced by injury in the length of dendritic material and the density of spines
FIGURE 13.2 Representative layer III pyramidal neurons age. Cells from the condominium-housed animals showed
from Zilles's area Parl in young adult rats that were housed more dendritic arbor but a reduced spine density relative to
in the condominiums from weaning until about 4 months of cells from lab-housed animals.
Brain development after early brain injury though the behavioral outcome is far better (Kolb and
Whishaw, 1981). Therefore, it must be the organization
In the course of studying the functional effects of early of the brain rather than its size that predicts recovery in
cortical injury in rats, we noticed that brain size in adult- the day-10 animal. Changes in organization can be in-
hood was directly related to the postnatal age at injury: ferred from an analysis of synaptic numbers, cortical
The earlier the injury, the smaller the brain and the connectivity, and evidence of neuro- and gliogenesis.
thinner the cortical mantle. Thus, rats with perinatal
lesions have very small brains whereas those with lesions SYNAPTIC SPACE By staining brains with a Golgi-type
at day 10 have larger brains. Curiously, however, the stain it is possible to measure and quantify the dendritic
day-10 brains still are markedly smaller than the brains structure of neurons. Although dendritic length does
of rats with lesions later in life, such as day 25, even not provide a direct measure of synapse number, den-
dritic length and spine density can be used as a reason- although only the youngest animals showed the en-
able estimate of synaptic number. Golgi analyses of hanced ipsilateral connections, the day-10 animals
cortical neurons of rats with perinatal lesions consis- showed the best functional outcome. Furthermore, ani-
tently show a general atrophy of dendritic arborization mals with day-4 lesions showed unexpected deficits on
and a drop in spine density across the cortical mantle cognitive tasks, such as the acquisition of a spatial nav-
(Kolb, Gibb, and van der Kooy, 1994). In contrast, rats igation task. It thus seems likely that the aberrant
with cortical lesions around 10 days of age show an in- corticospinal pathway interfered with the normal func-
crease in dendritic arbor and an increase in spine den- tioning of cortical areas that would not ordinarily be
sity relative to normal control littermates. Thus, animals involved in motor function. In a parallel set of studies
with the best functional outcome show the greatest syn- we showed that damage to the medial prefrontal region
aptic increase whereas animals with the worst functional produced a similar result: Animals with frontal lesions
outcome have a decrease in synapses relative to control on day 1 showed massive changes in cortical connectiv-
animals. Furthermore, factors that act to increase ity, but these animals had the worst behavioral outcome
dendritic space also enhance functional outcome (Kolb, Gibb, and van der Kooy, 1994). Furthermore, we
whereas those that act to decrease dendritic space act showed that the abnormal pathways did not reflect the
to retard functional outcome. creation of new connections so much as they reflected
a failure of pruning of connections that are normally
CHANGES IN CORTICAL CONNECTIVITY Perhaps the discarded during development. This was demonstrated
most extensive studies of changes in cortical connectiv- by our finding that newborn animals have extensive ab-
ity are those showing that after unilateral motor cortex errant pathways that die off during the first week of life.
lesions in infant rats or cats there is a major expansion If the cortex is damaged during this time, however,
of the ipsilateral corticospinal pathway from the undam- some of these pathways fail to die off, leaving the animal
aged hemisphere, which is correlated with partial re- with apparently novel circuitry that is not seen in nor-
covery of skilled forelimb use (Castro, 1990; Whishaw mally developing animals. Some of this unusual cir-
and Kolb, 1988). The initial studies concluded that cuitry could prove helpful after an injury; but given that
these aberrant connections were advantageous and normal animals shed such circuitry, it seems likely that
likely provided an explanation for the Kennard effect. it could prove equally disadvantageous to maintain such
It appears, however, that the anomalous corticospinal circuitry. Indeed, both hypotheses are confirmed. Rats
projections may be formed at a significant cost. For ex- with infant motor cortex lesions do show sparing of
ample, when we compared the effects of motor cortex some motor skills (Whishaw and Kolb, 1988) but ap-
lesions on postnatal days 4, 10, and 90, we found that parently at the price of impairments in other cognitive
FIGURE 13.6 Photographs of brains of representative control been filled with newly generated neurons in the lesion brain.
(left) and medial frontal lesion (right) rat. The lesion cavity has Arrows point to the lesion scar.
feral rats; thus, if we try to generalize to humans, we are increased acetylcholine levels. We reasoned that if this
left wondering how stimulating an environment would were so, then increasing acetylcholine levels more di-
have to be in order to induce functional changes in rectly might also enhance recovery. As shown by others,
human infants. One way to answer this question is to choline-enriched diets could alter cortical plasticity
consider what animals do in the complex environments. (Pyapali, 1998); therefore, we placed pregnant dams on
They are provided with novel objects to interact with a choline-enhanced diet during the last two weeks of
every few days, they get a lot of activity as they move pregnancy and kept the lactating mothers on the diet
about the compounds, and they have considerable so- until weaning (Tees and Kolb, in preparation). The
cial experience as they live in groups of 6-8 other ani- pups received medial frontal lesions on postnatal day 4
mals. The experience is thus perceptually and socially and were assessed on various behavioral tasks in adult-
stimulating, motorically demanding, and is continuous hood. Choline treatment significantly ameliorated the
for several weeks or months. Although we do not know expected functional deficits; in addition, there was a
which aspects of this experience are the most impor- significant increase in dendritic branching and spine
tant, our guess is that all contribute because they stim- density in the choline-treated lesion animals. In a
ulate different types of brain activity. Thus, our best follow-up study, we took the opposite approach and de-
guess as we generalize to human infants is that therapies pleted the forebrain of cholinergic neurons with intra-
would include perceptual stimulation, including nov- cerebral injections of IgG saporin on postnatal day 4,
elty, as well as social interaction and motor activities. then gave the animals frontal lesions on P7 (Kolb, Gibb,
Sherren, and Pappas, unpublished observations). In
NEUROMODULATORS Both acetylcholine and nor- this case, we blocked the functional recovery normally
adrenaline have been implicated in various forms of seen in P7 operates. Our anatomical analysis of the neu-
cortical plasticity (Bear and Singer, 1986), and as we rons of these animals is in progress, but we predict that
shall see, both modulate recovery from early cortical the cholinergic blockade will have the opposite effect
injury. We consider each separately. of the cholinergic stimulation—namely, an atrophy in
In the course of doing our experiments on tactile dendritic arborization relative to untreated controls.
stimulation and recovery, we noticed that tactile stimu- There are now many studies showing that noradrena-
lation during infancy produced a chronic enhancement line (NA) is necessary for various forms of experience-
of acetylcholinesterase density in the cortex (Kolb and dependent cortical plasticity (Kasamatsu, Pettigrew, and
Gibb, 1999). In addition, this increase was significantly Ary, 1979). We therefore depleted newborn rats of fore-
greater in the animals with frontal removals than in brain NA, which can be accomplished with subdural
sham controls. This result led us to speculate that the doses of the neurotoxin 6-hydroxydopamine in the
effect of tactile stimulation might be mediated through young animal. This is possible because the forebrain
GONADAL HORMONES There is accumulating evidence FIGURE 13.7 Summary of the Morris water task performance
that male and female brains differ in their structure, of rats. (A) Rats with frontal lesions on postnatal day 4 and
respond differently to environmental events, and re- pre- and postnatal choline supplements in their diets. The
spond differently to injury. For example, we have found choline supplements reduced the behavioral deficit (Tees and
that there are hormone-related structural differences in Kolb, 1999). (B) Rats with frontal lesions on postnatal day 7
and prelesion noradrenaline depletion. The noradrenaline
the prefrontal cortex of rats. Cells in the medial pre- depletion blocked the behavioral recovery (Kolb and Suth-
frontal regions have more dendritic arbor in animals erland, 1992).
exposed to testosterone, whereas cells in the insular
prefrontal regions have more dendritic arbor in animals
not exposed to testosterone (Kolb and Stewart, 1991).
These differences led us to wonder if early brain injury greater increase in spine density whereas females show
might influence the pattern of sex-related differences a larger increase in dendritic arborization (Kolb and
in cortical morphology. It does. As we have seen, rats Stewart, 1995). Thus, although both sexes show func-
with medial frontal lesions in postnatal day 7 show tional recovery, there are differences, and these differ-
marked functional recovery; but, in addition, this re- ences are related to morphological differences. To
covery is sexually dimorphic (figure 13.8). In particular, complicate matters more, there appear to be sex differ-
males show enhanced recovery of spatial-navigation be- ences in the effect of tactile experience or complex
haviors whereas females show enhanced recovery of housing on cortical neurons (Kolb, Gibb, and Gorny,
skilled forelimb reaching (Kolb and Cioe, 1996; Kolb, 2001). Thus, the question arises as to whether sex, ex-
Gibb, and Gorny, 2000; Kolb and Stewart, 1995). These perience, and lesion might all interact. This is an im-
behavioral differences are associated with differences in portant question if we are to design treatments for
cortical plasticity: Males with frontal lesions show a children with brain injuries.
ABSTRACT This chapter illustrates that learning and expe- also been assembled showing that the functional mat-
rience affect the development of perceptual and cognitive uration of the brain and the development of intellectual
abilities through mechanisms of reorganization of functional and emotional functioning are substantially influenced
brain architecture, that is, through neural plasticity on a mac-
roscopic scale. We begin by imparting some fundamentals by social and emotional experiences—learning pro-
about the mechanisms that allow the brain to organize and cesses that can occur early in life. Experiences may be
reorganize its functional structure. Then we describe basic as- beneficial, as is the case with environmental enrichment
pects of neural plasticity that underlie experience-dependent and (parental) language models; but they may also be
plasticity, focusing on cortical representational cortex as the disadvantageous, as in cases of parental loss, neglect,
best investigated model of macroscopic plasticity. Examples
for experience-induced cortical reorganization with its per-
lack of care, or even overuse of a certain skill or func-
ceptual and cognitive correlates and possible adaptive and tion. Particularly, early experiences play an important
maladaptive consequences follows. In reviewing the develop- role in the development of social and emotional func-
ment of language through the lens of neural plasticity, we tions as well as for perceptual, linguistic, and cognitive
point out elements of an emerging new paradigm in devel- development. Animal studies (Braun, 1996; Meaney et
opmental cognitive neuroscience. al., 1996; Meaney et al., 1988; Seckl and Meaney, 1993)
have demonstrated that early postnatal experiences,
Traditional concepts suggest that brain development
such as brief separation from the mother or the inten-
during the prenatal and early postnatal period follows
sity of maternal care (licking and grooming; Chaldji et
rigid rules whereby phases in development are initiated
al., 1998), affect structural, neural, endocrine, and
through a cascade of genetically determined programs.
functional brain development. These findings have con-
However, the impact of experience is significant. Even
sequences for various stress responses and learning abil-
in adulthood, the brain has the potential to reorganize
ities in later life. Human studies suggest that the impact
the functional organization of its neural networks de-
of experience and learning on later abilities is not
pending on the particular demands and afferent input.
confined to very early periods. For instance, Ramey
During development, the cortex may alter its functional
organization in conjunction with structure in response and Ramey (1998; see also Wickelgren, 1999) reported
a positive effect of preschool training on school
to experience. Perception and skills will vary with the
changes in neural network architecture. On a micro- achievement.
scopic scale, structural changes include alterations in Less is known about the influence of early experience
synaptic efficacy, synapse formation, synaptic plasticity, on intellectual and cognitive functioning, despite at-
spine density, and alterations in dendritic length. As a tempts to describe cognitive development in detail
consequence, supportive tissue elements such as astro- (Weinert, 1998; Weinert and Schneider, 1998). In con-
cytes and blood vessels are also changed (for review, see trast to the prevailing view of the last decades, there is
Kolb and Whishaw, 1998). growing evidence that the environment early in life af-
Many insights about aspects of neural plasticity have fects intellectual capacity as measured by the intelli-
been gained from studying representational zones. Evi- gence quotient (IQ) (for a brief summary, see
dence has been accumulated that the representational Wickelgren, 1999). An impact of early experience on
maps in the cortex have the ability to alter their func- cognitive abilities can also be deduced from animal
tional organization on a macroscopic range. And not studies. Meaney and colleagues (1988) observed im-
only are they highly plastic during development; they paired learning behavior in adult rats that had been
also maintain limited ability to respond to alterations in briefly separated from their mothers in the preweaning
afferent input throughout life. Increasing evidence has period. Braun (1996) reported that early sensory and
social deprivation affect synaptic restructuring of asso-
THOMAS ELBERT, SABINE HEIM, AND BRIGITTE ROCK- ciative networks, thereby preventing normal cognitive,
STROH Department of Psychology, University of Konstanz, affective, and social development. Reports of delayed
Konstanz, Germany. intellectual development and cognitive deficits in
meaning in one's native language. This alteration oc- 1996; Kamhi and Catts, 1989). For many children with
curs in linguistic experience and provides the basis of SLI and dyslexia, deficient auditory phoneme process-
rapid and effortless language acquisition. By the time ing has been confirmed by psychoacoustic studies
of their first birthday, infants speak several words; by the (Reed, 1989; Stark and Heinz, 1996; Tallal and Piercy,
age of two years, they have a vocabulary of about 50 1974, 1975; Tallal and Stark, 1981; Tallal, Stark, and
words and can produce two-word or even multiword Mellits, 1985a; for reviews see Farmer and Klein, 1995;
combinations (Bates, O'Connell, and Shore, 1987). But Tallal, Miller, and Fitch, 1993). Tallal and colleagues
if linguistic experience is not forthcoming in an early assumed an impaired integration of brief and rapidly
period of life, speech and language development are changing sounds to be at the basis of SLI, as deficits in
hampered. Deaf children who have never been exposed stop consonant perception turned out to be correlated
to either spoken or manually coded speech (Goldin- with language comprehension scores of SLI children
Meadow, 1997) or children who experience severe dep- (Tallal, Stark, and Mellits, 1985b). Furthermore, Kraus
rivation of social stimulation by humans (Brown, 1966; and colleagues (1996) observed a nearly absent mis-
Curtiss, 1977) do not acquire language properly. This match response to deviant stop consonant-vowel sylla-
indicates a "sensitive" period for the development of bles in language- and learning-impaired children.
language ("sensitive" implying the potential to acquire Moreover, the physiological dysfunction was corrobo-
speech discrimination, thus the potential of neural plas- rated by the impairment in discriminating these pho-
ticity, in contrast to "critical," which implies invariance; nemes (viz. /da/ versus /ga/). A relatively attenuated
see, for example, Tees and Werker, 1984). MMN to the phonetic contrast /ba/-/da/ in dyslexic
However, even if not suffering from hearing loss or boys was also reported by Schulte-Korne and colleagues
social isolation, some preschool children have great dif- (1998). From the explication of the MMN outlined
ficulty in acquiring language. In many cases, language above, it may be concluded that the neural memory
disabilities cannot be attributed to mental retardation, traces do not serve as reliable recognition patterns for
psychiatric disorders (e.g., childhood schizophrenia, in- rapidly changing phonetic contrasts in children with
fantile autism), or neurological causes (e.g., focal brain SLI and dyslexia. Determining the sources of the stop
lesions, seizure disorders). These language-based learn- consonant-vowel syllables /da/ and /ga/ in the audi-
ing disorders are referred to as specific language im- tory cortex of children with dyslexia from magneto-
pairment (SLI). When they reach school age, children encephalographic recordings, Heim and colleagues
with SLI are at risk for learning problems similar to (2000) found that the brain structures generating the
those seen in dyslexics (Aram, Ekelman, and Nation, magnetic waves at 200 ms were located about 1.5 cm
1984; Tallal, Curtiss, and Kaplan, 1988). Conversely, more anterior in children with dyslexia than in normal
children with developmental dyslexia are typically controls. This difference was also evident relative to the
found to be deficient in some linguistic tasks (see Catts, source location of the earlier component at 80 ms,
ABSTRACT The human auditory system undergoes dramatic conspired to constrain access to the functioning of the
improvement in sensitivity between the onset of function in auditory system. In addition, auditory phenomena only
the last trimester of gestation and the end of the first postnatal slightly more complex than basic psychoacoustic capac-
year. These early limitations may constrain various aspects of
sound discrimination and learning in young infants. Despite ities have not been easily incorporated into a general
these limitations, the newborn is capable of rapid learning of model of auditory processing.
suprathreshold auditory stimuli, rudimentary localization of How can a cognitive neuroscience perspective facili-
sound, and elementary processing of complex auditory pat- tate progress in understanding the developing auditory
terns such as speech. This chapter reviews these basic auditory system? As Johnson (1997, p. xv) pointed out, "Devel-
capacities, examining how various brain mechanisms process
both simple and complex sounds, and how these processing opmental cognitive neuroscience is specifically con-
mechanisms are elaborated by maturation and experience. cerned with understanding the relation between neural
and cognitive phenomena." It is hard to imagine a time
Several weeks before the human newborn is first ex- when studies of sensory and perceptual development
posed to visual stimulation, auditory inputs are capable ignored underlying neural substrates. Yet those sub-
of stimulating hair cells on the basilar membrane, gen1 strates remain largely inaccessible except in animal
erating neural signals that travel from the auditory studies and recent attempts at various forms of nonin-
nerve to brainstem, thalamic, and cortical areas of the vasive neuroimaging. The challenge is to meld together
central nervous system, triggering behavioral responses, findings from invasive neuroscience, noninvasive neu-
and establishing rudimentary representations that can roimaging, and behavioral studies of human infants and
affect behavioral preferences several days or weeks later. children to provide a coherent account of the mecha-
Newborns also exhibit preferences for certain classes of nisms by which auditory processing develops.
auditory stimulation and the ability to orient toward In this chapter we provide a summary of the devel-
sounds, neither of which could have been induced by opment of basic auditory capacities in infancy, and we
early experience. Because the apprehension of auditory discuss how infants perceive more complex auditory
stimulation does not require an overt attentional re- stimuli and how neural specializations for processing
sponse (e.g., turning the eyes to fixate a visual stimulus both simple and complex auditory signals emerge dur-
directly), ambient auditory information is readily avail- ing development. We also review evidence of functional
able to the developing infant. Given this combination plasticity in the processing of auditory signals, both dur-
of early maturation and ubiquitous stimulation, the au-
ing development and in adulthood. Our goal is to pro-
ditory system would seem to be an ideal candidate
vide a selective overview of classic and recent findings
for studying the sensory and perceptual mechanisms,
and their neural correlates, that enable cognitive on auditory system development that can serve as a
development. guide for future investigations of developmental cog-
Why, then, has relatively little attention been paid to nitive neuroscience in the auditory modality.
the development of the auditory system compared to
the visual system? And why do we know relatively little Fundamental auditory capacities in infancy
about how auditory information is processed by the de-
veloping brain? As we shall see, a variety of technical, The intricacies of the development of the auditory pe-
methodological, and neuroarchitectural factors have riphery and the auditory pathways are beyond the scope
of the present chapter. Fortunately, the interested
RICHARD N. ASLIN AND RUSKIN H. HUNT Department of Brain reader can consult several recent reviews that cover
and Cognitive Sciences, University of Rochester, Rochester, these issues in great detail (Rubel, Popper, and Fay,
New York. 1998; Werner and Marean, 1996; Werner and Rubel,
ABSTRACT Anatomical, physiological, and functional data what Teller (1984) referred to as linking propositions in
on early visual development are reviewed by discussing models visual science—formal statements that attempt to link
of brain-behavior relationships during this period. Models aspects of visual anatomy and physiology with observed
have been formulated at the retinal, subcortical, and cortical
levels. Constraints based on front-end (optical and photore- visual capacities or behaviors.
ceptoral) limitations have been reasonably successful in ex- A common theme that emerges from these models is
plaining the data on the development of spatial contrast the idea that different pathways within the visual system
vision. Cortically based models have relied on putative differ- may develop at different rates (Atkinson, 1998; Banton
ences in the rates of maturation of various subpathways within and Bertenthal, 1997; Johnson, 1990). For example, At-
the visual system to explain the order of emergence of various
visual functions. These cortically based models are inherently kinson (1998) proposed that the magnocellular path-
more difficult to evaluate because of the complexities associ- way may lag slightly behind the parvocellular pathway
ated with brain-behavior relationships at this level of the vi- early in postnatal development. Johnson (1990) and
sual system. Formulating developmental versions of these Banton and Bertenthal (1997) based their models on
cortical brain-behavior relationships adds another level of data showing that different layers within primary visual
subtlety to the problem when subpathways within the visual
system develop "on-line" postnatally. cortex develop at different rates. On a simpler level, it
has been proposed primarily on the basis of behavioral
Models of brain-behavior relationships in early visual data that the short-wavelength sensitive (sws) cones may
development have focused on several levels within the lag slightly in development over the first several weeks
visual pathway. These models address development at of postnatal life behind the medium- and long-wave-
the retinal (Banks and Bennett, 1988, 1991; Candy, length sensitive (mws and lws) cones (Banks and Ben-
Crowell, and Banks, 1998; Fulton et al., 1999; Hansen nett, 1988). It is important to realize that these models,
and Fulton, 1999), subcortical (Bronson, 1974), or cor- especially the cortical level ones, are arguing that vision
tical levels (Atkinson, 1984, 1992; Banton and Berten- in the early postnatal human infant is not simply a
thal, 1997; Johnson, 1990; Johnson and Vecera, 1996) scaled version of vision in the mature adult. It is not
and have attempted to link observed data on visual simply that there is more noise in the young visual sys-
function, whether behavioral or evoked electrical activ- tem or that the signals are generally weaker, but rather
ity, to properties of the underlying neural substrate. that the kinds of information available to the infant for
Such models of brain-behavior relationships in early purposes of object recognition and action may differ in
visual development typically make inferences either kind from that available to normal human adults.
from the known anatomy or physiology to observed be- The purpose of this chapter is to review the available
havior or from observed behavior to the anatomy and data on the early postnatal development of the visual
physiology. In the former case, data on the states of vari- system. I focus on the theme just noted—that different
ous neural elements within the visual system of human parts of the visual pathways develop at different rates.
or primate infants are used to explain observed behav- To organize this review, I discuss these various extant
ioral data (e.g., acuity or contrast sensitivity). In the lat- models of early visual development, then review the an-
ter case, observed behavioral data are used to make atomical and physiological data from human and non-
inferences about the presence and functioning of vari- human primate visual development in the context of
ous subpopulations of feature-selective neurons within discussing these models. I do not review the extensive
the visual pathway (e.g., orientation or direction selec- literature on plasticity and deprivation in early visual
tivity). These models are developmental examples of development (Daw, 1998; Pizzorusso and Maffei, 1996).
The interested reader may consult the chapter by
JAMES L. DANNEMILLER Department of Psychology, Univer- Maurer in this volume for a discussion of visual system
sity of Wisconsin at Madison, Madison, Wisconsin. plasticity.
ity. In contrast, models involving cortical function tend the corresponding cortical data. It is easier to make pre-
to make qualitative predictions. These models are con- dictions based on the lengths of cone outer segments
cerned with predicting the presence or absence of some than it is to make predictions based on the complexity
perceptual capacity thought to depend on cortical pro- of synaptic connectivity.
cessing (e.g., orientation sensitivity). Second, the ana- One exception to this statement that should be noted
tomical data on which the retinal models are based are is the attempt by Wilson (1993) to relate cortical
generally easier to relate to visual performance than are synaptic density, as reported by Huttenlocher and
terior parietal areas. The parvocellular or temporal- pathway from retina to visual cortex. Atkinson (1984)
directed stream is associated with areas V4 and echoed this proposal that much of the observed visual
inferotemporal cortex. It is important to keep in mind behavior in the first month is controlled subcortically
that this division into two primary streams does not im- and added the additional postulate that the descending
ply complete independence between the processing of pathways from visual cortex to superior colliculus ma-
information within these streams. As is evident below, tures later, leading to changes around 2 months in
even in area VI there is evidence for crosstalk between certain visual capacities (e.g., attentional switching,
these two streams. binocular convergence).
Given this brief review of the visual pathways, I now Using resting positron emission tomography (PET),
turn to the developmental models based on subcortical Chugani and Phelps (1986) concluded that prior to
and cortical processing. Bronson (1974) first proposed 3 months various subcortical areas (e.g., thalamus,
that early in postnatal life much of the visual capacity midbrain-brainstem) were probably more functionally
of the neonate could be explained by supposing that mature than various cortical areas (e.g., occipital, pari-
the subcortical pathway from retina to superior collic- etal, and temporal). Different cortical areas also mature
ulus matured or was functional earlier than the cortical earlier than others (e.g., temporal prior to frontal;
1986; Courage and Adams, 1990; Dobson et al., 1987; for normal adults is shown by the upper curve in figure
Mayer and Dobson, 1982; Miranda, 1970; van Hof-van 17.2. Contrast sensitivity peaks at 3-5 c/deg, drops off
Duin and Mohn, 1986), it improves rapidly over the first gradually for lower spatial frequencies (wider stripes),
few months of life, so that by 6 months of age it is only and declines sharply for higher spatial frequencies (nar-
about 8 times worse than that of a normal adult. There- rower stripes).
after, grating acuity improves more gradually and Several groups of investigators have used preferential
reaches adult values by 4-6 years of age (Courage and looking to measure contrast sensitivity in young infants
Adams, 1990; Mayer and Dobson, 1982; van Hof-van (Adams et al., 1992; Atkinson, Braddick, and Moar, 1977;
Duin and Mohn, 1986). Banks and Salapatek, 1978, 1981; Gwiazda et al., 1997;
Peterzell, Werner, and Kaplan, 1995). As with grating
CONTRAST SENSITIVITY Grating acuity provides a mea- acuity, stripes of a particular spatial frequency and con-
sure of the smallest visible stripe size. However, spatial trast are paired with a plain gray stimulus, and the tester
vision is limited not only by size, but also by the differ- looks for evidence that the baby has detected the stripes
ence in luminance between objects and their back- (e.g., that the baby looks first or most toward the side
ground. High contrast objects are more easily detected where the stripes appear). Over trials, contrast is varied
than are lower contrast objects, and the minimum to determine the minimum contrast for which babies
amount of contrast necessary to resolve an object varies show a preference for stripes of that spatial frequency.
with the size of the object. The contrast sensitivity func- The process is then repeated at different spatial fre-
tion plots contrast sensitivity (the inverse of the mini- quencies to derive the contrast sensitivity function.
mum contrast necessary to resolve the pattern) for All investigators agree that contrast sensitivity is very
stripes of various sizes (or, more accurately, for sine immature during early infancy. The lower curve in fig-
waves of different spatial frequency). A typical function ure 17.2 shows typical results from 2-month-olds. Al-
ABSTRACT The process by which new motor skills emerge new forms of movements arise, but also how movements
has been a fundamental question in developmental psychol- become more accurate, faster, more graceful, and more
ogy. One issue that must be faced when learning a new skill finely tuned to the environment (Adolph, 1997;
is the requirement that skilled movements be both stable and
flexible. In this chapter, we explore how infants resolve the Adolph, Eppler, and Gibson, 1993; von Hofsten, 1979;
inherent tension between newly emerging stability and flexi- Thelenetal., 1993).
ble adaptation. We illustrate these dual themes by examining In this chapter, we focus on a related aspect of motor
how infants learn to reach out and grab objects. We show that skill: the dual requirements that skilled behavior be
infants begin learning to reach with flexibility in the neuro- both stable and flexible. Consider for a moment the
motor system, as well as constraints on coordination patterns.
This combination allows for multiple solutions while still lim- achievements of baseball legend Mark McGuire, who
iting the nearly endless possibilities. Furthermore, we show broke the Major League record for home runs in a sin-
that stability and flexibility can be thought of as part of the gle season. To break that record, McGuire had to be
same dynamic system, reflecting the interconnectedness and consistent in the timing, placement, and force of the
redundancy of the neuromotor system.
swing of the bat. His skills were stable and repeatable.
But no two pitches are exactly the same. So McGuire
During their first year of life, human infants experi-
ence a dramatic transformation in their abilities to con- also had to be flexible in order to adjust his swing to hit
trol their bodies. Newborn infants are prisoners of the balls pitched at different angles and different speeds.
forces of gravity, unable even to raise their heads. Yet Most remarkably, the adaptation of the swing must be
after just one year, they are able to sit, reach for and done largely "on-line," only after the ball leaves the
manipulate objects, crawl, stand, walk, climb, gesture, pitcher's hand with a specific combination of speed and
and sometimes even speak a few words. These new skills direction. Estimates are that a batter has but 200 milli-
mark changes across all the systems of the body. More- seconds or so to adjust his swing to the oncoming ball.
over, these abilities completely reorganize infants' re- But despite such a small window of opportunity, Mc-
lationships with their physical and social worlds. The Guire was repeatedly able to make the necessary ad-
ability to control the head enables babies to choose justments to his swing to contact the ball with enough
what to look at. The ability to sit erect frees the hands force to hit it out of the ballpark.
for exploring and manipulating objects. The ability to Most tasks in everyday life do not require the speed
gesture and speak allows full communication with and precision seen in professional athletes, but the dual
others. qualities of stability and flexibility still apply. How are
There is more than a century of scientific study of movements controlled such that they show these two
these orderly developmental milestones. The recurrent qualities? How did they get that way during early devel-
theme of these descriptive and experimental studies has opment? In this chapter, we consider the seemingly
been to document the emergence and improvements simple, everyday skill of reaching for objects, the devel-
in motor skills and the processes that underlie these opment of that skill in infancy, and its neural founda-
changes (for reviews of motor development, see Ber- tions. We make the following points:
thenthal and Clifton, 1999; Catell, 1939; Fentress and
1. In adults, reaching is both highly repeatable and
McLeod, 1986; Jouen and Lepecq, 1990; Kirkpatrick,
very flexible.
1899; Schmidt and Fitzpatrick, 1996; Zelazo, 1998).
Thus, the concern has been not only when and how 2. These dual qualities develop gradually during the
first year of life.
MELISSA W. CLEARFIELD AND ESTHER THELEN Department of 3. Infants' first task is to acquire a measure of stability,
Psychology, Indiana University, Bloomington, Indiana. that is, a general category of successful movements.
a particular task. Indeed, there are seemingly limitless length in front of them. The developmental question
ways of organizing the combinations of muscles and has been to examine from what ongoing structures and
joints to bring a hand to a target. What solutions are processes this new behavior arises. For many years, fol-
chosen and how are these solutions retained? On the lowing Piaget (1952) and White, Castle, and Held
other hand, redundancy provides the substrate for flex- (1964), developmentalists believed that the primary is-
ibility. Without alternative solutions, behavior would be sue was infants' abilities to match the sight and feel of
rigid and inflexible. their hands to the sight of the target. It is now doubtful
This issue raises the question of how infants resolve that infants' vision of their own hands is the sole critical
the inherent tension between control (stability) and ad- element (Clifton et al., 1994), but clearly infants must
aptation (flexibility). We examine these dual themes of learn some sort of match between the properties of
stability and flexibility by looking closely at how infants their limbs and the visual properties of a target in space
learn to reach out and grab objects. (Berthier, 1996).
Infants first reach for and grab things when they are
about 4 months old. Until then, they may bring their FROM WIGGLING TO REACHING: SELECTING A STABLE
hands to their mouths, grasp objects placed in their PATTERN We have looked at the transition to reaching
hands, and wave and wiggle their arms and hands; but in a different way, by tracking changes from the pre-
they cannot, for instance, obtain a toy placed at arm's cursor movements—wavings and wigglings—to the first
FIGURE 18.2 Illustration of the linear synergy in infants. The paths. The middle panels plot the elbow joint angle versus the
three panels show the same infant sampled in the prereach shoulder joint angle. The bottom panel is the shoulder and
period, in the first month of reaching, and in the stable last elbow torques (calculated as in Gottlieb et al., 1996). (Re-
quarter of the first year. The top panels are the 3-D hand printed with permission from Thelen, in press.)
all antigravity strength to roll, sit, and stand. Nonethe- especially critical for the head, because a steady head
less, the same muscles that are used to attain these position is necessary for a stable visual image. Moreover,
postural milestones also provide the stability needed to the increasing strength and control of torso, shoulder,
reach successfully. If the head and trunk are not held and neck muscles allows infants to activate differentially
steady when the arm is lifted, the forces generated by those muscles needed for the movement of the arm and
the arm may, in turn, move the head and trunk. This is those needed to provide stability in posture.
months of age did the infants actually perseverate, and adjust on-line). Making fine adjustments requires
reaching correctly on the A trials and incorrectly on the better trajectory control, but better control could also
B trial. lead to the development of a motor habit under the
special circumstances of the A-not-B task.
THE THEMES OF STABILITY AND FLEXIBILITY REVISITED At about 1 year of age, infants do not perseverate at
These results, together with the kinematic results of the typical delays and circumstances of the A-not-B
Diedrich and colleagues (2000), confirm a shifting de- task. As their reaching improves with continued prac-
velopmental pattern between stability and flexibility of tice, they become flexible enough to follow the new cue
reaching. The first task for infants in learning to reach and reach to the second location. Where does this in-
is to select, from the many possibilities, a pattern of creased flexibility come from? At this point, we can only
movements that moves their hands into the vicinity of speculate. One possibility is that infants become better
the target. This represents an increase of stability over able to distinguish between the two identical targets so
the more random movements of the first months. Be- that they seem less novel and confusing. Another pos-
tween 4 and 7 or 8 months, this selective pattern is re- sibility is that increased practice in reaching provides
fined as variability is decreased and accuracy increases. more available motor patterns and thus less possibility
As described by Thelen and colleagues (1996), at 7 or of getting stuck in the pattern of what infants have just
8 months infants seem to discover a quite stable pattern done. Alternatively, infants may retain a stronger mem-
of speed and smoothness. It is at this time that persev- ory of the actual cue to the location and thus overcome
eration becomes predominant in the two-target task the tendency to repeat the previous action (Thelen et
(and infants also are able to reach for small objects al., in press).
ABSTRACT Behavioral studies have revealed much about the used in our native language makes it difficult for us to
development of speech perception and early word learning segment words, phrases, and sentences in nonnative
during infancy. In recent years, emerging neuroimaging and speech (Cutler et al., 1989). Our failure to distinguish-
magneto- and electrophysiological techniques have provided
a new window through which to examine neural correlates of
ing many nonnative phonemic contrasts (Werker, 1995)
language processing. In this chapter, we integrate findings may lead us to confuse similar sounding foreign words.
from the behavioral and neurophysiological literatures to ex- The difficulties we experience in processing unfamiliar
plore the brain's initial organization for language processing or even second languages suggest that experience plays
and the role of experience in modifying or maintaining this a role in the establishment or maintenance of special-
initial organization. We focus on three content areas in which
our lab has participated. At the most fundamental level we ask
ized linguistic processing substrates.
whether speech is perceived differently from nonspeech. Decades of behavioral studies have unraveled many
Within the realm of speech perception, we focus on the per- mysteries of the development of speech and language
ception of speech segments. Finally, we explore how speech processing. In recent years, neuroimaging and electro-
perception might intersect with language acquisition by re- physiological techniques have given us exciting new
viewing recent research on the use of phonetic detail in early tools to examine language processing. As a result, we
word learning. The pattern of findings across these three con-
tent areas provides converging evidence for preferred detec- now have the opportunity to sharpen our understand-
tion of some kinds of change over others in the young infant, ing, and perhaps even begin to resolve some of the theo-
and increasing neurocognitive specialization as a function of retical controversies that have defined the field. In this
age and experience. chapter, we bring together some of the findings from
the behavioral and neural processing literatures to ex-
To most adults, linguistic processing seems unre- plore how a developmental cognitive neuroscience ap-
markable. We talk a "mile a minute" and speak "without proach can be used to address old controversies and
thinking." We process speech information effortlessly, move the field on to addressing new questions. In par-
instantaneously calibrating for differences in gender, ticular, we explore the brain's initial organization for
age, accent, and speaking rate. We readily distinguish language processing and the role experience plays in
minimal differences in phonemes, parsing the words, modifying or maintaining this initial organization. We
phrases, and sentences of our native language with the focus on three content areas to which our lab has con-
greatest of ease. These abilities are so effortless and tributed: the processing of speech versus nonspeech,
rapid that they suggest specialized mental processes and phonetic perception, and early word learning. We
dedicated neural mechanisms. Not surprisingly, re- touch on the broad array of research in each of these
search has identified areas in the brain that respond areas, but direct our discussion to the specific questions
preferentially to language (Geschwind, 1970; Lenne- that have been approached from both behavioral and
berg, 1967), whether it be spoken or signed (Neville et neuroscience perspectives.
al., 1998).
At the same time, as any world traveler can confirm,
we encounter difficulties in processing unfamiliar lan- Processing of speech and nonspeech: A window
guages. This "foreign listening syndrome" is pervasive into linguistic foundations
(Mehler, Pallier, and Christophe, 1998). A relative in-
sensitivity to rhythmic structures that are unlike those Do dedicated neuroanatomical modules compute par-
ticular kinds of perceptual information, or can infor-
JANET F. WERKER AND ATHENA VOULOUMANOS Department of
mation processing be explained by general sensory and
Psychology, University of British Columbia, Vancouver, British learning mechanisms? This controversy has found ex-
Columbia, Canada. pression in a number of arenas, but was first explored
270 LANGUAGE
duces more frontal and temporal site activation, sug- perience. However, on the basis of the current evi-
gesting that processing of speech and nonspeech dence, it is also possible that such a mechanism exists
originates within different regions of the temporal lobe. but that the 3-formant sine wave stimuli are sufficiently
In another study, Mills and Neville (1997) showed that similar to engage it early in life. The robust preference
by 6 months of age, brain potentials to normal words for our nonsense words in infants 4 and 6 months of
and words presented backward begin to diverge by age, however, together with the neurophysiological
about 150-200 ms after stimulus onset, with backward evidence just reviewed, shows that, irrespective of the
words characterized by an attenuation in response. In organization at birth, by 4 months the system is suffi-
these studies, the counterparts to speech share some of ciently well tuned to the properties of speech to re-
its properties: The complex tones used by Dehaene- spond quite selectively.
Lambertz have energy in the same area of the spectrum
as speech sounds, and the backward words used by Speech perception: Listening to the sounds
Neville and Mills maintain the acoustic complexity of of language
forward speech and preserve many of its characteristics.
Thus the evidence for differential processing suggests From very early in life, human infants can categorically
that by 4 or 6 months of age, the brain encodes speech- discriminate information at various levels in speech.
specific properties faithfully enough to differentiate The categorical processing of linguistically relevant in-
speech from other highly similar acoustic signals. formation is remarkable for two reasons: (1) It provides
Evidence for specialized processing of speech also a kernel of evidence for arguments in favor of biological
comes from studies showing that infants listen differ- preparedness for language processing; (2) It readies
ently to speech than to nonspeech stimuli. For example, infants for ultimately breaking into language. This
newborn infants are better able to discriminate equally doesn't intimate that language is innate or special for
dissimilar sounds in a well-formed syllable ("tap" vs. newborns, but it does provide evidence of evolutionarily
"pat") than in a nonspeech form ("tsp" vs. "pst") (Ber- selected capabilities that can be exploited later for
toncini and Mehler, 1981). Infants as young as 4 months specific functional purposes (see Gottlieb, 1991). Our
of age show a preference for listening to speech over perceptual systems are set up to notice change, not
other sounds such as white noise (Colombo and Bundy, constancies, in the input. But if the system responded
1981) or music (Glenn, Cunningham, and Joyce, 1981), equally to any and all change that occurred, it would
be virtually impossible to leave the starting gate into
suggesting a potential initial processing bias that could
acquiring language. Fortunately, it appears that our per-
facilitate attention to specific properties of speech. This
ceptual systems are designed to maximally detect just
idea can be addressed more directly by comparing in-
those changes that are critical to the infant for the task
fants' behavioral responses to speech and nonspeech
at hand. Thus, one would predict age-related differ-
stimuli that are matched both in spectral complexity
ences in just which changes a listener detects.
and timing. And this is consistent with the pattern of findings.
To explore this idea more carefully, we tested infants Newborns have been shown to distinguish their own
on their preference for one-syllable infant-directed languages from other rhythmically distinct languages
nonsense words ("lif" and "neem") over their 3-formant (Mehler et al., 1988; Mehler and Christophe, 1994;
sine wave counterparts. (The sine wave stimuli were cre- Nazzi, Bertoncini, and Mehler, 1998), prefer their
ated by extracting the fundamental and the first three mother's voice to those of other women (DeCasper and
formants of the speech tokens, then replacing the cen- Fifer, 1980), differentiate one syllable type (Eimas et al.,
ter frequencies of each with simple sinusoidal waves.) 1971) or number (Bijeljac-Babic, Bertoncini, and Meh-
By 4 months of age, infants show a robust preference ler, 1993) from another, and differentiate grammatical
for speech over these highly matched nonspeech coun- words from lexical words (Shi, Werker, and Morgan,
terparts (Vouloumanos et al., 1999). However, in our 1999).
ongoing work there is, as yet, no evidence of a prefer- For an older infant who is facing the challenge of
ence for speech over the matched nonspeech stimuli in constructing a phonological system in her or his native
newborn infants (Vouloumanos, Werker, Bhatnager, language, native-language distinctions are most impor-
and Burns, in preparation). tant (Werker and Tees, 1984a). Accordingly, the broad-
The similarity of our nonspeech stimuli to speech based sensitivities to speech shown by the newborn
sounds makes it difficult to conclude whether or not infant prior to significant listening experience are rap-
there is a specialized speech processor in the newborn. idly molded by the input to which the infant is exposed,
Such a specialization might emerge only through ex- unleashing a barrage of abilities. Within the first days
272 LANGUAGE
ing. A shortened interstimulus interval (Werker and (1997) examined the MMN (mismatched negativity) re-
Logan, 1985) or repeated training (Tees and Werker, sponse to a continuum of vowels sounds spanning the
1984; Werker and Tees, 1984b) allows English listeners /e/-/6/-/o/ phoneme categories in Finnish and the
to discriminate the Hindi /Da/-/da/ distinction. Sim- /e/-/6/-/6/-/o/ categories in Estonian. Although
ilarly, with training, Japanese listeners improve on MMN amplitude typically increases with increased
their ability to discriminate the English /r/-/l/ acoustic difference between stimuli, Naatanen and col-
(McClelland, personal communication; Pisoni et al., leagues found that for these vowels, the size of the MMN
1982; Strange and Dittman, 1984). Another condition response varied more as a function of phonemic cate-
under which nonnative discrimination is possible is gory difference than absolute acoustic difference. The
when the nonnative distinctions are not heard as magneto- and electrophysiological studies are consis-
"speech-like." For example, English speakers can be tent with the behavioral studies, and point to the exis-
tricked into discriminating two truncated Thompson tence of specialized processing systems for the native
syllables when told they are drops of water (Werker and language. Moreover, neurophysiological studies, like
Tees, 1984b). Moreover, like 10-12-month-old infants, the behavioral studies, can also reveal the impact of
English adults can discriminate acoustically similar Zulu training on nonnative discrimination (Tremblay et al.,
clicks (Bestetal., 1989). 1997).
We have alluded to Best's hypothesis (1995) account- One group of neurophysiological studies has in-
ing for the continued ability to make these discrimina- vestigated adult processing of native and nonnative
tions: Discrimination is maintained because Zulu clicks, phonetic contrasts using the same stimuli set, first
or truncated Thompson syllables, do not engage a na- developed for testing infant speech perception in the
tive language phonological processing system. This hy- Werker and Lalonde (1988) study. These studies com-
pothesis was recently confirmed in two types of studies. pared adult brain potentials to three place of articula-
In a dichotic listening task, both English and Zulu/ tion contrasts: (1) a native bilabial/alveolar /ba/-/da/
Xhosa speakers were able to discriminate the Zulu clicks contrast that spans different phonetic categories (native
(as reported by Best, McRoberts, and Sithole, 1988), but AC), (2) a nonnative Hindi retroflex/dental /Da/-
only the Zulu/Xhosa speakers showed a right ear ad- /da/ contrast also spanning different phonetic cate-
vantage (Best and Avery, 1999). And in an fMRI study, gories (nonnative AC), and (3) an acoustic contrast
speakers of click languages, such as Zulu and Xhosa, within the same phonetic category /daj/-/da 2 / (WC).
showed greater activation in the LH than the RH in Using these three contrasts, Dehaene-Lambertz (1997)
response to the clicks, whereas English listeners showed conducted an ERP study with French-speaking adults
bilateral activation (Best et al., 1999). These findings and found significant early (within 200 ms) negative
provide striking evidence in support of organized neu- responses to the native AC change, primarily in the tem-
ral systems for processing the native language. poral lobe of the LH. There was no differential evoked
Together, infant and adult studies support the idea response to the nonnative AC change, and only a small,
that experience helps hone specialized neurocognitive bilateral late positive response to the WC change. More
systems for processing speech. A number of neurophys- recently, Rivera-Gaxiola and colleagues (2000) tested
iological studies examine this more directly by assessing (British) English speakers on the same three contrasts
the neural systems involved in native and nonnative and obtained slightly different results. The ERPs again
speech discrimination in adults. Sharma and Dorman revealed a pronounced response to the native AC
(1999) conducted an ERP study of a Hindi voicing con- change and some sensitivity to the WC change. How-
tinuum which spans the Hindi /ba/-/pa/ category ever, unlike the Dehaene-Lambertz study, the nonnative
boundary but stays within the English /ba/ category. AC change also elicited a response, which differed both
Their results suggest that the neural signature reflects in latency and topography from the native AC change.
only native language phonemic processing. Similarly, in It is not clear whether differences in participants'
a magnetoencephalography (MEG) study, Marantz and linguistic background (French vs. British English) or
colleagues (1996) tested English and Japanese listeners differences in methodology can account for the incon-
on a contrast that is used in both languages (/d/-/t/) sistent results obtained in these studies.
and one that is used only in English (/r/-/l/). The As a whole, these neurophysiological studies suggest
English subjects showed a significantly enhanced MMF that when language processing is invoked, the brain re-
(mismatched field) in response to deviant trials in both sponds only to native language phonemic categories.
the /d/-/t/ and /r/-/l/ comparisons, whereas the Indeed, so far the neurophysiological and behavioral
Japanese listeners showed it only in response to the studies are equally insensitive in revealing nonnative
/d/—/t/ contrast. Finally, Naatanen and colleagues discrimination. Of the studies reviewed here, only the
274 LANGUAGE
can learn the association between two words and two spheres and shows up at frontal, temporal, and parietal
moving objects when the objects are physically dissimi- scalp electrode sites. By 20 months, it is restricted to the
lar and the words are phonetically dissimilar (e.g., "lif" LH over the temporal and parietal lobes, suggesting the
and "neem"). If familiarized to only one word-object emergence of specialized brain systems for processing
pairing (e.g., Aa), infants as young as 8 months (and words by that age. In their work, Mills and colleagues
probably younger) can discriminate a change to a new (1993) reported that 17-month-old infants who were
word (e.g., Ba) or a new object (Ab), showing that they post-vocabulary spurt (>150 words) showed the local-
can pick up the details of both the word and the object ized LH response, whereas infants with smaller vocab-
in a discrimination task (Werker et al., 1998). However, ularies (<50 words) showed the more distributed
the failure of the younger infants in the two-word/two- response, indicating that the increasingly local nature
object switch task shows that they lack the ability to pair of the differential ERP signature seen in older infants
novel words with novel objects in this testing situation. reflects greater sophistication in word knowledge. Mills
In a follow-up series of studies, Stager and Werker and colleagues interpreted these results as indicating
(1997) used the switch and discrimination procedures that the word spurt comes about as a result of the es-
to test infants aged 8-14 months on their ability to learn tablishment of specialized neural systems for word-
the association between phonetically similar words learning (see also Schafer and Mills, 1997).
(e.g., "bih" vs. "dih") and objects. Whereas infants of 8 In collaboration with the labs of Helen Neville (Uni-
months (who cannot be tested in the switch task) easily versity of Oregon) and Debbie Mills (University of Cali-
discriminate the phonetically similar "bih" from "dih" fornia at San Diego), we used the ERP technique to
in the discrimination task, infants of 14 months fail in examine the phonetic representation of words in novice
both the switch and discrimination tasks. and experienced word learners. We tested infants of 14
We interpret these results to suggest that 8-month- and 20 months in the same ERP procedure used by
olds and 14-month-olds use different processing strate- Mills, Coffey-Corina, and Neville (1993). This time,
gies in these tasks: Inasmuch as the 8-month-olds however, in addition to presenting the infants with 10
approach the discrimination task as one of discrimina- known and 10 unknown words, we included 10 un-
tion, they are able to discriminate "bih" from "dih"; but known words that are phonetically similar to the known
14-month-olds, in treating the discrimination task as words. For example, we would present the word "dog,"
one of word-learning, fail to detect the fine phonetic which is known to infants of this age, the phonetically
difference between "bih" and "dih." To provide a strong similar unknown word "bog," and the dissimilar un-
test of this hypothesis, we tested 14-month-olds in a vir- known word "lif."
tually identical task but used a stationary unbounded The results to date with 9 subjects at each age are
nonobject visual stimulus (see Spelke, 1990) that was unequivocal. Infants of both 14 and 20 months show
unlikely to encourage word learning. When the word precisely the same pattern for the known (e.g., "dog")
"bih" (or "dih") was presented together with a station- and unknown (e.g., "lif") words as reported previously
ary checkerboard that filled the screen, following fa- by Mills and colleagues. However, the response to the
miliarization, infants of 14 months easily detected a phonetically similar unknown words (e.g., "bog") was
change to the contrasting word. These results confirm quite different for the two ages. For the 14-month-olds,
that infants of 14 months are able to detect the phonetic the ERP responses to the phonetically similar unknown
difference between "bih" and "dih." The difficulty words looked just like those to the known words. In
comes when they attempt to attach meaning to the other words, the infants of 14 months confused the pho-
word. When first engaging in a computationally more netically similar words, and treated a word like "bog" as
complex task such as word-learning, infants may have one that they knew. For the infants of 20 months, the
to temporarily drop detail—either in the encoding and ERP responses to the phonetically similar words pat-
representation phase or in the access and use phase. terned just like those to the unknown words. By 20
The potential contributions of cognitive neurosci- months, these now more proficient word-learning in-
ence techniques to understanding behavior are well il- fants were no longer confusing the phonetically similar
lustrated in subsequent studies designed to explain the unknown words with known words (Prat et al., 1999).
"failure" of 14-month-olds and to predict when infants These findings were useful for two reasons. First, they
would again have access to fine phonetic detail in word- provided confirmation that infants of 14 months do in-
learning. In previous work, Mills, Coffey-Corina, and deed have difficulty distinguishing similar sounding
Neville (1993, 1997) observed that infants show differ- words. On the basis of our behavioral work using the
ent ERP responses to known and unknown words. At switch design, we had to draw this conclusion by com-
13-17 months, this difference is detected in both hemi- paring a series of positive findings to the failures of the
276 LANGUAGE
It is in the use of phonetic detail in word learning English. In Advances in Infancy Research, Vol. 9, C. Rovee-
and lexical representation that we see the least evidence Collier and L. P. Lipsitt, eds. Greenwich, Conn.: Ablex,
pp. 217-230.
for initial specialization and the clearest evidence of ex- BEST, C. T., and R. A. AVERY, 1999. Left-hemisphere advantage
periential influences on the establishment of a distinct for click consonants is determined by linguistic significance
neural system. Here, the behavioral and electrophysio- and experience. Psychol. Sci. 10(1):65-70.
logical work both point to the emergence, at around BEST, C. T., H. HOFFMAN, and B. B. GLANVILLE, 1982. Devel-
18-20 months of age, of rapid and efficient word learn- opment of infant ear asymmetries for speech and music.
Percept. Psychophysics 31:75-85.
ing. Accompanying the achievement of this language BEST, C. T., G. D. MCROBERTS, R. LAFLEUR, and J. SILVER-
explosion milestone appears to be a specialized neural ISENSTADT, 1995. Divergent developmental patterns for in-
system for word learning (Mills, Coffey-Corina, and fants perception of two nonnative consonant contrasts.
Neville, 1993, in press; Schafer and Mills, 1997), and Infant Behav. Dev. 18:339-350.
with the establishment of this system, attention to fine BEST, C. T., G. W. MCROBERTS, and N. M. SITHOLE, 1988. Ex-
amination of perceptual reorganization for nonnative
phonetic detail is again possible (Prat et al., 1999; Wer- speech contrasts: Zulu click discrimination by English-
ker, Corcoran, and Stager, 1999; Werker et al., under speaking adults and infants.J. Exp. Psychol.: Hum. Percept.
review). The consistency in the brain areas involved Perform. 14:345-360.
from individual to individual suggests that, even in word BEST, C. T., E. MENCL, K. PUGH, T. CONSTABLE, R. FULBRIGHT,
learning, some areas in the brain are more suited for, J. GORE, C. LACADIE, P. SKUDLARSKI, A. FABER, and D. HAR-
RISON, 1999. Native-language phonetic and phonological con-
and more likely to take on this task, than others (Neville
straints on perception of non-native speech contrast. Paper
etal., 1998). presented at the March Meeting of Acoustical Society of
In summary, we have focused on infants' sensitivity to America, Berlin, Germany.
change as a critical first entry into the construction of BEST, C. T., M. STUDDERT-KENNEDY, S. MANUEL, andj. RUBIN-
a functional language processing system. At birth, the SPITZ, 1989. Discovering phonetic coherence in acoustic
system appears to be set up to notice some rather than patterns. Percept. Psychophysics 45(3):237-250.
BIJELJAC-BABIC, R.,J. BERTONCINI, and J. MEHLER, 1993. How
other changes in the environment, which, with experi- do 4-day-old infants categorize multisyllabic utterances?
ence, become more specific and attuned to, for exam- Dev. Psychol. 29(4):7ll-721.
ple, speech over nonspeech, and the characteristics of CHEOUR, M., K. ALHO, R. CEPONIENE, K. REINIKAINEN, K. SAI-
the native language. This attunement may help main- NIO, M. POHJAVUORI, O. AALTONEN, and R. NAATANEN,
tain and extend functionally relevant sensitivities while 1998. Maturation of mismatch negativity in infants. Int. J.
Psychophysiol. 29:217-226.
others regress. It also may provide a perceptual foun- CHEOUR, M., K. ALHO, K. SAINIO, K. REINIKAINEN, M. REN-
dation, which, together with the emergence of special- LUND, O. AALTONEN, O. EEROLA, and R. NAATANEN, 1997.
ized word-learning systems, allows the child to rapidly The mismatch negativity to changes in speech sounds at the
learn words in the native language. age of three months. Dev. Neuropsychol. 13(2): 167-174.
CHEOUR-LUHTANEN, M., K. ALHO, K. SAINIO, T. RINNE, K. REI-
NIKAINEN, M. POHJAVUORI, M. RENLUND, O. AALTONEN, O.
ACKNOWLEDGMENTS We thank the Natural Sciences and EEROLA, and R. NAATANEN, 1996. The ontogenetically ear-
Engineering Research Council of Canada (NSERC) for re- liest discriminative response of the human brain. Psycho-
search grant support toj. Werker (1103-95), and an NSERC physiology 33:478-481.
graduate fellowship to A. Vouloumanos. Special thanks to M. COLOMBO, J., and R. S. BUNDY, 1981. A method for the mea-
Fung and E. Job for their assistance throughout. surement of infant auditory selectivity. Infant Behav. Dev.
4:219-223.
CUTLER, A., J. MEHLER, D. NORRIS, and J. SEGUI, 1989. Limits
on bilingualism. Nature 340(6230) :229-230.
REFERENCES DECASPER, A. J., and W. P. FIFER, 1980. Of human bonding:
ASLIN, R. N., D. R. PISONI, and B. L. HENNESSY, 1981. Discrim- Newborns prefer their mother's voices. Science 208:1174-
ination of voice onset time by human infants: New findings 1176.
and implications for the effects of early experience. Child DEHAENE-LAMBERTZ, G., 1997. Electrophysiological correlates
Dev. 52:1135-1145. of categorical phoneme perception in adults. NeuroReport
BATES, E., D. THAL, D. TRAUNER,J. FENSON, D. ARAM,J. EISELE, 8(4):919-924.
and R. MASS, 1997. From first words to grammar in children DEHAENE-LAMBERTZ, G., 2000. Cerebral specialization for
with focal brain injury. Developmental Neuropsychology 13:275- speech and non-speech stimuli in infants. J. Cogn. Neurosci.
343. 12(3):449-460.
BERTONCINI, J., andj. MEHLER, 1981. Syllables as units in in- DEHAENE-LAMBERTZ, G., and S. BAILLET, 1998. A phonologi-
fant speech perception. Infant Behav. Dev. 4:247-260. cal representation in the infant brain. NeuroReport 9(8):
BERTONCINI, J., J. MORAIS, R. BIJELJAC-BABIC, S. MCADAMS, 1885-1888.
I. PERETZ, and J. MEHLER, 1989. Dichotic perception and DEHAENE-LAMBERTZ, G., and S. DEHAENE, 1994. Speed and
laterality in neonates. Brain Lang. 37:591-605. cerebral correlates of syllable discrimination in infants. Na-
BEST, C. T., 1995. Learning to perceive the sound pattern of ture 370:292-295.
278 LANGUAGE
MOON, C., R. COOPER, and W. FIFER, 1993. Two-day-olds prefer organization be data driven? Stage-like behavior in a simple
their native language. Infant Behav. Dev. 16:495-500. associative network. Infancy.
MYERS, J., P. W. JUSCZYK, D. G. KEMLER-NELSON, J. CHARLES- SCHAFER, G., and D. MILLS, 1997. Early Word Learning: From
LUCE, A. WOODWARD, and K. HIRSH-PASEK, 1996. Infants' Cerebral Specialisation to Underlying Mechanism. Paper pre-
sensitivity to word boundaries in fluent speech.J. Child Lang. sented at the Society for Research in Child Development,
23:1-30. Washington, D.C.
NAATANEN, R., A. LEHTOKOSKI, M. LENNES, M. CHEOUR, M. SCHAFER, G., and K. PLUNKETT, 1998. Rapid word learning by
HUOTILAINEN, A. I IVONKN, M. VAINIO, P. ALKU, R. J. ILMON- fifteen-month-olds under tightly controlled conditions.
IEMI, A. LUUK, J. ALLIK, J. SINKKONEN, and K. ALHO, 1997. ChildDevel. 69(2):309-320.
Language-specific phoneme representations revealed by SHAFER, V. L., D. W. SHUCARD, J. L. SHUCARD, and L. GERKEN,
electric and magnetic brain responses. Nature 385A32.-434. 1998. An electrophysiological study of infants sensitivity to
NAZZI, T., J. BERTONCINI, and J. MEHLER, 1998. Language dis- the sound patterns of English speech. J Speech Lang. Hear.
crimination by newborns: Towards an understanding of Res. 41:874-886.
the role of rhythm. J. Exp. Psychol.: Hum. Percept. Perform. SHARMA, A., and M. DORMAN, 1999. Neurophysiologic corre-
24(3):756-766. lates of cross-language perception of phonetic categories.J.
NELSON, C. A., 1999. Neural plasticity and human develop- Acoust. Soc. Am. 105 (2): 1096.
ment. Curr. Direct. Psychol. Sci. 8(2):42-45. SHI, R., J. F. WERKER, andj. L. MORGAN, 1999. Newborn in-
NEVILLE, H. J., 1995. Developmental specificity in neurocog- fants sensitivity to perceptual cues to lexical and grammat-
nitive development in humans. In The Cognitive Neurosci- ical words. Cognition. 72(2):B11-B21.
ences, M. S. Gazzaniga, ed. Cambridge, Mass.: MIT Press, SPELKE, E., 1990. Principles of object perception. Cogn. Sci.
pp. 219-231. 14(l):29-56.
NEVILLE, H. J., D. BAVELIER, D. CORINA, J. RAUSCHECKER, STAGER, C. L., and J . F. WERKER, 1997. Infants listen for more
A. KARNI, A. LALWANI, A. BRAUN, V. CLARK, P. JEZZARD, and phonetic detail in speech perception than in word-learning
R. TURNER, 1998. Cerebral organization for language in tasks. Nature 388:381 -382.
deaf and hearing subjects: Biological constraints and effects STRANGE, W., and S. DITTMANN, 1984. Effects of discrimina-
of experience. Proc. Natl. Acad. Sci. 95:922-929. tion training on the perception of /r-1/ by Japanese adults
NOVAK, G. P., D. KURTZBERG, J. A. KREUZER, and H. G. learning English. Percept. Psychophysics 36:131-145.
VAUGHANJR., 1989. Cortical responses to speech sounds and STRANGE, W., and J. J.JENKINS, 1978. Role of linguistic expe-
their formants in normal infants: Maturational sequence rience in the perception of speech. In Perception and Expe-
and spatiotemporal analysis. Electroencephalogr. Clin. Neuro- rience, R. D. Walk and H. J. Pick, eds. New York: Plenum
physiol. 73:295-305. Press, pp. 125-169.
PEGG, J. E., and J. F. WERKER, 1997. Adult and infant percep- STREETER, L. A., 1976. Kikuyu labial and apical stop discrimi-
tion of two English phonemes.J. Acoust. Soc. Am. 102(6): nation.J. Phonetics 4:43-49.
3742-3753. STUDDERT-KENNEDY, M., and D. SHANKWEILER, 1970. Hemi-
PHILLIPS, C., T. PELLATHY, and A. MARANTZ, 1999. Magnetic spheric specialization for speech perception. J Acoust. Soc.
mismatch field elicited by phonologicalfeature contrast. Paper pre- Am. 48:599-602.
sented at the 6th Annual Cognitive Neuroscience Society SWINGLEY, D., and R. N. ASLIN, 2000. Spoken word recogni-
Meeting, Washington, D.C. tion and lexical representation in very young children. Cog-
PIAGET, J., 1980. Language and cognition. In Language and nition 76:147-166.
Learning: The Debate between Jean Piaget and Noam Chomsky, TEES, R. C., and J. F. WERKER, 1984. Perceptual flexibility-
M. Piattelli-Palmarini, ed. Cambridge, Mass.: Harvard Uni- Maintenance or recovery of the ability to discriminate non-
versity Press. native speech sounds. Can. J. Psychol. 38(4):579-590.
PISONI, D. B., R. N. ASLIN, A. J. PEREY, and B. L. HENNESSY, TREHUB, S. E., 1976. The discrimination of foreign speech
1982. Some effects of laboratory training on identification contrasts by infants and adults. Child Dev. 47:466-472.
and discrimination of voicing contrast in stop consonants. TREMBLAY, K., N. KRAUS, T. D. CARRELL, and T. McGEE, 1997.
/. Exp. Psychol: Hum. Percept. Perform. 8(2):297-314. Central auditory system plasticity: Generalization to novel
POLKA, L., andj. F. WERKER, 1994. Developmental changes in stimuli following listening training.J. Acous. Soc. Am. 102(6):
perception of non-native vowel contrasts. J Exp. Psychol.: 3762-3773.
Hum. Percept. Perform. 20:421-435. VOULOUMANOS, A., M. D. BHATNAGAR, D. E. GIASCHI, and
PRAT, C. S., C. L. STAGER, T. V. MITCHELL, A. M. ADAMSON, J. F. WERKER, 1999. Do infants prefer speech over spectrally similar
and L. D. SANDERS, 1999. Semantic processing of phonetically non-speech sounds?Paper presented at the 29th annual meet-
similar words in infants: Indications from event-related potentials. ing of the Society for Neuroscience, Miami, Florida.
Poster presented at the Society for Research in Child De- WADA, J. A., R. CLARKE, and A. HAMM, 1975. Cerebral hemi-
velopment, Albuquerque, New Mexico. spheric asymmetry in humans: Cortical speech zones in 100
REPP, B. H., 1984. Categorical perception: Issues methods adult and 100 infant brains. Arch. Neurol. 32(4):239-246.
findings. In Speech and Language: Advances in Basic Rresearch WERKER, J. E, 1995. Exploring developmental changes in
and Practice, Vol. 10, N. J. Lass, ed. New York: Academic cross-language speech perception. In An Invitation to Cog-
Press, pp. 243-335. nitive Science, 2nd ed. Language: Vol. 1, D. N. Osherson, L. R.
RIVERA-GAXIOLA, M., G. CSIBRA, M. H.JOHNSON, and A. KAR- Gleitman, and M. Liberman, eds. Cambridge, Mass.: MIT
MILOFF-SMITH, 2000. Electrophysiological correlates of Press, pp. 87-106.
cross-linguistic speech perception in native English speak- WERKER, J. E, L. B. COHEN, V. L. LLOYD, M. CASASOLA, and
ers. Behav. Brain Res. lll(l-2):13-23. C. L. STAGER, 1998. Acquisition of word-object associations
SCHAFER, G., and D. MARESCHAL, in press. Can functional re- by 14-month-old infants. Dev. Psychol. 34(6):1289-1309.
280 LANGUAGE
20 Language Development in Children
with Unilateral Brain Injury
ELIZABETH BATES
AND KATHERINE ROE
ABSTRACT Aphasia (defined as the loss or impairment of than injuries to the right, suggesting in turn that
language abilities following acquired brain injury) is strongly (2) the left hemisphere plays a privileged role in lan-
associated with damage to the left hemisphere in adults. This guage processing by normal adults. The second conclu-
well-known finding has led to the hypothesis that the left hemi-
sphere is innately specialized for language, and may be the sion has been independently confirmed in the 20th
site of a specific "language organ." However, we have known century by methods ranging from sodium amytal
for more than a century that young children with left- (WADA) tests and/or point-to-point electrical stimula-
hemisphere damage (LHD) do not suffer from aphasia, and tion in adult candidates for neurosurgery (Ojemann,
in most studies do not differ significantly from children with 1991) to neural imaging studies of normals, including
right-hemisphere damage (RHD). This result provides strong
evidence for plasticity—brain reorganization in response to
positron emission tomography (PET), functional mag-
experience—and constitutes a serious challenge to the lan- netic resonance imaging (fMRI), magnetoencephalog-
guage organ hypothesis. This chapter reviews the history of raphy (MEG), and event-related brain potentials (ERP).
research on language outcomes in children versus adults with (For reviews, see Brown and Hagoort, 1999; Xiong et
unilateral brain injury. In that review we address some of the al., 1998.)
discrepancies in the literature to date, including the meth-
odological confounds that may be responsible for those dis- The privileged status of the left hemisphere for lan-
crepancies. We also review recent prospective studies of guage processing is now beyond dispute (with estimates
children with unilateral injury as they pass through the first averaging from 95% to 98% of normal adults, indepen-
stages of language development. Prospective studies have dent of handedness), but the origins and development
demonstrated specific correlations between lesion site and of this specialization are still poorly understood. There
profiles of language delay, but they look quite different from
lesion-symptom correlations in adults, and gradually disap- must be something about the left hemisphere that
pear across the course of language development. The classic makes it especially suited for language—but what is that
pattern of brain organization for language observed in normal "something"? Is it present at birth, or does it develop
adults may be the product rather than the cause of language gradually? Is it possible to develop normal language in
learning, emerging out of regional biases in information pro- the absence of a normal left hemisphere? And if an
cessing that are relevant for language, but only indirectly re-
lated to language itself. If those regions are damaged early in intact left hemisphere is not required for language de-
life, other parts of the brain can emerge to solve the language- velopment, then when, how, and why does it become nec-
learning problem. essary for language use in adults? Finally, if alternative
forms of brain organization for language can emerge
Aphasia, the loss of language abilities following brain in the presence of early left-hemisphere damage, is
injury, has been studied systematically in adults for there some critical period in which this must occur?
more than a century, and its existence has been docu- The sparse but growing body of evidence on language
mented since the first Egyptian surgical papyrus more development in children with left- versus right-hemi-
than 4000 years ago (Goodglass, 1993; O'Neill, 1980). sphere damage is relevant to all these points, and it has
There is now a large body of research on adult aphasia, yielded two very puzzling results: (1) Most children with
and although there is still substantial controversy re- early left-hemisphere damage go on to acquire lan-
garding its nature and causes, consensus has emerged guage abilities within the normal range (although per-
on at least two points: (1) Injuries to the left hemi- formance is often at the low end of the normal range),
sphere are overwhelmingly more likely to cause aphasia and (2) most studies fail to find any significant differ-
ences in language outcomes when direct comparisons
ELIZABETH BATES AND KATHERINE ROE Center for Research are made between children with left- versus right-hemi-
in Language, University of California at San Diego, La Jolla, sphere damage. These unexpected findings in children
California. are hard to reconcile with one of the most popular ideas
BATES AND ROE: LANGUAGE DEVELOPMENT IN CHILDREN WITH BRAIN INJURY 281
in neuropsychology: that the left hemisphere of the hu- fact is clear: In the absence of other confounding fac-
man brain contains an innate and highly specialized tors (e.g., severe and intractable seizures), the language
organ for language (Fodor, 1983; Gopnik, 1990; Gopnik deficits observed in children with early left-hemisphere
and Crago, 1991; Newmeyer, 1997; Pinker, 1994; Rice, injury are (if they exist at all) far less pronounced than
1996). The language-organ hypothesis is appealing on the aphasic syndromes seen in adults (Bates, 1999;
many grounds. Aside from its value in explaining left- Bates, Vicari, and Trauner, 1999; Eisele and Aram, 1995;
hemisphere specialization, the existence of a special- Vargha-Khadem, Isaacs, and Muter, 1994; Vargha-
ized language organ might help to explain why all Khadem et al., 1992). Other conclusions are still con-
normal adults are virtuosi in this domain. For example, troversial, regarding the time course of recovery, the
adult speakers of English produce an average of 150 nature of the mechanisms that support it, and whether
words per minute, each rapidly selected from a pool of there are ultimately any significant differences (i.e.,
20-40 thousand lexical options. As quickly as these mild deficits) between children with left- versus right-
words are spoken (often blurred together, without well- hemisphere damage.
marked boundaries), the average listener can parse Our ability to answer these questions is limited by a
these unbroken streams of sound into words and number of factors. First, focal lesions in young children
phrases, accessing the meaning of each word (from that are very rare, so that generalizations are sometimes
same large pool) while simultaneously processing all based on samples too small to support them. Second,
the complex grammatical cues necessary for compre- results across studies are often in direct conflict, due to
hension. This is an ability no other species on the planet methodological variations including sample size, etiol-
appears to have, and one that today's largest and fastest ogy (e.g., stroke, tumor, trauma, and conditions that
computers have yet to master. might predispose children to any of these injuries), age
Perhaps even more phenomenal than the speed and of lesion onset, age of testing, the developmental sen-
ease with which we produce and perceive speech is the sitivity (or insensitivity) of the instruments used to eval-
speed and ease with which we learn how to do it. Most uate language, and the kinds of statistical comparisons
4-year-olds cannot tie their own shoes, but they can eas- that were made (e.g., whether children with LHD and
ily ask someone else to help them. In fact, most 4-year- RHD are compared directly, versus indirect compari-
olds have a vocabulary of 6000 words or more, and sons in which each clinical group is evaluated against a
produce well-formed sentences as grammatically com- separate set of normal controls).
plex as those observed in any adult (Bates, in press; Due in part to these troubling methodological fac-
Bates et al., in press; Fletcher and MacWhinney, 1995). tors, research on language outcomes following early
Children master their native language (or languages, for brain injury has swung back and forth between two ex-
that matter) without formal instruction, without ex- treme views: equipotentiality (site or side of injury does
plicit corrections, and seemingly, without effort. Per- not matter at all in young children, because both sides
haps we are the only animals on earth that can manage of the brain are equivalent at birth) and irreversible de-
this feat because we have an innate language organ. But terminism (the left hemisphere is innately and irrevers-
the organ metaphor carries a number of assumptions ibly specialized for language, precluding the possibility
that are contradicted by research on language devel- of complete and normal language development if it is
opment in children with early brain injury: (1) The severely damaged). We argue that the bulk of the evi-
brain in general and the left hemisphere in particular dence supports a compromise view between these two
are specialized for language at birth; (2) this speciali- extremes, in which the two hemispheres are character-
zation involves compact and well-defined regions of the ized at birth by innate but "soft" biases in information
left hemisphere that are dedicated to language (and processing that are relevant to language but not specific
language alone); (3) this specialization is irreversible, to language, permitting both neural and behavioral re-
so that normal levels of language are precluded if the organization across the course of language develop-
language organ is severely damaged at birth; (4) even ment (see also Satz, Strauss, and Whitaker, 1990). On
if some degree of language learning does take place this argument (which we will call the emergentist view), we
(presumably through compensatory mechanisms), chil- would expect to see left/right-hemisphere differences
dren with early left-hemisphere injuries should display early in life, but these differences will decrease with
persistent deficits that are not observed with early in- time and may eventually disappear.
juries to homologous areas on the right side of the We will review the evidence in three partially overlap-
brain. ping phases in the history of this field: an equipoten-
All of these assumptions are in peril. Although these tiality phase, an irreversible determinism phase, and
issues are not yet settled to everyone's satisfaction, one (after a brief stop to consider the contribution of meth-
282 LANGUAGE
odological factors) the current move toward an erner- all its claims regarding the autonomy, innateness and
gentist view. A summary of evidence involving measures "unlearnability" of language (Botha, 1989; Newmeyer,
of verbal and nonverbal IQ is presented in table 20.1. 1980). Another round of studies of children with early
Evidence based on more specific measures of language brain injury rapidly ensued, leading to an entirely dif-
is summarized in table 20.2. ferent view.
Not long after the first 19th-century studies linking In response to Lashley and Lenneberg's controversial
aphasia to left-hemisphere damage in adults, studies ap- ideas about equipotentiality, a number of studies ap-
peared suggesting that children with the same kinds of peared between 1960 and 1980 suggesting that early
damage have little or no difficulty with language (Co- brain injury does lead to subtle but persistent language
tard, 1868; Clarus, 1874; both cited in Woods and Teu- impairments, deficits that are more likely following left-
ber, 1978), or that they show temporary deficits that hemisphere damage (LHD) than right-hemisphere
quickly disappear (Bernhardt, 1897). damage (RHD). For example, Woods and colleagues
In this century, Basser (1962) reported on 34 chil- (Woods, 1980; Woods and Carey, 1979; Woods and Teu-
dren with severe epilepsy who underwent a radical pro- ber 1973, 1978) concluded that LHD in children does
cess called hemispherectomy (removal of the damaged lead to speech and language problems, especially if le-
side of the brain) to control intractable seizures. Results sion onset occurs after one year of age, and they attrib-
were consistent with those from the century before: all uted earlier evidence for equipotentiality to limitations
but one of these children developed speech abilities in in medical knowledge at that time (Woods and Teuber,
the normal range (see also Rasmussen and Milner, 1978). In the same vein, Dennis and colleagues (Den-
1977). It was Basser's study that led Lenneberg (1967) nis, 1980; Dennis and Kohn, 1975; Dennis and Whi-
to his controversial notion that the brain is "equipoten- taker, 1976, 1977; Dennis, Lovett, and Wiegel-Crump,
tial" at birth, with lateralization determined gradually 1981) reported that left-hemispherectomized children
across the course of development. As a corollary, Len- are more likely to have phonological and grammatical
neberg also argued that this period of equipotentiality problems than children with right hemispherectomies
and plasticity is brought to an end at puberty, providing (although the reported deficits were quite subtle).
the first systematic argument in favor of a "critical pe- Although these studies were influential (and are cited
riod" for language. Lenneberg's views were quite com- in many textbooks), most of them do not include direct
patible with an earlier proposal by Lashley (1950), who statistical comparisons of children with LHD and chil-
interpreted lesion studies of animals to indicate that dren with RHD (see tables 20.1 and 20.2). Some looked
loss of learning is predicted by the size of the lesion exclusively at LHD children and controls, while others
rather than its location (see also Irle, 1990). Lenne- compared each group to its own set of age-matched
berg's critical period proposal was compatible not only controls (a practice followed in many of the studies re-
with the evidence on recovery from unilateral damage viewed below). The latter practice is common, but it is
(i.e., the difference between children and adults with also problematic: Authors infer that effects of LHD are
comparable injuries), but also with (1) the difficulty quantitatively and perhaps qualitatively different than
that adults display in acquiring a second language with- the effects of RHD, but this supposed difference in pat-
out an accent and (2) some influential "wild child" stud- terning assumes an untested statistical interaction (i.e.,
ies, especially the famous case of Genie (Curtiss, 1977), that the difference between LHD and their controls is
which seemed to suggest that acquisition of a first lan- statistically greater than the difference between RHD
guage is also precluded if normal input is delayed until and their controls). As we shall see, studies that have
late childhood or puberty. looked for such statistical interactions (or compared
However, Lenneberg's equipotentiality hypothesis LHD and RHD directly) have generally failed to find
did not sit well with some of his contemporaries, who the predicted effects.
were persuaded by the research of Sperry, Gazzaniga, As evidence accumulated, the picture became more
Geschwind, and others that the two hemispheres are complex, and more confusing. For example, Alajouanine
too different to support a complete change of roles and Lhermitte (1965) reported that children with LHD
even early in life (Gazzaniga and Sperry, 1967; Gesch- do have initial difficulty with some aspects of language,
wind and Kaplan, 1962; Levy, Nebes, and Sperry, 1971). especially expressive language, but these difficulties
Equipotentiality was also difficult to reconcile with were far less pronounced than those seen in adults
Noam Chomsky's theory of generative grammar, with and disappeared within 6 months to 2 years after
(text continues on page 296)
BATES AND ROE: LANGUAGE DEVELOPMENT IN CHILDREN WITH BRAIN INJURY 283
TABLE 20.1
Summary of IQ data from studies of children with focal brain injury
Patients
RHD
Neuro- Age at vs.
Authors Site imaging Age of Onset Testing Seizures LHD Results
Woods and 19 RHD No PN- 15 years NR NR Yes FIQ:
Teuber (1973) 17 LHD LHD < controls
RHD < controls
LHD = RHD
PIQ:
LHD < controls
RHD < controls
VIQ:
LHD < controls
RHD = controls
Woods (1980) 23 RHD No Early (<lyr): RE: XT = 14; 10 RE: 5(-S); 5( + S) No VIQ:
27 LHD 10 RHD = RE LE: XT = 17; 2 LE: 8(-S); 3( + S) RE < population mean & controls
11 LHD - LE RL: Xr = 10; 3 RL: ll(-S); 2( + S) LE < population mean & controls
LL:X T = 8;6 LL: 13(-S); 3( + S) RL = population mean & controls
Late (>1 yr):
LL < population mean & controls
13 RHD = LR Comparisons
XQ = 6; 4 between + S and — S PIQ:
6 LHD = LL not made RE < population mean & controls
Xo = 5; 7 LE < population mean & controls
RL < population mean & controls
LL < population mean & controls
Vargha-Khadem, 28 RHD Some 33 Prenatal (<2 mos.) NSR NR No Prenatal RHD: VIQ = PIQ
O' Gorman, and 25 LHD (16 RHD, 18 LHD) Prenatal LHD: VIQ < PIQ
At least >2 yrs.
Walters (1985)
1 1 Early postnatal post-onset Early postnatal RHD: VIQ = PIQ
(2 mos. to 5 yrs.) Early postnatal LHD: VIQ < PIQ
(5 RHD, 6 LHD)
Late postnatal RHD: VIQ = PIQ
9 Late postnatal Late postnatal LHD: VIQ = PIQ
(5-14 yrs.)
(5 RHD, 4 LHD)
Aram et al. 8 RHD Yes RHD: RHD: None No FIQ:
(1985) 8 LHD Xo = 1; 4 XT = 3; 10 LHD < controls
(2 mos. to 9; 0 yrs.) (1; 8-5; 10) RHD < controls
LHD: LHD: P1Q:
Xo = 1;6 XT = 4; 6 LHD < controls
(1 mo. to 6; 2 yrs.) (2 mos. to 8; 2 yrs.) RHD < controls
VIQ:
LHD = controls
RHD < controls
Riva and 26 RHD Some Early (<1 yr): Early (<1 yr): None No FIQ:
Cazzaniga 22 LHD 13 RHD = RE XT = 8; 5 LE — controls
(1986) 10 LHD = LE LL = controls
Late (>1 yr):
RE < controls
Late (>1 yr): XT = 1 1 ; 8
RL < controls
13 RHD = LR
12 LHD = LL PIQ:
Xo = 6; 6 LE < controls
LL = controls
RE < controls
RL < controls
VIQ:
LE < controls
LL = controls
RE = controls
RL = controls
Riva et al. 8 RHD Yes RHD: RHD: X T = 1 0 ; 7 5RHD(-S) No PIQ:
(1986) 8 LHD Xo = 4; 6 3RHD( + S) LHD < controls
(l;8-8; 3) LHD: XT = 11; 1 RHD < controls
LHD: 5LHD(-S) VIQ:
Xo = 6; 4 3LHD( + S) LHD < controls
(0; 8-12; 8) Comparisons RHD = controls
between + S and — S
not made
Nass, Peterson, 12 RHD Yes Pre- or perinatal RHD: XT = 8; 6 Yes FIQ: LHD > RHD
and Koch (1989) 14 LHD PIQ: LHD = RHD
LHD: XT = 6; 8 VIQ: LHD > RHD
LHD: PIQ< VTQ
RHD: PIQ = VIQ
(continued)
TABLE 20.1 (Continued)
Summary of IQ data from studies of children with focal brain injury
Patients
RHD
Neuro- Age at vs.
Authors Site imaging Age of Onset Testing Seizures LHD Results
Vargha-Khadem 40RHD Yes Pre- or perinatal RHD(-S): 24RHD(-S) Yes FIQ: RHD = LHD
et al. (1992) 42LHD XT=11;3 16RHD( + S) + S < controls
— S < controls
RHD( + S): 28RHD(-S) + S< -S
XT = 11; 6 14RHD( + S)
LHD(-S): PIQ: RHD = LHD
XT = 12; 9 + S < controls
— S < controls
+ S < -S
LHD( + S): VIQ: RHD = LHD
XT = 11; 8 + S < controls
— S controls
+ S < -S
Eisele and Aram 12 RHD Yes RHD: RHD: None Yes VIQ: LHD = LHD
(1993) 21 LHD Xo = 1; 06 XT = 5; 8
(PN-9; 08) (2; 06-14; 7) PIQ:
Patients
RHD
Neuro- Age at vs.
Authors Site imaging Age of Onset Testing Seizures LHD Tests Used Results
McFie (1961) 24RHD No 35 infantile (before NR NR NR Clinical/observational 7 RH infantile with clinical
14 LHD lyr.) (19RHD, dysphasia
11 LHD) 1 RH juvenile with clinical
dysphasia
8 juvenile (after 1 yr.)
4 LH infantile with clinical
(5 RH, 3 LH); range
dysphasia
of onset 15 mos. to
4 LH juvenile with clinical
5 yrs.
dysphasia
Weigl Sorting test LHD: Impaired on Weigl
Sentence Learning test Sorting and Sentence
Memory for Designs test Learning, but not on
Memory for Designs
RHD: Impaired on Weigl
Sorting and Sentence
Learning, but not on
Memory for Designs
Basser (1962) 48 LHD No 72 prespeech onset NR NSR No Clinical/observational None of the RHD or LHD
54 RHD (34L, 38R) had dysphasia
30 postspeech (14 L, 8/48 LHD had stuttering
16 R) or articulation problems
8/48 LHD had reduced or
nonexistent speech
7/54 RHD had stuttering
or articulation problems
11/54 RHD had reduced
or nonexistent speech
34 prespeech onset LHD
had delayed first word
production
38 prespeech onset RHD
had delayed first word
production
Alajouanine and 32 LHD No 6-15yrs. NR NR N/A Clinical/observational 32/32 had impaired
Lhermitte (1965) patients expression (written,
only spoken, gestural)
7/32 exhibited paraphasias
10/32 had impaired
comprehension
18/32 had alexia
Woods and Teuber 31 RHD No LHD: NR NR No Clinical/observational 25/34 LHD had language
(1978) 34 LHD AQ = 6;3 (1;4-15; 2) disturbances post-onset
RHD:
4/31 RHD had language
XQ = 7;2 (1; 9-14; 4)
disturbances post-onset
(3 were premorbid left-
handers)
Woods and Carey 27 LHD No Early LHD (before 1 Early LHD: Early LHD: N/A Oldfield-Wingfield Early LHD = controls
(1979) patients yr.),N= 11 XT = 17; 8 yrs. 2 ( - S ) , 9 ( + S) Picture Naming Late LHD < controls
only
Late LHD (after 1 yr.), Late LHD: Late LHD: Spelling Early LHD < controls
N= 16 XT = 15; 8 yrs. 15(-S), 1( + S) Late LHD < controls
Xo = 5; 7 yrs.
Comparison between Rhymes (complete 6 Early LHD = controls
(1;2-15; 1)
+ S and — S nursery rhymes) Late LHD = controls
groups not made
Sentence Completion Early LHD = controls
(from the BDAE) Late LHD < controls
Vargha-Khadem, 28 LHD Some 33 prenatal (< 2 mos) NR NR Yes Token Test RHD > LHD
O'Gorman, and 25 RHD (16 RHD, 18 LHD) All 3 RHD groups
At least < 2 yrs.
Walters (1985) = controls
11 early postnatal post-onset
All 3 LHD groups
(2 mos. to 5 yrs.)
< controls
(5 RHD, 6 LHD)
Late postnatal LHD
9 late postnatal < prenatal and
(5-14 yrs.) late RHD
(5 RHD, 4 LHD)
(continued)
TABLE 20.2 (Continued)
Summary of data from studies of children with focal brain injury using various language assessment measures
Patients
RHD
Neuro- Age at vs.
Authors Site imaging Age of Onset Testing Seizures LHD Tests Used Results
% Error
LH < controls
RH = controls
Rapid Automatized LHD:
Naming Test: Objects < numbers £
Repeatedly name letters
visually presented Colors < number
exemplars in 4 RHD:
categories (objects, Objects < numbers &
number, letters, colors) letters
Colors < number
(continued)
TABLE 20.2 (Continued)
Summary of data from studies of children with focal brain injury using various language assessment measures
Patients
RHD
Neuro- Age at vs.
Authors Site imaging Age of Onset Testing Seizures LHD Tests Used Results
Aram, Meyers, and 13 RHD Yes RHD: RHD: None No Free speech analysis of:
Ekelman (1990) 20 LHD X0 = 2; 2 XT = 7; 0 Normal-type nonfluencies LHD = controls
(PN-5; 3) (4; 7-17; 3) (whole-word & phrase RHD = controls
repetititions, revisions/
LHD: LHD:
incomplete phrases,
Xo = 4- 8 XT = 9; 10
interjections, unfilled
(PN-15; 11) (3; 10-16; 5)
pauses, parenthetical
remarks)
Stuttering-type LHD < controls
nonfluencies (part- RHD = controls
word repetititions,
prolongations, tense
pauses)
Speech rate (syllables/ sec) LHD < controls
RHD = controls
Eisele and Aram 12 RHD Yes RHD: RHD: None Yes PPVT LHD < controls
(1993) 21 LHD X0 = 1; 06 XT = 5; 08 RHD < controls
(PN-9; 08) (2; 06-14; 07) RHD < LHD
LHD: LHD: EOWPVT No significant differences
XQ = 2; 03 XT = 7; 05 between groups
(PN-11;07) (2; 07-14; 02)
Eisele and Aram 9 RHD Yes RHD: RHD: None No Imitation of: LHD < controls
(1994) 16 LHD X0 = 2; 08 XT = 12; 11 Coordinate sentences RHD < controls
(PN-9; 08) (9; 01 -23; 01) Passive sentences LHD: coordinates <
Relative sentences passives < relatives
LHD: LHD:
RHD: passives <
Xo = 4; 05 XT = 14; 2
coordinates < relatives
(PN-15; 11) (7; 02-22; 4)
Comprehension of: LHD = controls
Coordinate sentences RHD = controls
Passive sentences LHD: coordinates <
Relative sentences passives = relatives
RHD: passives <
coordinates = relatives
Eiselc and Aram LHD: comprehension >
(1994) (com.) production
RHD: comprehension =
production
7/16 LHD < controls on
imitation only
2/16 LHD of these were
also < controls on both
imitation and
comprehension
2/9 RHD < controls on
comprehension only
1/9 RHD < controls on
imitation only
1/9 RHD < controls on
both imitation and
comprehension
Reilly, Bates, and 13 RHD Yes < 6 mos. RHD: 1( + S) Yes Narrative speech sample
Marchman (1998) 18 LHD XT = 6; 6 (4; 1-9; 1) analyzed:
Propositional length RHD = LHD
LHD:
R/LHD < controls
XT = 6; 4 (3; 7-9; 4)
Older = younger
Younger (<5 yrs.)
Total word token/types RHD = LHD
(7 LHD, 4 RHD)
R/LHD < controls
Older (>5 yrs.)
Older = younger
(11 LHD, 9 RHD)
Type/token ration RHD = LHD
R/LHD = controls
Older = younger
Use of mental verbs or RHD = LHD
terms for mental state/ R/LHD = controls
motivation Older > younger
Forms of reference RHD = L H D
(noun/pronoun ratio) R/LHD = controls
Older = younger
Use of complex, complete RHD = LHD
referring expressions R/LHD = controls
(noun and antecedent Older > younger
within a sentence)
Proportion of RHD = LHD
morphological errors RHD < controls
LHD controls
Older > younger
Younger + Temp <
— Temp
(continued)
TABLE 20.2 (Continued)
Summary of data from studies of children with focal brain injury using various language assessment measures
Patients
RHD
Neuro- Age at vs.
Authors Site imaging Age of Onset Testing Seizures LHD Tests Used Results
Reilly, Bates, and Proportion of syntactic Older > younger
Marchman (1998) complexity (coordinates, RHD > LHD
(cont.) passives, relative clauses, Younger RHD > younger
sentences with adverbial LHD
clauses and/or verb Younger RHD = controls
complements) Younger LHD < controls
Older RHD = older LHD
Older LHD, RHD <
controls
Syntactic diversity Older > younger
RHD > LHD
Younger RHD > younger
LHD
Younger RHD = controls
Younger LHD = controls
Older RHD = older LHD
Older LHD, RHD <
controls
Proportion of pragmatic RHD = LHD
connectors R/LHD = controls
Older = younger
Inclusion of major story RHD = LHD
components (8) R/LHD < controls
Older > younger
If/how often theme of RHD = LHD
narrative is mentioned R/LHD < controls
Older > younger
Bates et al. (1997) 10 RHD Yes Under 6 mos. 10-17 mos. 7RHD(-S) Yes MacArthur GDI word LHD = RHD
16 LHD 3RHD( + S) comprehension LHD = controls
RHD < controls
14LHD(-S)
R/LHD < controls
2LHD( + S)
9 RHD Yes Under 6 mos. 10-17 mos. NSR Yes MacArthur GDI gesture LHD > RHD
11 LHD R/LHD < controls
10 RHD Yes Under 6 mos. 10-17 mos. 7RHD(-S) Yes MacArthur GDI % of LHD = RHD
16 LHD 3RHD( + S) receptive vocabulary R/LHD < controls
produced + LTemp < - LTemp
14LHD(-S)
2LHD( + S)
12 RHD Yes Under 6 mos. 19-31 mos. 9RHD(-S) Yes MacArthur GDI word LHD = RHD
17LHD 3RHD( + S) production R/LHD < controls
+ LTemp < - LTemp
15LHD(-S)
+ Frontal < - Frontal
2LHD( + S)
MacArthur GDI LHD = RHD
expressive grammar R/LHD < controls
+ LTemp < - LTemp
+ Frontal< -Frontal
8RHD Yes Under 6 mos. 19-31 mos. NSR Yes MacArthur GDI % LHD < RHD
8LHD function words LHD = controls
RHD > controls
6 RHD Yes Under 6 mos. 20-44 mos. 2RHD(-S) Yes Free speech: Mean length LHD = RHD
24 LHD 4RHD( + S) of utterance in R/LHD < controls
morphemes + LTemp < - LTemp
21 LHD(-S)
3LHD( + S)
Bates, Vicari, and 12 RHD Yes Under 6 mos. 3-13yrs. None Yes Peabody Picture R/LHD < controls
Trauner (1999) 18 LHD Vocabulary Test (R/LHD = controls with
mental age controlled)
12 RHD Yes Under 6 mos. 3-13 yrs. None Yes Boston Naming Test R/LHD < controls
10 LHD (R/LHD < controls with
mental age controlled)
15 RHD Yes Under 6 mos. 3-13yrs. None Yes Semantic Fluency R/LHD < controls
19 LHD (R/LHD = controls with
mental age controlled)
8 RHD Yes Under 6 mos. 5- 13 yrs. None Yes Token Test R/LHD < controls
9 LHD (R/LHD = controls with
mental age controlled)
5 RHD Yes Under 6 mos. 4-8 yrs. None Yes Test for Receptive LHD = RHD = controls
9 LHD Grammar (TROG)
RHD, right-hemisphere damage; LHD, left-hemisphere damage; PN, prenatal; + S patients suffering seizures; — S, patients with no seizure activity; + Temp, injury includes
damage to the temporal lobe; -Temp, injury spares the temporal lobe; + Frontal, injury includes damage to the frontal lobe; -Frontal, injury spares the frontal lobe; NR,
not reported; N/A, not applicable; .Xo, mean age of onset (years; months); XT, mean age of testing (years; months).
Note: < indicates worse performance regardless of the specific measure used; = indicates that groups did not differ significantly.
lesion onset. Note that Alajouanine and Lhermitte did patients typically show the opposite pattern. The extent
not study RHD patients. Riva and colleagues (1986) to which findings for children fit this pattern varies
found that while left-hemispherectomized children per- from study to study, due in part to methodological
formed more poorly than right-hemispherectomized confounds.
children on some grammatical comprehension tests, In one study, Woods (1980) found that results for VIQ
LHD and RHD children were equally impaired on mea- and PIQ depended both on side of lesion and on the
sures—of—vocabulary production and comprehension. age at which the lesion was acquired. He found that (1)
Similar findings have been reported in a series of stud- children with LHD scored significantly below normal
ies by Aram and colleagues (Aram et al., 1985; Aram, on both VIQ and PIQ, regardless of the age at which
Ekelman, and Whitaker, 1986; Aram, Meyers, and Ek- the lesion was acquired; (2) children with RHD also
elman, 1990) and Eisele (Eisele and Aram, 1993, 1994, scored below normal on both subscales, but only if their
1995). While Aram and colleagues (1985) and Eisele lesions were acquired before one year of age; (3) if chil-
and Aram (1993) found that on measures of lexical dren with RHD acquired their lesions after the first year,
competence, RHD and LHD children were both im- they scored in the normal range for language but below
paired relative to age-matched controls, it appeared normal on performance IQ. This complex nest of find-
that children with LHD performed worse than their ings led Woods, Teuber, and colleagues to propose the
normal controls on a number of other language mea- "crowding" hypothesis: In an effort to salvage language
sures, including tests of both grammatical comprehen- in the presence of LHD, language functions are moved
sion and production, phonological discrimination tests, to the right hemisphere, where they interfere with the
and tests of lexical fluency. By contrast, children with spatial tasks normally conducted in those areas of the
RHD showed no statistical difference from their own brain.
controls on nearly all such measures. However, later Riva and colleagues (1986) also reported differential
studies by the same research team reached a different effects of age of onset and lesion side, but their results
conclusion. For example, Eisele and Aram (1994) re- were virtually the opposite of Woods (1980). Children
ported no differences between LHD and RHD on a test with early LHD were significantly lower than controls
on syntax comprehension, although several children on both VIQ and PIQ, but only if their lesions occurred
from both groups performed at chance. Based on a de- before one year of age; children with later lesions did not
tailed qualitative examination of lesion data (albeit
differ significantly from normal controls on either sub-
without a statistical test), the authors concluded that
scale. Children with RHD scored reliably below normal
subcortical involvement to either hemisphere may be
controls on PIQ, but not on VIQ, regardless of the age
the most important determiner of failure on this syntax
at which damage occurred. More recently, Ballantyne,
task (Eisele and Aram, 1995).
Scarvie, and Trauner (1994) found that brain-injured
A similar history can be traced in research by Vargha-
Khadem and colleagues. For example, Vargha-Khadem, children as a group performed below controls on all IQ
O'Gorman, and Watters (1985) reported that perfor- subscales; VIQ was no worse than PIQ for LHD chil-
mance on grammatical comprehension tests was more dren, but VIQ was better than PIQ for RHD children.
impaired in children with LHD. However, as they added Note that none of these studies (Woods, 1980; Riva et
more cases to their sample, this difference disappeared al., 1986; Ballantyne et al., 1994) reports a direct statis-
(Vargha-Khadem, Isaacs, and Muter, 1994). It now ap- tical comparison of LHD and RHD.
pears from studies by this research group that seizure Nass, Peterson, and Koch (1989) did conduct direct
history is the most important predictor of language im- comparisons of children with congenital LHD and
pairments in brain-injured children, regardless of side RHD, with surprising results: Children with LHD actu-
or size of injury or of the age at which the lesion was ally did better on VIQ than PIQ, and they also performed
acquired. better than children with RHD on the verbal scale. Ei-
Variations in the tests used to assess language (see sele and Aram (1993) also compared groups of brain-
tables 20.1 and 20.2) may be responsible for some of injured children directly. They found the adult pattern
the discrepancies seen between studies. However, even for PIQ (with RHD performing worse than both LHD
when standardized tests of IQ are used, studies differ in and controls), but there were no effects of lesion side
factors like age of onset, subcortical involvement, and on VIQ (where LHD and RHD were both indistinguish-
presence or absence of seizures. When IQ scores are able from controls). Vargha-Khadem (Vargha-Khadem
broken down into verbal and nonverbal (performance) et al., 1992) and Muter (Muter, Taylor, and Vargha-
quotients, adult LHD patients typically have higher PIQ Khadem, 1997) found no differences between RHD and
scores compared to their VIQ scores, whereas RHD LHD groups on either VIQ or PIQ, although children
296 LANGUAGE
with seizures were more impaired on both scales than have shown that infant monkeys with bilateral removal
children without seizures. of area TE (the ventral temporal areas that are the final
As we move out of the 1990s and into the next mil- way station of the "what is it?" visual system in mature
lennium, some of the confusion that has characterized animals) perform only slightly below unoperated con-
research in this area has begun to lift. Most investigators trols on a task that measures memory for new visual
now embrace a "third view" midway between equipoten- objects, whereas mature animals with the same lesions
tiality and irreversible determinism, a bidirectional re- display severe visual amnesia. The accumulated evi-
lationship between brain and behavioral development dence strongly suggests that cortical specialization is (at
in which initial biases and subsequent reorganization least in part) driven by cortical input, and that new
are both acknowledged. This consensus is due in no forms of organization can emerge following early brain
small measure to methodological improvements, in- injury. Based on this evidence, we should expect to find
cluding the availability of imaging techniques to clarify that early injuries in humans are followed by substantial
the relationship between lesion type and language out- reorganization, for language and other cognitive func-
comes. But improved neural imaging is not the only tions as well (Stiles et al., 1998).
relevant factor. Before reviewing a final set of studies in This well-attested finding leads to a prediction that
support of this emergentist view, let us consider several seems, at first glance, to be quite obvious: If plasticity is
crucial methodological factors and their theoretical greater in the young brain, we ought to find a mono-
consequences: timing of lesion onset, lesion type (both tonic relationship between cognitive outcomes and age
site and size), lesion etiology, sample size, and the im- of lesion onset. Although the shape of this function
portance of prospective studies that employ develop- might vary in a number of theoretically interesting ways
mentally sensitive measures. (dropping sharply at some point in a nonlinear pattern,
or decreasing gradually from birth to puberty), later
Intermezzo: Methodological confounds lesions ought to produce worse outcomes than early
ones under any scenario. In fact, the shape of the func-
TIME OF LESION ONSET AND ITS IMPLICATIONS FOR tion governing loss of plasticity in humans is still en-
PLASTICITY There is now a large body of evidence tirely unknown, and it may not even be monotonic (i.e.,
demonstrating that the brains of young animals (espe- plasticity may fall, and then rise again). Many of the
cially mammals) are quite plastic, and that many aspects studies summarized in tables 20.1 and 20.2 have con-
of cortical specialization are activity dependent. That is, flated cases of congenital injury with lesions that were
cortical specialization is determined not by endogenous acquired at points later in childhood. Other studies
growth plans under direct genetic control but by the have divided age of lesion onset into broad epochs, with
input that cortical areas receive from the animal's own mixed and often contradictory results. For example,
body (before and after birth) and from the outside Woods and Teuber (1978) concluded that injuries in
world (for reviews, see Kolb and Gibb, this volume; El- the first year of life are actually more dangerous than
bert, Heim, and Rockstroh, this volume; see also Dea- injuries acquired after age 1, a finding that seems to fly
con, 1997; Elman et al., 1996; Johnson, 1997; Quartz in the face of accumulated evidence for early plasticity
and Sejnowski, 1994, 1997). It has been shown, for ex- in animal models.
ample, that if the cortex of a fetal ferret is rewired so Even more puzzling findings come from Goodman
that input from the eye is fed to auditory cortex, then andYude (1996) and from unpublished data by Vargha-
auditory cortex takes on retinotopic maps (Pallas and Khadem and colleagues (personal communication, July
Sur, 1993). And if slabs of fetal tissue are transplanted 1996, cited in Bates, Vicari, and Trauner, 1999). The
from visual to somatosensory areas (and vice versa), the latter two studies employed relatively large samples (by
transplanted cortex takes on representations appropri- the standards of this field), and both revealed a result
ate to the input received in its new home, as opposed that would not be predicted either by the theory of
to the representations typically seen in their regions of equipotentiality or the theory of irreversible determin-
origin (O'Leary and Stanfield, 1985, 1989; Stanfield ism: In the absence of severe seizures (which seem to
and O'Leary, 1985). Lesion studies of animals also pro- preclude recovery to normal levels of language in most
vide striking evidence for plastic reorganization. For ex- cases), the best outcomes in both verbal and nonverbal
ample, Payne (1999) has shown that cats with early IQ are seen either with congenital lesions (pre- or peri-
bilateral removal of primary visual cortex are virtually natal) or with lesions that occur between 4 and 12 years
indistinguishable from normal on visual tasks, whereas of age! It is possible, of course, that this U-shaped func-
mature cats with the same operation are functionally tion is an artifact of other methodological factors, in-
blind. Webster, Bachevalier, and Ungerleider (1995) cluding etiology (e.g., the medical conditions that lead
BATES AND ROE : LANGUAGE DEVELOPMENT IN CHILDREN WITH BRAIN INJURY 297
to unilateral injury, including stroke, may be quite dif- midsize lesions are large enough to lead to permanent
ferent in infants, preschool children, and children in behavioral impairments, but not quite large enough to
the elementary school years) and the developmental precipitate/cause the brain to reorganize; large lesions
status of the child when testing occurs (e.g., grade result in a better outcome because the animal makes a
school children may have more sophisticated behav- "fresh start," abandoning the inefficient strategies that
ioral strategies at their disposal, permitting them to per- an animal with a midsized lesion still struggles to apply.
form better on standardized tests in the short run, and Preliminary evidence by Thai and colleagues appeared
to exploit their residual plasticity and recover to higher to provide support for the fresh-start hypothesis, re-
levels in the long run). It is also possible that this result flected in a significant U-shaped effect of lesion size on
would not replicate with even larger samples (e.g., ac- early language outcomes (i.e., small lesions or very large
cording to Vargha-Khadem, personal communication lesions were both associated with better language abil-
June 1999, the significant U-shaped function reported ities than those observed in children with lesions in
for her unpublished data by Bates, Vicari, and Trauner, the middle range). However, this U-shaped function
1999, dropped below significance when the same was dropped below significance when the sample size was
expanded to include more than 300 cases). For present doubled (Bates et al., 1997); hence the fresh-start hy-
purposes, we can only conclude that the limits of plas- pothesis still awaits confirmation, and our understand-
ticity and capacity for recovery in young children are ing of the effects of lesion size on language outcomes
still unknown, and that there is ample reason for fami- is still very tenuous.
lies of children with unilateral injury to be hopeful
about their children's chances for recovery. LESION ETIOLOGY AND ITS NEUROLOGICAL CORRE-
LATES The prospective studies reviewed below have
LESION TYPE: SITE AND SIZE Earlier studies (including concentrated entirely on children with congenital in-
most of the studies reviewed in tables 20.1 and 20.2) juries (before 6 months postnatal age) that are usually
were restricted to a global distinction between left- and due to pre- or perinatal stroke (although it is not always
right-hemisphere damage, often established via exter- possible to make a definitive diagnosis of the cause or
nal neurological signs such as hemiparesis. More recent timing of congenital injuries). We should not be sur-
studies have taken advantage of structural brain imag- prised to find that these studies yield different results
ing, and have begun to qualify the crude distinction from those that have included children with trauma or
between LHD and RHD with further distinctions re- tumor (Anderson et al., 1999). Results may also differ
volving around lesion size, the presence/absence of from studies of children who suffered postnatal strokes
subcortical damage, and the lobes of the damaged secondary to cardiac catheterization (which is often as-
hemisphere that are involved. Nevertheless, the term sociated with a lifetime of inadequate oxygen intake),
"focal brain injury" is still defined quite broadly in most and from studies of outcomes following hemispherec-
studies, referring to a single (contiguous) lesion re- tomy in children who have suffered for many years from
stricted to one half of the brain, of any size, cortical intractable seizures. In fact, as Vargha-Khadem and her
and/or subcortical. colleagues have recently reported (see also Ballantyne
Variations in lesion size merit consideration, al- and Trauner, 1999), seizures are the single greatest risk
though evidence on the contribution of lesion size to factor for language and cognitive outcomes in children
language outcomes is still mixed. Lashley's principle of with unilateral brain injury. We also need to consider
mass action (the complement of equipotentiality) pre- when the seizure condition appeared and its subse-
dicts that larger lesions will have greater behavioral re- quent course. For example, Thai, Bates, and others
percussions, with less chance for functional recovery. found no effects of seizure history were found in pro-
His experiments with adult rats supported this idea. spective studies of early language development. How-
However, when Irle (1990) carried out a meta-analysis ever, such studies necessarily conflate relatively benign
of more than 200 lesion studies in monkeys, she found neonatal seizure conditions with more severe and per-
that while lesion size did affect skill reacquisition, the sistent forms of epilepsy that may not appear for
function was curvilinear: Midsized lesions were signifi- months or even years after birth.
cantly more likely to cause permanent damage than
were small or large lesions, the latter including lesions SAMPLE SIZE Sample size is a banal but potentially pow-
of up to 60% of total brain tissue. At first glance, this erful factor to consider when evaluating studies with
result is counterintuitive. But Irle suggests a compelling discrepant results. There are massive individual differ-
explanation, which she calls the "fresh-start" hypothesis: ences in the rate and nature of language development
Small lesions have little effect because they are small; in perfectly normal children (Bates, Dale, and
298 LANGUAGE
Thai, 1995). Unilateral injuries are superimposed upon brain, relying primarily on studies by the San Diego
this landscape of variation, which means that single-case group and their collaborators.
studies or small-sample studies must be interpreted with
caution. Consider a recent report by Stark and Mc- Phase III: The emergentist view
Gregor (1997) on two cases of childhood hemispher-
ectomy, one to the left and the other to the right All of the studies considered here involve children with
hemisphere. These authors report an "adult-like" pat- congenital injuries (prior to 6 months of postnatal age),
tern: Selectively greater deficits for language in the case producing a single contiguous lesion (though often
of LHD, compared with a more even profile of delay in very large) confined to one side of the brain. These
the case of RHD, results interpreted to support a mild lesions are due primarily to pre- or perinatal stroke and
variant of innate/irreversible determinism. However, have, in all cases, been confirmed by CT or MRI. Chil-
these two cases contrast sharply with Vargha-Khadem's dren were excluded if the lesion was due to tumor,
case study of Alex, a child with severe LHD and intrac- trauma, arteriovenal malformation, or any form of dif-
table seizures who was virtually mute when he under- fuse or multifocal brain damage, or if they suffered
went hemispherectomy at 8 years of age (Vargha- from any serious medical conditions (other than sei-
Khadem et al., 1997). After an initial delay, Alex went zures subsequent to the lesion itself). All children come
on to attain fluent control over language (with no ar- from families in which the predominant language is
ticulatory problems or specific delays in grammar) com- English, and although they represent a broad socio-
mensurate with his mental age. Although case studies economic spectrum, children of middle-class parents
can be quite informative in showing us the range of tend to predominate (as they do in much of the behav-
outcomes that are possible following various forms of ioral literature in developmental psychology).
unilateral injury, they should not be used as the basis The San Diego group and their collaborators have
for generalizations about the correlation between vari- conducted cross-sectional and longitudinal studies of
ous forms of injury and their linguistic sequelae. this clinical group for approximately 15 years, focusing
on many aspects of development including visual-spatial
DEVELOPMENTAL SENSITIVITY AND TIMING OF LAN- cognition, attention and hemispatial neglect, and per-
GUAGE TESTING Two related factors are at issue here. ception and production of facial and vocal affect. We
First, the amount of time that has elapsed since lesion will concentrate here on studies of speech and lan-
onset may influence how "recovered" a child appears guage. (For reviews of development in other domains,
during testing. That is, when children are tested in the see Stiles, this volume; also see Stiles, 1995; Stiles et al.,
middle school years or beyond, those who suffered their
1998. For more detailed reviews of language develop-
lesions earlier in life have also had more time to reor-
ment in this population, see Bates et al., in press; Bates,
ganize and recover. Second, there may be specific ef-
Vicari, and Trauner, 1999; Broman and Fletcher, 1999;
fects of lesion type that are evident only in particular
phases of development, when children start to come to Elman et al., 1996.)
terms with the demands of a new language task. For We will start with results of cross-sectional studies that
both these reasons, studies that focus on the early stages focus on development after 5 years of age, which largely
of language may yield qualitative information about the confirm results of other large-sample studies of lan-
initial state of the system, and about the processes in- guage outcomes in this population. Then we will end
volved in plastic reorganization of language and other with studies that have examined the acquisition of lan-
cognitive functions. guage in this population, starting in the first year of life.
Most of the studies summarized in tables 20.1 and These studies demonstrate that side- and site-specific
20.2 have been retrospective in nature, testing children biases are present early in life; although the lesion-
well after the period in which language is usually ac- symptom correlations observed in these studies do not
quired and (we presume) after much of the recovery map directly onto the patterns observed in adults, dif-
for which this population is so famous has already oc- ferent lesions have different effects on early language
curred. For the remainder of this chapter, we will con- learning that must be overcome. The fact that they are
centrate on developmental studies of children with overcome (disappearing entirely by 5-7 years of age in
focal brain injury that take the children's level of de- the domain of language) provides powerful evidence
velopment into account, tracking change over time for the plastic and experience-dependent nature of
using age- and stage-appropriate language outcome brain and behavioral development. Furthermore, the
measures. In particular, we will focus on prospective stud- evidence suggests that language learning itself is the cat-
ies of children with congenital injuries to one side of the alyst for this reorganization.
BATES AND ROE: LANGUAGE DEVELOPMENT IN CHILDREN WITH BRAIN INJURY 299
Starting with studies of language outcomes at later pared directly in an age-by-side-of-lesion design, using
stages of development, Bates, Vicari, and Trauner age-based z-scores derived from relatively large samples
(1999) summarize performance by 43 English-speaking of age-matched controls on the van Lancker and
children from the San Diego sample (28 LHD and 15 Kempler Familiar Phrases Test. As can be seen in figure
RHD) and 33 Italian-speaking children (18 LHD and 20.1, RHD and LHD adult patients display a double dis-
15 RHD) from Rome, tested cross-sectionally between sociation on this task. LHD patients have more difficulty
3 and 14 years of age. Mean full-scale IQs were in the on familiar phrases, whereas patients with RHD are sig-
low-normal range (94-97), although the range was nificantly worse on idioms or familiar phrases matched
quite broad (from 40 to 140). There were also more for length and complexity. As figure 20.1 also shows,
cases in the below-80 range (which some investigators child patients displayed absolutely no evidence for a
use as a cut-off for mild mental retardation) than we double dissociation: Children with LHD versus RHD
would expect if we were drawing randomly from the both performed significantly below normal controls as
normal population. However, there were absolutely no a group, but did not differ significantly from each other.
differences between LHD and RHD children in full- Even more important, the child patients performed
scale, verbal or nonverbal IQ. For the Italian sample, within the low-normal range on both measures, while
Bates and colleagues also summarize performance on the adult patients performed many standard deviations
several language tests, including lexical comprehension below their age-matched controls on their weakest mea-
(an Italian version of the Peabody Picture Vocabulary sure (i.e., novel phrases for patients with LHD; familiar
Test), lexical production (an Italian adaptation of the phrases for patients with RHD). In other words, the chil-
Boston Naming Test), grammatical comprehension dren were not significantly impaired (i.e., their perfor-
(the Token Test and an Italian version of the Test of mance did not reach criteria required to establish the
Receptive Grammar), and semantic category fluency. existence of a language deficit) following either right-
Again, although brain-injured children performed sig- or left-hemisphere damage, and no selective effects of
nificantly below normal controls on all language mea- lesion side were detected.
sures except the TROG, there was no evidence for a The second study, by Dick and colleagues (1999),
difference between LHD and RHD on any measure. compared performance by children and adults with
Furthermore, when mental age was controlled in anal- unilateral brain injury and their age-matched controls
yses of covariance, the difference between brain-injured in an on-line auditory sentence comprehension test
children and normal controls disappeared for every that contrasts syntactically simple sentences (active and
measure except the Boston Naming Test. subject clefts that follow canonical word order) with syn-
These cross-sectional results suggest that the plastic tactically complex sentences (passives and object clefts
reorganization for which this population is known takes that violate canonical word order). All sentences were
place prior to 5-7 years of age. As a result, children with fully grammatical, and semantically reversible. All
early focal brain injury recover far better (relative to groups (including normal controls) displayed the same
age-matched controls) than do adults with comparable basic profile of lower accuracy on noncanonical sen-
injuries. Although this conclusion has been around for
quite a while, and there is a large body of evidence on
plasticity from animal research to support it, adults and
children have rarely been compared directly on a com-
mon set of measures. More direct comparisons would
be helpful in assessing the nature and magnitude of this
presumed plasticity. We are aware of just three studies
(all by the San Diego group and their collaborators)
that have compared school-age children and adults di-
rectly on the same measures (other than verbal and
nonverbal IQ), using z-scores based on data from age-
matched controls.
The first study in this series, by Kempler and col-
leagues (1999), compared adults with RHD and LHD
to a sample of 6-12-year-old children who had suffered
comparable injuries (also due to cerebrovascular acci-
dents, or CVA) during the pre/perinatal period. Child FIGURE 20.1 Comprehension of novel and familiar phrases
and adult patients with LHD versus RHD were com- by children and adults with left versus right hemisphere injury.
300 LANGUAGE
tences (object clefts and passives). Among the children, qualitative variations in their symptoms, reflecting dif-
group by sentence type interactions were obtained in- ferent aphasia subtypes (e.g., more morphological and
dicating that (1) the youngest normal children were at omission errors in Broca's aphasics, more lexical errors
a greater disadvantage than older children on the more in Wernicke's aphasics). One small illustration of these
difficult noncanonical sentence types; (2) as a group, results can be seen in figure 20.2a, which plots the total
brain-injured children showed a greater disadvantage number of errors per proposition in children versus
on the difficult sentences than their age-matched con- adults within each lesion group, and figure 20.2b,
trols; however, (3) the brain-injured children were still which plots the same data for LHD and RHD children
within the normal range for their age; and, most im- and adults in z-scores based on performance by age-
portant for our purposes here, (4) there were no sig- appropriate controls. Figure 20.2a shows that error
nificant differences between children with LHD and rates are certainly higher for children than adults (as
children with RHD on any of the sentence types. In we have known for many years), but figure 20.2b shows
contrast with these findings for children, adults with that LHD and RHD children are very close to normal
unilateral brain injury were severely impaired, espe- (with z-scores close to zero) while the worst aphasics
cially on the noncanonical sentences. Direct compari- produce error rates that are orders of magnitude
sons of adults and children with LHD clearly higher than normal controls (whose error rate is ex-
demonstrate that LHD is associated with receptive tremely small, leading to very small standard devia-
agrammatism in adults but not in children. tions) . Although these results are not surprising, in view
The third study in this series focused on language of the accumulated evidence for plasticity following
production instead of comprehension, based on sam- early brain injury in humans and in other species, they
ples of free speech (Bates et al., 1999) collected within document this phenomenon with exceptional clarity.
the framework of a biographical interview tailored to This brings us to a summary of evidence by the same
reflect the different interests of children and adults. research group looking at the first stages of language
Participants included 38 brain-injured children (24 development, prior to 5-7 years of age.
LHD, 14 RHD) between 5 and 8 years of age, 38 normal In a study focusing on the earliest stages of language
controls matched for age and gender, 14 adults with development, Thai and colleagues (1991) describe re-
LHD (including 3 Broca's aphasics, 3 Wernicke's apha- sults for 27 congenitally brain-damaged infants between
sics, 5 anomic aphasics, and 3 nonaphasic patients), 7 12 and 35 months of age, using an early version of the
adults with RHD, and 12 adult controls in the same MacArthur Communicative Development Inventories,
range of age and education. The structured interviews or GDI (Fenson et al., 1993), a parent-report instru-
were videotaped and transcribed following conventions ment for the assessment of early lexical and grammati-
of the Child Language Data Exchange System, and cal development. Delays in word comprehension in the
coded into various categories assessing amount of very first stages of development were actually more
speech (number of word types, word tokens, mor- common in children with RHD. Delays in first word
phemes, and utterances), length (mean length of utter- production occurred for almost all the brain-injured
ance in morphemes, or MLU), grammatical complexity children, regardless of lesion side or site, but tended to
(number of complex syntactic structures, in both types be more severe in children with left posterior damage—
and tokens), and errors (word omissions, morphologi- an apparent reversal of the expected association be-
cal errors, lexical errors). Although it was generally true tween comprehension deficits and damage to Wer-
that children talk far less than adults (including adult nicke's area.
aphasics), when proportion scores were used to correct Bates and colleagues (1997) followed up on Thai and
for overall amount of output, results were exceedingly colleagues (1991) with a larger sample, using a combi-
clear: (1) There were absolutely no differences between nation of GDI data and free speech to assess early lan-
children with LHD versus RHD on any measure. (2) In guage development in 53 children between 10 and 40
this open-ended free-speech task, there were also very months of age (36 LHD, 17 RHD), including 18 of the
few differences between brain-injured children (com- 27 cases from Thai's study. The study was divided into
bining LHD and RHD) and their controls (the excep- three cross-sectional epochs (although many of the chil-
tions were small but significant disadvantages for FL dren participated in more than one): a period focusing
children as a group in number of word omission errors on the dawn of word comprehension, word production,
and in number of word types). (3) In striking contrast and gesture (26 children from 10 to 17 months); a sec-
to the child data, there were huge differences between ond substudy focusing on word production and the
adults with LHD versus RHD on virtually every measure, emergence of grammar (29 children from 19 to 31
in the predicted directions. (4) LHD adults also showed months); and an analysis of grammatical development
BATES AND ROE: LANGUAGE DEVELOPMENT IN CHILDREN WITH BRAIN INJURY 301
from free-speech samples (30 children from 20 to 44
months). Performance at these various stages of devel-
opment was evaluated in comparisons based on lesion
side, size, and site (i.e., whether or not the frontal lobes
or temporal lobes were involved). There were no effects
of lesion size in any of these analyses (including a failure
to replicate the U-shaped effect of lesion size described
by Thai and colleagues). Interesting effects of lesion
side and intrahemispheric lesion site did emerge, but
in complex patterns that are surprising from the point
of view of the adult aphasia literature.
Between 10 and 17 months, delays in receptive lan-
guage were particularly evident in children with RHD
(i.e., more RHD cases than we would expect by chance
fell into bottom tenth percentile for word comprehen-
sion) . By contrast, the LHD children performed within
the normal range on word comprehension, even if their
lesions involved the temporal lobe (the presumed site
of Wernicke's area, which is implicated in moderate to
severe forms of receptive aphasia in adults). However,
there was no significant difference between LHD and
RHD on direct statistical comparisons, so the RHD dis-
advantage is not robust and should be investigated fur-
ther. There was also a significant RHD disadvantage in
the development of communicative and symbolic ges-
ture, and this time the RHD disadvantage did reach sig-
nificance in a direct LHD/RHD comparison. This result
is also surprising, since deficits in the production of
symbolic gestures are atypically associated with left-
hemisphere damage when they occur in adults (Good-
glass, 1993). Finally, Bates and colleagues do report a
selective delay in expressive vocabulary for children
with LHD. However, in line with the earlier report by
Thai and colleagues, this disadvantage was only evident
in children whose lesions involved the temporal lobe.
The second substudy followed children's language
development between 19 and 31 months, when the so-
called vocabulary burst is said to occur (e.g., an intense
period of development for vocabulary/lexical produc-
tion) and when children's comprehension is often so
vast it is difficult to measure. This is also the period in
which children typically start to combine words, fol-
lowed by the emergence of grammatical inflections and
function words. For the 29 children whose scores on
this scale were obtained, a selective disadvantage for
children with LHD appeared both for expressive vocab-
ulary and the emergence of grammar (with no evidence
FIGURE 20.2 (A) The mean number of errors per proposi- whatsoever for a dissociation between grammatical and
tion for normally developing children, adult controls, and lexical production). However, this LHD disadvantage
children and adults with right or left hemisphere damage. was due once again to children with left temporal in-
LHD patients are further subdivided by aphasia type. (B) Z-
score error rates per proposition for children versus adults volvement, in contrast with the typical adult pattern in
with right or left hemisphere damage. LHD patients are fur- which expressive deficits (especially nonfluent aphasia)
ther subdivided by aphasia type. are usually associated with left frontal involvement (i.e.,
302 LANGUAGE
Broca's area and adjacent cortical and subcortical tribution of the right hemisphere may be much less
regions). Similar delays in expressive vocabulary and important, so that control may shift (in the undamaged
grammar appeared when children with frontal lobe in- brain) to rapid, automatic processes mediated primarily
volvement were compared with children whose lesions by the left hemisphere.
spared the frontal lobe. However, in contrast with the In the third and final subgroup of children, Bates and
asymmetric left temporal disadvantage that we have just colleagues (1997) collected free-speech samples be-
discussed, this frontal effect was perfectly symmetric: tween 21 and 44 months of age. As their GDI scores
Delays were equally severe with left frontal or right fron- predicted, the MLU scores of children with damage that
tal lesions. encompassed the left temporal region of their brain
Curiously, an abnormal proportion of the children were significantly lower than normal, and significantly
with RHD were also producing a higher than normal lower than scores for brain-injured children whose le-
number of function words for their vocabulary size. As sions spared this region (including all children with
described in some detail by Bates, Bretherton, and Sny- RHD and the subset of LHD children with no temporal
der (1988) and by Bates and colleagues (1994), such involvement). Children with right or left frontal dam-
overuse of function words for children in the early age also still looked delayed, but this difference was not
stages of vocabulary development (i.e., under 400 statistically significant in the 21-44-month subsample.
words) is definitely not a sign of precocious grammar. Vicari and colleagues (in press) attempted a partial
In fact, children who overuse pronouns and other func- replication of the Bates results for early lexical devel-
tion words in the early stages tend to be relatively slow opment, administering an Italian version of the Mac-
in grammatical development later on. For these chil- Arthur GDI to the parents of 43 children between 13
dren, function words tend to appear in frozen or rote and 46 months of age. Their study differed from the
expressions like "I wan dat," a style of early expressive methods used by Bates in two crucial respects: Children
language that has been called "pronominal style," or beyond the age range covered by the MacArthur GDI
"holistic style." At the opposite end of the continuum were included in the study (which means that they
are children who avoid function words in their first could not use age-based percentile scores), and parents
word combinations, producing telegraphic utterances were given the Infant or the Toddler version of the Mac-
like "Adam truck" or "Mommy sock". This style of early Arthur based not on age but on their child's current
expressive language has been referred to as "nominal level of linguistic ability (children who were still in the
style" or "analytic style." Given the terms "holistic" and one-word stage were assigned the Infant form, but chil-
"analytic," which are often attributed to right- versus dren who were starting to combine words were assigned
left-hemisphere processing, respectively, one might the Toddler form). For these reasons, the studies are
have predicted that holistic style would be more com- not entirely comparable, but results replicate and ex-
mon in children with LHD (who are presumably relying tend Bates's findings in some interesting directions.
more on holistic right-hemisphere processes to acquire First, Vicari and colleagues also report a massive across-
language). This prediction is roundly contradicted by the-board delay in early vocabulary development for
the Bates study, in which holistic style was robustly as- brain-injured children as a group. Hence, even though
sociated with RHD (indicating that the overproduction the long-term prospects for these children are relatively
of function words in early speech reflects reliance on good, it is obviously hard to get language off the ground
the intact left hemisphere). Bates and colleagues when significant damage has occurred to either hemi-
(1997) suggest that children with RHD are relying sphere. Second, Vicari and colleagues report a large
heavily on the more precise acoustic analysis and/or and significant interaction between side of lesion (LHD
greater acoustic memory available in the left hemi- versus RHD) and stage of language development (sin-
sphere, storing up frozen expressions that they are un- gle word versus multiword). Among children who were
able to segment or understand beyond a relatively still in the one-word stage, LHD were significantly
superficial level of analysis (rather like an American slower in vocabulary development than RHD (because
who says "Gesundheit" when someone sneezes, with no 10 out of 12 of the one-word-stage children with LHD
idea regarding the structure or meaning of that word had temporal lobe involvement, a specific replication
in German). This would mean, in turn, that right-hemi- of Bates's left temporal findings was not possible). By
sphere processes are very important in the early stages contrast, among children who were now in the multi-
of language learning for the breakdown of acoustic ma- word stage, the LHD disadvantage had disappeared en-
terial and its integration into a larger cognitive system. tirely. In fact, LHD children in the multiword group had
However, once the material has been analyzed, under- a numerical advantage over their RHD counterparts.
stood, and integrated into a larger framework, the con- This advantage was not statistically significant, but it
BATES AND ROE: LANGUAGE DEVELOPMENT IN CHILDREN WITH BRAIN INJURY 303
contributed to the robust interaction between language emergence of word comprehension and communica-
stage and lesion side. Vicari and colleagues suggest that tive gesture, progress in expressive language (both lex-
recovery from this initial delay may begin very early for ical and grammatical) seems to be delayed with frontal
some children, and may be forced in part by the delay damage (to either side of the brain) and with temporal
itself. That is, children who are particularly disadvan- damage (but in this case, temporal damage restricted
taged in the first stages of language acquisition (e.g., to the left hemisphere). In other words, there is an early
LHD cases) may be forced to abandon a failing strategy bias that predisposes the left hemisphere to "take over"
in favor of some alternative approach, leading to earlier rapid and efficient production of words and sentences,
and (ultimately) more successful language learning. a development that may also result in the emergence of
Reilly, Bates, and Marchman (1998) conducted a left-hemisphere specialization for many aspects of re-
cross-sectional study of 15 RHD and 15 LHD children, ceptive language as well. In the absence of evidence to
between 3 and 12 years of age, using a story-telling for- the contrary, one might have assumed that this early
mat (the well-known Frog Story narratives—Herman left-hemisphere advantage for speech/language pro-
and Slobin, 1994) to assess lexical, grammatical, and duction (but not reception) has a motor base. And yet
discourse development. For children between 3 and 6 several studies by the San Diego group suggest that the
years of age, the now-familiar left temporal disadvan- source of this left-hemisphere bias lies primarily within
tage was observed in syntactic complexity and in persis- the temporal lobe, a region that is supposed to be spe-
tence of morphological errors. However, this effect of cialized for perception rather than production.
lesion site was not observed in children between 6 and In this regard, Bates and colleagues note that some
12 years of age. Among the older children, there were children with severe otitis media (i.e., middle-ear infec-
still significant differences between focal lesion chil- tions) also show selective delays in the emergence of
dren (LHD and RHD combined) and their age- expressive (but not receptive) language. Why would
matched controls on a number of measures, but the middle-ear impedance have greater effects on language
focal lesion children were nevertheless performing production than on comprehension? The answer may
within the normal or low-normal range. Hence the lie in a simple fact: Language learning is not the same
Reilly results for grammar suggest a later variant of the thing as fluent language use. When a child is trying to
recovery pattern that Vicari's group observed within break into the language system for the first time, the
the lexical domain. amount of perceptual analysis required to produce her
Because the Reilly and Vicari studies are both cross- own versions of a new word is greater than the amount
sectional, it would be very useful to replicate these re- of perceptual analysis that she needs to recognize the
sults with longitudinal samples. Although their results word (especially if she is asked to recognize that word
are still preliminary, based on a relatively small sample, in a richly supportive social and physical context, which
Reilly and colleagues (Losh, Reilly, and Bates, 1996) can be integrated with the acoustic signal to achieve
have tested a longitudinal subgroup across the 5-7-year comprehension). If these assumptions are correct, we
age range, which seemed to be a watershed in their can put the story together as follows: Left temporal
cross-sectional study. They report that children with left regions may be particularly well suited (perhaps at or
temporal involvement do indeed move sharply upward before birth) for the extraction of perceptual detail. In-
in syntax and morphology across this age range, scoring deed, there is ample evidence from visual-spatial pro-
numerically above children with RHD at the later time cessing in adults to support this view; hence the
point. The general picture seems to be one in which hypothesized "perceptual detail advantage" would not
children with LHD display sharper or "steeper" growth be specific to language, or even to audition. However,
functions, while children with RHD show a flatter pro- such a bias would be particularly relevant in the first
file of growth in the language domain. stages of language learning, leading (in the absence of
injury) to the establishment of left-hemisphere domi-
Conclusions nance. What these prospective studies do clearly show
is that this bias is "soft," and can be overcome. Indeed,
Putting these lines of evidence together, we may con- by 5-7 years of age the initial disadvantages associated
clude (or perhaps hypothesize) that the infant brain with left-hemisphere damage seem to have disappeared,
contains strong biases that, in the absence of early brain or at least fallen below the levels detectable with the
damage, guarantee the eventual emergence of left- measures that we have developed so far.
hemisphere specialization for language. Although (if Finally, it appears from these studies that the emer-
anything) the right hemisphere seems to play a more gence of organization for language (in the undamaged
important (or at least equally important) role in the brain) and reorganization for language (in the dam-
304 LANGUAGE
aged brain struggling to overcome initial biases) both BATES, E., D. THAL, B. L. FINLAY, and B. CLANCY, in press.
occur within the period in which language is acquired, Early language development and its neural correlates. In
Handbook of Neuropsychology, Vol. 7: Child Neurology, 2nd ed.,
i.e., somewhere between birth and 5 years of age. We F. Boiler andj. Grafman, series eds.; I. Rapin and S. Sega-
may speculate that this correlation between brain and lowitz, vol. eds. Amsterdam: Elsevier.
behavioral development is no accident. In fact, we pro- BATES, E., D. THAL, D. TRAUNER,J. FENSON, D. ARAM,J. EISELE,
pose that learning itself plays a major role in organizing and R. NASS, 1997. From first words to grammar in children
the brain for efficient language use, as children struggle with focal brain injury. Dev. Neuropsychol. 13:447-476.
BATES, E., S. VICARI, and D. TRAUNER, 1999. Neural mediation
to find an optimal solution to the challenges associated of language development: Perspectives from lesion studies
with language and communication. of infants and children. In Neurodevelopmental Disorders, H.
Tager-Flusberg, ed. Cambridge, Mass.: MIT Press, pp. 533-
581.
REFERENCES BATES, E., B. WULFECK, M. OPIE, J. FENSON, S. KRIZ,J. REILLY,
N. DRONKERS, L. MILLER, R. JEFFRIES, and K. HERBST, 1999.
ALAJOUANINE, T., and F. LHERMITTE, 1965. Acquired aphasia
Comparing free speech in children and adults with left- vs.
in children. Brain 88:653-662.
right-hemisphere injury (Abstract). Brain Lang. 69:377-
ANDERSON, S. W., A. BECHARA, H. DAMASIO, D. TRANEL, and
379.
A. R. DAMASIO, 1999. Impairment of social and moral be-
BERMAN, R. A., and D. I. SLOBIN, 1994. Relating Events in Nar-
havior related to early damage to human prefrontal cortex.
rative: A Cross-linguistic Developmental Study [in collaboration
Nature Neurosci. 2:1032-1037.
with Ayhan Aksu-Koc et al.]. Hillsdale, N.J.: Erlbaum.
ARAM, D. M., B. EKELMAN, D. ROSE, and H. WHITAKER, 1985.
BERNHARDT, M., 1897. Die Erkrankungen der peripherischen
Verbal and cognitive sequelae following unilateral lesions
Nerven [Peripheral nerve diseases]. In Spezielle Pathologie
acquired in early childhood.J. Clin. Exp. Neuropsychol. 7:55-
und Therapie 9 [Specific Pathology and Therapy 9], H. Noth-
78.
nagel, ed. Vienna: Alfred Holder.
ARAM, D., B. EKELMAN, and H. WHITAKER, 1986. Spoken syn-
BOTHA, R. P., 1989. Challenging Chomsky: The Generative Garden
tax in children with acquired unilateral hemisphere lesions.
Game. Oxford, UK: Basil Blackwell.
Brain Lang. 27:75-100.
BROMAN, S., andj. M. FLETCHER, eds., 1999. The ChangingNer-
ARAM, D. M., B. EKELMAN, and H. WHITAKER, 1987. Lexical
vous System: Neurobehavioral Consequences of Early Brain Disor-
retrieval in left and right brain-lesioned children. Brain
Lang. 31:67-87. ders. New York: Oxford University Press.
BROWN, C. M., and P. HAGOORT, eds., 1999. TheNeurocognition
ARAM, D. M., S. C. MEYERS, and B. L. EKELMAN, 1990. Fluency
of conversational speech in children with unilateral brain of Language. New York: Oxford University Press.
lesions. Brain Lang. 38:105-121. CLARUS, A., 1874. Ueber Aphasie bei Kindern [Aphasia in chil-
BALLANTYNE, A. O., K. M. SCARVIE, and D. TRAUNER, 1994. dren] . Jahresb. Kinderheilkd. 7:369-400.
Verbal and performance IQ patterns in children after peri- COTARD, J., 1868. Etude sur I'atophie partielle du cerveau. These
natal stroke. Dev. Psychol. 10:39-50. de Paris.
BALLANTYNE, A. O., and D. A. TRAUNER, 1999. Neuropsycho- CURTISS, S., 1977. Genie: A Psycholinguistic Study of a Modern-
logical profiles of infants and children with early focal brain Day Wild Child. New York: Academic Press.
damage. JINS 5:145. DEACON, T, 1997. The Symbolic Species: The Co-evolution of Lan-
BASSER, L., 1962. Hemiplegia of early onset and the faculty of guage and the Brain. New York: Norton.
speech with special reference to the effects of hemispher- DENNIS, M., 1980. Capacity and strategy for syntactic compre-
ectomy. Brain 85:427-460. hension after left or right hemidecortication. Brain Lang.
BATES, E., 1999. Plasticity, localization and language develop- 10:287-317.
ment. In The Changing Nervous System: Neurobehavioral Con- DENNIS, M., and B. KOHN, 1975. Comprehension of syntax in
sequences of Early Brain Disorders, S. Broman andj. M. Fletcher, infantile hemiplegics after cerebral hemidecortication.
eds. New York: Oxford University Press, pp. 214-253. Brain Lang. 2:472-482.
BATES, E., in press. On the nature and nurture of language. DENNIS, M., M. LOVETT, and C. A. WIEGEL-CRUMP, 1981. Writ-
In Frontiere della Biologia [Frontiers of Biology]. The brain of homo ten language acquisition after left or right hemidecortica-
sapiens, R. Levi-Montalcini, D. Baltimore, R. Dulbecco, and tion in infancy. Brain Lang. 12:54-91.
F. Jacob, series eds.; E. Bizzi, P. Calissano, and V. Volterra, DENNIS, M., and H. A. WHITAKER, 1976. Language acquisition
vol. eds. Rome: Giovanni Trecanni. following hemidecortication: Linguistic superiority of the
BATES, E., I. BRETHERTON, and L. SNYDER, 1988. From First left over the right hemisphere. Brain Lang. 3:404-433.
Words to Grammar: Individual Differences and Dissociable Mech- DENNIS, M., and H. WHITAKER, 1977. Hemispheric equipoten-
anisms. New York: Cambridge University Press. tiality and language acquisition. In Language Development and
BATES, E., P. S. DALE, and D. THAL, 1995. Individual differ- Neurological theory, S. J. Segalowitz and F. A. Gruber, eds. New
ences and their implications for theories of language de- York: Academic Press, pp. 93-106.
velopment. In Handbook of Child Language, P. Fletcher and DICK, E, B. WULFECK, E. BATES, D. SALTZMAN, N. NAUCLER,
B. MacWhinney, eds. Oxford, UK: Basil Blackwell, pp. 96- and N. DRONKERS, 1999. Interpretation of complex syntax
151. in aphasic adults and children with focal lesions or specific
BATES, E., V. MARCHMAN, D. THAL, L. FENSON, P. S. DALE, language impairment (Abstract). Brain Lang. 69:335-336.
J. S. REZNICK, J. REILLY, andj. HARTUNG, 1994. Develop- EISELE, J., and D. ARAM, 1993. Differential effects of early
mental and stylistic variation in the composition of early hemisphere damage on lexical comprehension and pro-
vocabulary.J. Child Lang. 21:85-123. duction. Aphasiology 7:513-523.
BATES AND ROE: LANGUAGE DEVELOPMENT IN CHILDREN WITH BRAIN INJURY 305
EISELE, J., and D. ARAM, 1994. Comprehension and imitation NEWMEYER, F. J., 1980. Linguistic Theory in America. New York:
of syntax following early hemisphere damage. Brain Lang. Academic Press.
46:212-231. NEWMEYER, F. J., 1997. Genetic dysphasia and linguistic the-
EISELE, J., and D. ARAM, 1995. Lexical and grammatical de- ory.J. Neurolinguistics 10:47-73.
velopment in children with early hemisphere damage: A OJEMANN, G. A., 1991. Cortical organization of language. J.
cross-sectional view from birth to adolescence. In The Hand- Neurosci. 11:2281-2287.
book of Child Language, P. Fletcher and B. MacWhinney, eds. O'LEARY, D. D., and B. B. STANFIELD, 1985. Occipital cortical
Oxford, UK: Basil Blackwell, pp. 664-689. neurons with transient pyramidal tract axons extend and
ELMAN, J. L., E. BATES, M.JOHNSON, A. KARMILOFF-SMITH, D. maintain collaterals to subcortical but not intracortical tar-
PARISI, and K. PLUNKETT, 1996. RethinkingInnateness: A Con- gets. Brain Res. 336:326-333.
nectionist Perspective on Development. Cambridge, Mass.: MIT O'LEARY, D. D., and B. B. STANFIELD, 1989. Selective elimi-
Press. nation of extended by developing cortical neurons is de-
FENSON, L., P. S. DALE, J. S. REZNICK, D. THAL, E. BATES, J. pendent on regional locale: Experiments utilizing fetal
HARTUNG, S. PETHICK, and J. REILLY, 1993. The MacArthur cortical transplants.J. Neurosci. 9:2230-2246.
Communicative Development Inventories: User's Guide and Tech- O'NEILL, Y. V., 1980. Speech and Speech Disorders in Western
nical Manual San Diego: Singular Publishing Group. Thought Before 1600. Westport, Conn.: Greenwood Press.
FLETCHER, P., and B. MACWHINNEY, eds., 1995. Handbook of PALLAS, S. L., and M. SUR, 1993. Visual projections induced
Child Language. Oxford, UK: Basil Blackwell. into the auditory pathway of ferrets: II. Corticocortical
FODOR, J. A., 1983. The Modularity of Mind: An Essay on Faculty connections of primary auditory cortex. J Comp. Neurol.
Psychology. Cambridge, Mass.: MIT Press. 337:317-333.
GAZZANIGA, M. S., and R. W. SPERRY, 1967. Language after PAYNE, B. R., 1999. Immature visual cortex lesions: Global re-
section of the cerebral commissures. Brain 90:131-148. wiring, neural adaptations, and behavioral sparing. In The
GESCHWIND, N., and E. KAPLAN, 1962. A human cerebral de- Changing Nervous System: Neurobehavioral Consequences of Early
connection syndrome. Neurology 12:675-685. Brain Disorders, S. Broman andj. M. Fletcher, eds. New York:
GOODGLASS, H., 1993. Understanding Aphasia. San Diego: Ac- Oxford University Press, pp. 114-145.
ademic Press. PINKER, S., 1994. The Language Instinct: How the Mind Creates
GOODMAN, R., and C. YUDE, 1996. IQ and its predictors in Language. New York: William Morrow.
childhood hemiplegia. Dev. Med. Child Neural. 38:881-890. QUARTZ, S. R., and T. J. SEJNOWSKI, 1994. Beyond modularity:
GOPNIK, M., 1990. Feature-blind grammar and dysphasia. Na- Neural evidence for constructivist principles in develop-
ture 344(6268) :715. ment. Behav. Brain Sci. 17:725—726.
GOPNIK, M., and M. B. CRAGO, 1991. Familial aggregation of QUARTZ, S. R., and T. J. SEJNOWSKI, 1997. The neural basis of
a developmental language disorder. Cognition 39:1-50. cognitive development: A constructivist manifesto. Behav.
IRLE, E., 1990. An analysis of the correlation of lesion size, Brain Sci. 20:537.
localization and behavioral effects in 283 published studies RASMUSSEN, T, and B. MILNER, 1977. The role of early left
of cortical and subcortical lesions in old-world monkeys. brain injury in determining lateralization of cerebral
Brain Res. Rev. 15:181-213. speech functions. Ann. N.Y. Acad. Sci. 229:355-369.
JOHNSON, M. H., 1997. Developmental Cognitive Neuroscience: An REILLY, J., E. BATES, and V. MARCHMAN, 1998. Narrative dis-
Introduction. Cambridge, Mass.: Blackwell Publishers. course in children with early focal brain injury. Brain Lang.
KEMPLER, D., D. VAN LANCKER, V. MARCHMAN, and E. BATES, 61:335-375.
1999. Idiom comprehension in children and adults with RICE, M., ED., 1996. Toward a Genetics of Language. Mahwah,
unilateral brain damage. Dev. Neuropsychol. 15:327-349. N.J.: Erlbaum.
LASHLEY, K. S., 1950. In search of the engram. In Symposia of RIVA, D., and L. CAZZANIGA, 1986. Late effects of unilateral
brain lesions before and after the first year of life. Neuro-
the Society for Experimental Biology. No. 4: Physiological Mecha-
psychologia 24:423-428.
nisms and Animal Behaviour. New York: Academic Press, pp.
RIVA, D., L. CAZZANIGA, C. PANTALEONI, N. MILANI, and E.
454-482.
FEDRIZZI, 1986. Acute hemiplegia in childhood: The neu-
LENNEBERG, E. H., 1967. Biological Foundations of Language.
ropsychological prognosis.J Fed. Neurosci. 2:239-250.
New York: Wiley.
SATZ, P., E. STRAUSS, and H. WHITAKER, 1990. The ontogeny
LEVY, J., R. D. NEBES, and R. W. SPERRY, 1971. Expressive lan- of hemispheric specialization: Some old hypotheses revis-
guage in the surgically separated minor hemisphere. Cortex ited. Brain Lang. 38:596-614.
7:49-58. STANFIELD, B. B., and D. D. O'LEARY, 1985. Fetal occipital
LOSH, M., J. S. REILLY, and E. BATES, 1996. Language develop- cortical neurones transplanted to the rostral cortex can
ment: Neural plasticity in children with early focal brain injury. extend and maintain a pyramidal tract axon. Nature
Poster session presented at the 8th Annual Meeting of the 313(5998):135-137.
American Psychological Association, San Francisco. STARK, R. E., and K. K. MCGREGOR, 1997. Follow-up study of
McFiE, J., 1961. Intellectual impairment in children with lo- a right- and left-hemispherectomized child: Implications for
calized post-infantile cerebral lesions. J Neurol. Neurosurg. localization and impairment of language in children. Brain
Psychiatry 24:361 -365. Lang. 60: 222-242.
MUTER, V., S. TAYLOR, and F. VARGHA-KHADEM, 1997. A lon- STILES, J., 1995. Plasticity and development: Evidence from
gitudinal study of early intellectual development in hemi- children with early focal brain injury. In Maturational Win-
plegic children. Neuropsychologia 35:289-298. dows and Adult Cortical Plasticity: Is There Reason for an Opti-
NASS, R., H. PETERSON, and D. KOCH, 1989. Differential effects mistic View'? Proceedings of the Santa Fe Institute Studies in the
of congenital left and right brain injury on intelligence. Sciences of Complexity, Vol. 23, B. Julesz and I. Kovacs, eds.
Brain Cogn. 9:258-266. Reading, Mass.: Addison-Wesley, pp. 217-237.
306 LANGUAGE
STILESJ., E. BATES, D. THAL, D. TRAUNER, andj. REILLY, 1998. early language learning in children with congenital brain
Linguistic, cognitive and affective development in children injury. Cortex 36:31-46.
with pre- and perinatal focal brain injury: A ten-year over- WEBSTER, M. J., J. BACHEVALIER, and L. G. UNGERLEIDER,
view from the San Diego Longitudinal Project. In Advances 1995. Development and plasticity of visual memory circuits.
in Infancy Research. C. Rovee-Collier, L. Lipsitt, and H. In Maturational Windows and Adult Cortical Plasticity: Is There
Hayne, eds. Norwood, N.J.: Ablex, pp. 131-163. Reason for an Optimistic View ? Proceedings of the Santa Fe Insti-
THAL, D., V. MARCHMAN, J. STILES, D. ARAM, D. TRAUNER, R. tute Studies in the Sciences of Complexity, Vol. 23, B. Julesz and
NASS, and E. BATES, 1991. Early lexical development in chil- I. Kovacs, eds. Reading, Mass.: Addison-Wesley, pp. 73-86.
dren with focal brain injury. Brain Lang. 40:491-527. WOODS, B. T, 1980. The restricted effects of right-hemisphere
VARGHA-KHADEM, R, L. CARR, E. ISAACS, E. BRETT, C. ADAMS, lesions after age one: Wechsler test data. Neuropsychologia
and M. MISHKIN, 1997. Onset of speech after left hemi- 18:65-70.
spherectomy in a nine-year-old boy. Brain 120:59-182. WOODS, B. T, 1983. Is the left hemisphere specialized for lan-
VARGHA-KHADEM, R, E. ISAACS, S. VAN DER WERE, S. ROBB, and guage at birth? Trends Neurosci. 6:115—117.
J. WILSON, 1992. Development of intelligence and memory WOODS, B. T, and S. CAREY, 1979. Language deficits after ap-
in children with hemiplegic cerebral palsy: The deleterious parent clinical recovery from childhood aphasia. Ann. Neu-
consequences of early seizures. Brain 115:315-329. rol. 6:405-409.
VARGHA-KHADEM, R, E. ISAACS, and V. MUTER, 1994. A review WOODS, B. T, and H. L. TEUBER, 1973. Early onset of comple-
of cognitive outcome after unilateral lesions sustained dur- mentary specialisation of the cerebral hemispheres in man.
ing childhood./. Child Neurol. 9(suppl.):2S67-2S73. Trans. Am. Neurol. Assoc. 98:113-117.
VARGHA-KHADEM, R, A. O'GORMAN, and G. WATTERS, 1985. WOODS, B. T, and H. L. TEUBER, 1978. Changing patterns of
Aphasia and handedness in relation to hemispheric side, childhood aphasia. Ann. Neurol. 3:272-280.
age at injury and severity of cerebral lesion during child- XIONG, J., S. RAO, J.-H. GAO, M. WOLDORFF, and P. T. Fox,
hood. Brain 108:677-696. 1998. Evaluation of hemispheric dominance for language
VICARI, S., A. ALBERTONI, A. CHILOSI, P. CIPRIANI, G. CIONI, using functional MRI: A comparison with positron emission
and E. BATES, 2000. Plasticity and reorganization during tomography. Hum. Brain Mapp. 6:42-58.
BATES AND ROE: LANGUAGE DEVELOPMENT IN CHILDREN WITH BRAIN INJURY 307
This page intentionally left blank
21 Experience-Dependent Plasticity
and the Treatment of Children
with Specific Language Impairment
or Dyslexia
GAIL C. BEDI
ABSTRACT The temporal processing theory of speech and theory of speech and language and research suggesting
language processing suggests that accurate perception of that both specific language impairment and dyslexia
rapid acoustic cues (auditory temporal processing) is critical are characterized by impaired temporal processing.
to speech perception. It has been hypothesized that both spe-
cific language impairment (SLI) and dyslexia are character- This review is followed by a discussion of the neural
ized by impaired auditory temporal processing. Training basis of experience-dependent plasticity and experience-
programs that use methods aimed at remediating these de- dependent plasticity of speech perception, including a
velopmental disorders by promoting experience-dependent summary of principles drawn from this work and their
neural plasticity have been effective in developing auditory application to training programs for children with spe-
temporal processing and have resulted in improved auditory
processing and language comprehension. In these programs,
cific language impairment or dyslexia. Finally, the im-
neural plasticity was facilitated by training until a behavioral plications of this research and direction for future
asymptote or criterion was achieved, as well as by maintaining research are addressed.
attention to training. Application of these training principles
holds promise for increasing the effectiveness of cognitive re-
mediation and rehabilitation of both children and adults. Temporal processing theory of speech
Recently, research findings in neuroscience, psychol- A considerable body of research has examined the the-
ogy, and speech and language have been integrated to ory that the brain relies upon temporal cues to code
develop new treatments for developmental language or process speech (for review, see Kraus, McGee, and
and language-based disorders. The fact that speech and Koch, 1998; Steinschneider, Arezzo, and Vaughan,
language processing is dependent upon the ability to 1990). According to temporal processing theory,
perceive rapid acoustic cues (auditory temporal pro- speech perception is dependent upon accurate percep-
cessing) and that many children with language-based tion of very short duration acoustic cues and/or the
disorders exhibit impaired auditory temporal process- perception of rapidly sequenced acoustic stimuli (Tal-
ing has been extensively studied. Within this context, lal, Miller, and Fitch, 1993). Certain sounds in English
training programs have been developed that are aimed incorporate rapidly changing or rapidly occurring
at improving auditory temporal processing and thus re- acoustic cues. For example, during the production of
mediating language-based disorders of development. stop consonants (b, d, g, p, t, k) a rapid rise and/or fall
What makes these programs unique is their use of meth- in frequency occurs within approximately 40 ms. Pre-
ods that neuroscience research has shown to be effec- cisely timed transitional elements characterize not only
tive in inducing experience-related changes in the certain consonant sounds but also voice onset time and
brain. This chapter reviews the temporal processing place of articulation. It has been further theorized that
the hemisphere dominant for processing and produc-
tion of speech and language (most commonly, the left
GAIL c. BEDI Director, Manhattan Neuropsychology; Assis- hemisphere) is specialized for processing these rapid
tant Clinical Professor, The Mount Sinai School of Medicine, acoustic cues, not speech and language per se (Nuss-
New York, New York. baum et al., 1999; Tallal, Miller, and Fitch, 1993).
310 LANGUAGE
(r = .81) between errors on the auditory temporal pro- would interfere with the development of a clear neural
cessing tests and errors in reading pseudowords. representation of each phoneme of the child's native
Some findings to the contrary notwithstanding (Jo- language, making speech discrimination problematic.
hannes et al., 1996; Mody, Studdert-Kennedy, and Then poor speech discrimination would compromise
Brady, 1997; Smith, Early, and Grogan, 1986), impaired the development of language and phonological aware-
temporal processing has been well documented in chil- ness, which would, in turn, impede progress in learning
dren with dyslexia (Bedi and Halperin, 1995; Eden et to read phonetically. Reading impairment may indeed
al., 1995; Lovegrove, Martin, and Slaghuis, 1986; Reed, develop due to impaired phonological awareness skills
1989; Slaghuis and Lovegrove, 1985; Tallal 1980b; (Liberman et al., 1974; Stanovich, 1986), but impaired
Wright et al., 1997). Auditory temporal processing def- phonological awareness itself may develop due to im-
icits in stimulus individuation (Wright et al., 1997), dis- paired speech discrimination subsequent to impaired
crimination (Elliott et al., 1990; Reed, 1989), and auditory temporal processing.
temporal order judgment of nonverbal (Bedi and Hal-
perin, 1995) and verbal stimuli (Reed, 1989) have been Experience-driven or training-induced plasticity
documented.
In summary, there is empirical evidence that children Training is considered effective if its implementation
with SLI and children with dyslexia have many charac- results in improvement or decrement of a target behav-
teristics in common. Although dyslexia is denned by im- ior. It is assumed that the behavior changes measured
pairment in reading, children with SLI are at very high following training are the result of experience-driven
risk for reading impairment. In both children with SLI brain plasticity or changes in brain chemistry, mor-
and children with dyslexia (without a previous diagnosis phology, and/or physiology that correlate with the spe-
of SLI), a positive relationship between reading impair- cific training or experience.
ment and impaired phonological awareness skills has Indeed, within the audiovocal and other sensory sys-
been found. Most children with SLI and many children tems, experience-induced plasticity has been demon-
with dyslexia are impaired in their ability to process rap- strated in human and nonhuman primates through
idly presented auditory stimuli (Bedi and Halperin, measurement of changes in topographical representa-
1995; Reed, 1989; Tallal, 1980b). Both the receptive lan- tion of cortical functions (Recanzone, Merzenich, and
guage of children with SLI and the ability of children Schreiner, 1992; Recanzone, Schreiner, and Merzenich,
with dyslexia to perform phonological awareness tasks 1993). More specifically, improved auditory discrimi-
are highly correlated with deficits in discriminating nation, tactile discrimination, and performance of mo-
stimuli that incorporate rapidly changing acoustic cues tor tasks have been shown to be correlated with changes
(Tallal, 1980b). in neural representation in sensory relevant or related
motor areas (Cansino and Williamson, 1997; Kami et
Does a common mechanism underlie impaired al., 1998; Recanzone, Merzenich, and Schreiner, 1992;
phonological awareness and reading impairment'? Recanzone, Schreiner, and Merzenich, 1993; Xerri et
al., 1996).
SLI and dyslexia may have a cognitive process in com- It is theorized that a distributed series of Hebb-like
mon, which could, if impaired, affect phonological synapses and dynamic cortical cell assemblies underlie
awareness, language, and reading. It has been theorized experience-induced plasticity (Merzenich and Same-
that the ability to discriminate between speech sounds shima, 1993; Rose, 1992a; Rosenzweig and Bennett,
is critical to the ability to develop sound-symbol rela- 1996; Rosenzweig et al., 1992). Hebb (1949) hypothe-
tionships (Goswami, 1990; Tallal, 1980b). SLI and dys- sized that an association could be formed between neu-
lexia have been shown to be related to a deficit in rons such that a neuron induced to discharge following
discriminating rapidly presented acoustic information, specific neuronal input would have increased synaptic
specifically within the time window essential for pro- effectiveness if this association was repeated succes-
cessing brief, rapidly changing components of speech sively. Repeated or successive coincidental input and
(Reed, 1989; Tallal 1980b), and herein may lie the com- discharge between the two neurons would result in met-
mon mechanism. Impaired phonological awareness abolic and structural changes in one or both neurons.
may be related to a neural timing deficit which con- These changes would account for the neuron's higher
strains the processing of stimuli that incorporate short- likelihood of discharge to future inputs.
duration, rapidly sequenced, or rapidly changing During training, neurons that fire in response to spe-
acoustic components (Stark and Tallal, 1988). Accord- cific environmental stimuli will become more respon-
ing to this hypothesis, a temporal processing deficit sive with repeated exposure to the same environmental
312 LANGUAGE
ing effect quickly saturates and is followed by the slow and changes in central auditory cortex neurophysiology
incremental emergence of a large performance gain as measured by changes in MMN evoked potentials.
(Kami et al., 1998). This larger performance gain is Specifically, following speech perception training, an
triggered by continued training (Kami et al., 1998). increased magnitude of MMN responses was demon-
However, it is important to note that overtraining can strated (Kraus et al., 1995a; Tremblay et al., 1997; Trem-
result in a reduction of task-specific responding in the blay, Kraus, and McGee, 1998). Furthermore, this
cortex (Aizawa et al., 1990). Merzenich and Sameshima change in neurophysiology preceded or occurred co-
(1993) suggested that this reduction in task-specific cor- incidentally with behavioral changes (Tremblay, Kraus,
tical response is a function of decreased attention sec- and McGee, 1998). These results indicated that the
ondary to training that was conducted well past task training changed the neural representation of speech
mastery. Thus, training until a task has been mastered sounds and that behavioral changes followed these neu-
or until change in behavior has reached a plateau seems rophysiological changes. Both the behavioral and neu-
to be an important feature of training aimed at induc- rophysiological changes associated with speech sound
ing experience-driven neural plasticity. However, train- discrimination training generalized to novel or un-
ing well past task acquisition may negatively affect trained stimuli (Tremblay et al., 1997).
experience-driven neural plasticity, most likely because Native speakers of Japanese cannot discriminate be-
the experience becomes routine or boring and, in tween the /r/ and /!/ sounds that occur in American
consequence, results in decreased attention during English. Instead, they identify both sounds as the /r/
training. sound in Japanese. Using an adaptive training proce-
dure based upon a Hebbian experience-induced model
Experience-dependent plasticity following speech of plasticity, native speakers of Japanese were trained to
perception training discriminate between the /r/ and /!/ phonemes that
occur in American English but not in Japanese (Mc-
Speech discrimination training has been shown to im- Candliss et al., 1999). Participants were presented with
prove speech perception and has been shown, using words that began with one of these phonemes (e.g.,
mismatched negativity (MMN), to alter the physiology load, road) and were required to respond whether the
of central auditory cortex. Mismatched negativity is a word began with an /r/ or /1/. No feedback about re-
passively elicited evoked auditory response that is re- sponse accuracy was provided. Two conditions were
corded when a deviant or "oddball" stimulus is pre-
studied: One group heard natural speech samples; the
sented within the context of several hundred identical
other group was trained using a continuum of stimuli
(standard) stimuli (for review, see Naatanen, 1995). A
that progressed from a signal that was modified to en-
"difference wave" is obtained by subtracting the re-
hance discrimination to natural speech. This second
sponse elicited by the standard stimuli from the re-
type of training involved an adaptive training approach.
sponse elicited by the deviant stimuli (Naatanen, 1995).
The difference wave shows a negativity at the 100-200 Specifically, after eight consecutive correct responses,
ms latency range that is dependent upon the magnitude stimuli from the next step along the continuum were
of the deviation between the standard and oddball stim- presented. After three 20-minute training sessions,
uli (Naatanen, 1995). It reflects specific processes in the significant improvement in the ability to discriminate
primary and association auditory cortex (Naatanen, between /r/ and /l/ was seen in the training group that
Gaillard, and Mantysalo, 1978) and preattentive dis- heard the modified speech but not in the natural
crimination of acoustic input (Naatanen et al., 1978; speech group.
Tremblay et al., 1997). Thus, it is a neurophysiological In summary, adaptive training modeled upon the
response that indicates auditory discrimination without principles of Hebbian experience-driven plasticity re-
the participant's having to pay attention. sults in improved speech discrimination in normal
Normal adults were trained to discriminate two adults. Speech discrimination training in normal adults
speech contrasts which, prior to training, would not results in experience-driven plasticity as demonstrated
have been recognized as distinct from each other or by changes in neurophysiology that correlate with train-
would infrequently have been recognized as being dis- ing. Next, research investigating the effectiveness of
tinct from each other (Kraus et al., 1995a; Tremblay et training that incorporates features aimed at facilitating
al., 1997; Tremblay, Kraus, and McGee, 1998). In this experience-driven plasticity to train speech perception
series of studies, training resulted in significant im- deficits associated with the developmental disorders of
provement in discriminating between the two stimuli SLI and dyslexia will be reviewed.
314 LANGUAGE
pair. Each of these three types of syllable pairs was was conducted to assess phonemic segmentation. All of
trained using an eight-step progression from long (400 the children with dyslexia who had received discrimi-
ms) to short (10 ms) ISIs nation training showed significant improvement in
This program also reflected training aimed at pro- phonemic segmentation skills while the performance of
moting experience-induced plasticity in that behavior the untrained children with dyslexia was not signifi-
was trained until a criterion was achieved and reinforce- cantly different from their pretraining levels.
ment was employed to maintain attention. For each
child, training was conducted until each child achieved Discussion
eight consecutive correct responses at the shortest ISI
for each syllable type. The training was individually The results from the training programs for children
adaptive such that each child started training for a par- with SLI and dyslexia suggest that, once an underlying
ticular syllable type at the longest ISI. He or she was cognitive deficit for a disorder has been determined,
required to achieve a criterion of eight consecutive cor- training that incorporates characteristics important to
rect responses before the next shorter ISI was pre- promoting experience-induced plasticity can effectively
sented. If this criterion was not met within 24 trials, remediate the disorder. Important training features for
syllable pairs with an ISI halfway between the last ISI facilitating experience-driven plasticity are (1) intense
passed and the failed ISI were presented. This iterative training that is maintained until a mastery criterion or
procedure was applied in an ascending or descending behavior asymptote is achieved and (2) maintaining at-
sequence until the initially failed ISI was passed or stim- tention during task performance.
uli with an ISI within 5 ms of the initially failed ISI were It is presumed that these training programs obtained
passed. The children required 3-4 days of 30-45- their results because the training produced neural re-
minute sessions to complete the training sequence. To organization. However, it is important to document that
maintain attention, immediate feedback was provided this training does indeed promote reorganization of the
and correct responses were rewarded; the child was pre- brain or experience-induced plasticity. Although fMRI
sented with a smiling face for each correct response. and PET have been used to assess cortical activation
After determining that a group of second and fourth during cognitive tasks and could presumably be used to
graders with dyslexia were indeed impaired in making assess change in cortical activation as a function of train-
same/different judgments for pairs of stop consonant- ing, there can be limitations to employing these two
vowel syllables relative to grade-matched average read- methods with children as participants (but see Casey,
ers, Hurford (1990) randomly assigned the children Thomas, and McCandliss, this volume). PET is too in-
with dyslexia to one of two treatment groups. One vasive to employ with children. Functional MRI has
group received the discrimination training just de- been successfully employed to study cortical activation
scribed while the other group received "training" that in children 6 years of age and older. However, it can be
was not aimed at remediating discrimination. Following difficult to keep children quiet/still during the length
training, discrimination for pairs of syllables was reas- of time needed to conduct many fMRI procedures.
sessed using novel pairs of stop consonant-vowel sylla- Thus, imaging artifacts that are secondary to movement
bles. The performance of the children with dyslexia may complicate data interpretation. Therefore, consid-
who received discrimination training was significantly eration should be given to also employing alternative
improved, to the point that they were no longer signifi- methods to determine whether change in neurophysi-
cantly different from average readers. However, the per- ology occurs following these specific training programs
formance of the children with dyslexia who did not for SLI and dyslexia. The Mismatched Negativity tech-
receive discrimination training did not differ signifi- nique appears to be a viable option (Kraus et al.,
cantly from that at pretraining levels, remaining signifi- 1995a,b; Naatanen, 1995).
cantly below that of the average readers. Discussions of neural plasticity encompass the com-
In another treatment study, Hurford and Sanders monly held view that a young nervous system is more
(1990) identified a group of second- and fourth-grade capable of recovery of function following brain damage
children with dyslexia who had deficits in phonemic than is an older nervous system (Lenneberg, 1967).
segmentation relative to grade-matched average read- This view is supported by findings of normal language
ers. These children were then randomly divided into development in prepubescent children following focal,
two groups, with one group receiving the same discrim- unilateral brain injury unaccompanied by seizure dis-
ination training that was employed by Hurford (1990) order (Dall'Oglio et al., 1994; Muter, Taylor, and
and the other group receiving the training control in- Vargha-Khadem, 1997), or following hemispherectomy
tervention. Upon completion of training, post-testing (Vargha-Khadem et al., 1991). In contrast, adults with
316 LANGUAGE
BRADLEY, L., and P. BRYANT, 1983. Categorizing sounds and LEVI, G., F. CAPOZZI, A. FABRIZI, and E. SECHI, 1982. Language
learning to read: A causal connection. Nature 301:419-421. disorders and prognosis for reading disabilities in develop-
BRYANT, P. E., and L. BRADLEY, 1983. Auditory organization mental age. Percept. Motor Skills 54:1119-1122.
and backwardness in reading. In Developmental Neuropsychi- LIBERMAN, I. Y, D. SHANKWEILER, F. W. FISCHER, and B. CA-
atry, M. Rutter, ed. New York: Guilford Press, pp. 489-497. TER, 1974. Explicit syllable and phoneme segmentation in
CALFEE, R. C., P. LINDAMOOD, and C. LINDAMOOD, 1973. the young child.J Exp. Child Psychol. 18:201-212.
Acoustic-phonetic skill and reading, kindergarten through LOVEGROVE, W.J., R. MARTIN, and W. SLAGHUIS, 1986. A theo-
12th grade.J. Ed. Psychol. 64:293-298. retical and experimental case for a visual deficit in specific
CANSINO, S., and S. J. WILLIAMSON, 1997. Neuromagnetic reading disability. Cogn. Neuropsychol. 3(2):225-267.
fields reveal cortical plasticity when learning an auditory MANN, V. A., 1991. Phonological abilities: Effective predictors
discrimination task. Brain Res. 764:53-66. of future reading ability. In Learning to Read, L. Rieben and
CATTS, H. W., 1993. The relationship between speech- C. A. Perfetti, eds. Hillsdale, N.J.: Lawrence Erlbaum, pp.
language impairments and reading disabilities. /. Speech 121-133.
Hear. Res. 36:948-958. MCCANDLISS, B. D., J. A. FIEZ, M. CONWAY, and J. L. MC-
DALL'OGLIO, A. M., E. BATES, V. VOLTERRA, M. DICAPUA, and CLELLAND, 1999. Eliciting adult plasticity for Japanese
G. PEZZINI, 1994. Early cognition, communication and lan- adults struggling to identify English /r/ and /l/: Insights
guage in children with focal brain injury. Dev. Med. Child from a Hebbian model and a new training procedure.
Neural. 36 (12): 1076-1098. Cogn. Neurosci. Soc. Ann. Meeting Prog.,J Cogn. Neurosd.
EDEN, G. E, H. M. STEIN, H. M. WOOD, and F. B. WOOD, 1995. (suppl.)
Temporal and spatial processing in reading-disabled and MERZENICH, M. M., W. M. JENKINS, P. JOHNSTON, C. SCHREI-
normal children. Cortex 31:451-468. NER, S. L. MILLER, and P. TALLAL, 1996. Temporal process-
ELLIOTT, L. L., M. E. SCHOLL, J. O. GRANT, and M. A. HAM- ing deficits of language-learning impaired children
MER, 1990. Perception of gated, highly familiar spoken ameliorated by training. Science 271:77-81.
monosyllabic nouns by children with and without learning MERZENICH, M. M., and K. SAMESHIMA, 1993. Cortical plastic-
disabilities.J Learn. Dis. 23(4):249-252, 259. ity and memory. Cogn. Neurosci. 3:187-196.
GOSWAMI, U., 1990. A special link between rhyming skill and MODY, M., M. STUDDERT-KENNEDY, and S. BRADY, 1997.
the use of orthographic analogies by beginning readers. J Speech perception deficits in poor readers: Auditory pro-
Child Psychol. Psychiatry 31:301-311. cessing or phonological coding? J Exp. - Child Psychol.
HEBB, D. O., 1949. The Organization of Behavior: A Neuropsycho- 64(2):199-231.
logical Theory. New York: Wiley. MUTER, V., S. TAYLOR, and F. VARGHA-KHADEM, 1997. A lon-
HURFORD, D. P., 1990. Training phonemic segmentation abil- gitudinal study of early intellectual development in hemi-
ity with a phonemic discrimination intervention in second- plegic children. Neuropsychologia 35(3):289-298.
and third-grade children with reading disabilities. J Learn. NAATANEN, R., 1995. The mismatch negativity: A powerful tool
Dis. 23(9):564-569. for cognitive neuroscience. Ear Hearing 16(1) :6-18.
HURFORD, D. P., and R. F. SANDERS, 1990. Assessment and re- NAATANEN, R., A. GAILLARD, and S. MANTYSALOA, 1978. Early
mediation of a phonemic discrimination deficit in reading- selective attention effect on evoked potential reinterpreted.
disabled second and fourth graders. J Exp. Child Psychol. Acta. Psychol. 42:313-329.
50:396-415. NUSSBAUM, A. O., C. Y. TANG, G. C. BEDI, E.S. DORSETT,
JOHANNES, S., C. L. KUSSMAUL, T. F. MUNTE, and G. R. MAN- M. BUCHSBAUM, and S. W. ATLAS, 1998. Functional MRI
GUN, 1996. Developmental dyslexia: Passive visual stimula- during rapid auditory processing in normal adults: Corre-
tion provides no evidence for a magnocellular processing lation to task performance and implication for the study of
deficit. Neuropsychologia. 34(11):1123-1127. dyslexia. Proceedings of the 84th Sci. Assembly Ann. Meeting Ra-
KARNI, A., 1996. The acquisition of perceptual and motor diol. Soc. N. Amer.
skills: A memory system in the adult human cortex. Cogn. RECANZONE, G. H., M. M. MERZENICH, M and C. E. SCHREINER,
Brain Res. 5:39-48. 1992. Changes in the distributed temporal response prop-
KARNI, A., G. MEYER, C. REY-HIPOLITO, P. JEZZARD, M. M. erties of SI cortical neurons reflect improvements in per-
ADAMS, R. TURNER, and L. G. UNGERLEIDER, 1998. The ac- formance of a temporally based tactile discrimination task.
quisition of skilled motor performance: Fast and slow / Neurophysiol 67(5):1071-1091.
experience-driven changes in primary motor cortex. Proc. RECANZONE, G. H., C. E. SCHREINER, and M. M. MERZENICH,
Natl Acad. Sci. USA 95:861-868. 1993. Plasticity in the frequency representation of primary
KRAUS, N., T. McGEE, T. D. CARRELL, C. KING, K. TREMBLAY, auditory cortex following discrimination training in adult
and T. NICOL, 1995a. Central auditory system plasticity as- owl monkeys.J Neurosci. 13(1):87-103.
sociated with speech discrimination training.J Cogn. Neu- REED, M. A., 1989. Speech perception and the discrimination
rosci. 7(l):25-32. of brief auditory cues in reading-disabled children. J Exp.
KRAUS, H., T. McGEE, T. D. CARRELL, and A. SHARMA, 1995b. Child Psychol. 48:270-292.
Neurophysiologic bases of speech discrimination. EarHear- ROBIN, D. A., D. TRANEL, and H. DAMASIO, 1990. Auditory
mg!6(l):19-37. perception of temporal and spectral events in patients with
KRAUS, N., T. J. McGEE, and D. B. KOCH, 1998. Speech sound focal left and right cerebral lesions. Brain Lang. 39:539-555.
representation, perception and plasticity: A neurophysio- ROSE, S. P. R., 1992a. Synaptic plasticity, learning and memory.
logic perspective. Audiol. Neurootol. 3:168-182. In Synaptic Plasticity: Molecular, Cellular and Functional Aspects,
LENNEBERG, E. H., 1967. Biological Foundations of Language. M. Baudry, R. F. Thompson, and J. L. Davis, eds. Cambridge,
New York: John Wiley and Sons. Mass.: MIT Press, pp. 209-229.
318 LANGUAGE
VI
COGNITION
This page intentionally left blank
22 Attention in Young Infants:
A Developmental
Psychophysiological Perspective
JOHN E. RICHARDS
ABSTRACT This chapter reviews the development of atten- view is limited to the use of heart rate as an index of
tion in young infants, emphasizing heart rate changes in psy- the development of sustained attention, which is a gen-
chophysiological experiments as a measure of an arousal brain eral arousal system affecting a wide number of behav-
system. The neural systems affecting attention that may be in-
dexed by psychophysiological measures are briefly reviewed. ioral and cognitive functions controlled by the brain.
Heart rate, electroencephalogram (EEG), event-related po- These experiments are related to changes occurring in
tentials (ERP), and other physiological measures are reviewed the neural systems underlying attention.
that have been used for the study of attention development
in young infants. The developmental changes in infant atten-
tion are related to changes occurring in the neural systems Brain systems involved in attention
underlying attention. Several studies are reviewed that show
how heart rate may be used as a measure of a general arousal AROUSAL ATTENTION SYSTEM One emphasis in the
system in young infants. cognitive neuroscience of attention has been on the
arousal associated with energized cognitive activity (Pos-
Attention, generally defined, shows dramatic devel- ner, 1995). The arousal emphasis has focused upon the
opment over the period of infancy. At birth infants at- increased behavioral performance that occurs when at-
tend primarily to salient physical characteristics of their tention is engaged. This increased behavioral perfor-
environment or attend with nonspecific orienting mance is associated with shortening of reaction times
(Berg and Richards, 1997). Between birth and two years in detection tasks, increased focus of performance on
the development of alert, vigilant sustained attention specific tasks, and the sustaining of performance over
occurs. At the end of the first two years, infants' exec- extended periods of time. The arousal emphasis is non-
utive attention system is beginning to function (Ruff specific, affecting multiple modalities, cognitive sys-
and Rothbart, 1996; also see Rothbart and Posner, this tems, and cognitive processes. This arousal emphasis
volume). These dramatic changes in infants are com- characterizes attention's energizing effect on cognitive
monly thought to be based predominantly on age- and behavioral performance. Attention also may have
related changes in brain structures responsible for
a selective effect on specific cognitive processes or be-
attention control.
havior without arousal properties. In fact, selective at-
This chapter has three objectives. First, brain systems
tention may serve in some situations to inhibit behavior
that may be involved in attention and show develop-
if such inhibition is appropriate for the goal of the task.
ment in infancy are reviewed. These systems include a
Specific locations or systems in the brain control the
general arousal system that affects many cognitive func-
arousal aspect of attention. The brain systems underly-
tions as well as specific attention systems that are limited
ing the arousal aspect of attention have been detailed
in their effect on cognition and attention. Second, psy-
in the theoretical and empirical research literature for
chophysiological measures that have been useful in the
a number of years. An example of this arousal emphasis
study of brain-attention relations in infants are pre-
is a model of neuroanatomical connections between the
sented. The use of heart rate as a measure of the general
arousal system is emphasized. Finally, several studies are mesencephalic reticular activating system and the cor-
examined that used these psychophysiological methods tex (Heilman et al., 1987; Mesulam, 1983). This model
to study the development of infant attention. This re- (diagrammed in figure 22.1) presumes that there are
centers broadly scattered throughout the mesence-
JOHN E. RICHARDS Department of Psychology, University of phalic reticular activating system that are activated
South Carolina, Columbia, South Carolina. by sensory stimulation. In turn, the mesencephalic
reticular activating system directly influences the limbic action times, better detection, and sustaining of cogni-
system, thalamus, and cortex. The cingulate cortex tive performance for extended periods of time.
receives information from areas of the limbic system Another perspective on the arousal aspect of atten-
such as the basolateral nucleus of the amygdala and the tion is a model based on the neurochemical systems
subicular portion of the hippocampus. The cingulate involved in arousal. Robbins and Everitt (1995) distin-
cortex is a major afferent relay center that projects to guish four neurochemical systems that form the basis
parietal area PG, visual association cortical areas, and for the arousal functions of attention: noradrenergic,
other cerebral cortex centers involved in complex cog- cholinergic, dopaminergic, and serotoninergic. Figure
nitive functions. This neuroanatomical system acts in 22.2 (see color plate 14) shows the projections from
synchrony to "energize" primary sensory areas in the midbrain nuclei for these four brain systems. The nuclei
cortex and increase the efficiency of responding in that give rise to these four neurochemical systems are
those areas. This system also influences association ar- located in brain regions adjacent to the mesencephalic
eas and other attention systems, such as the posterior reticular activating system. Robbins and Everitt (1995)
attention system described by Posner (Posner, 1995; Pos- review the evidence linking these neurochemical pro-
ner and Petersen, 1990). The nonspecificity of this sys- jection systems to attention and arousal. The noradren-
tem is implied by its interconnections with multiple ergic and cholinergic systems are thought to be the
areas that influence cognitive processing. This arousal neurochemical systems that are most closely involved in
system "invigorates" or "energizes" cognitive processes, cortical arousal as it is related to attention. The dopa-
leading to increased processing efficiency, shorter re- minergic system affects the motivational and energetic
322 COGNITION
FIGURE 22.2 The neurochemical systems involved in atten- nucleus of the vagus; H, hypothalamus; LC, locus ceruleus; C,
tion and arousal. Abbreviations: III, oculomotor nucleus; T, caudate nucleus; P, putamen; S, septal nuclei; V, ventral stria-
thalamus; HC, hippocampal formation; RF, reticular forma- turn. (Reprinted with permission from Nolle, J., and J. B. An-
tion; PSG, parasympathetic ganglion cell; X, dorsal motor gevine, 1995. The Human Brain. St. Louis, MO: Mosby.)
aspects of cognitive processing and the serotonin system tions. Therefore, these systems have only a narrow and
affects the overall control of state. These four neuro- selective impact on attention-based cognitive function-
chemical systems are closely linked so that more than ing. Two of these are worth mentioning in this respect.
one is likely to be operating during an aroused state. The enhancement of visual receptive fields during at-
tention to visual stimuli has been widely studied in
SELECTIVE ATTENTION The second manner in which invasive preparations (Desimone and Duncan, 1995;
the brain affects the development of attention in infants Maunsell and Ferrera, 1995). This type of attention is
is found in brain systems specific to selected functions. selective for particular objects, particular spatial loca-
These brain areas show enhanced functioning under tions, or particular tasks. For example, the responses of
attention but affect only a single (or few) cognitive func- visual receptive fields are enhanced in tasks requiring
324 COGNITION
tion termination" phase. These periods are simply the The phases of sustained attention and attention ter-
period of time before the presentation of the stimulus mination are markers of the nonspecific arousal system
(preattention) and before heart rate returns to its pres- of the brain (Richards and Casey, 1992; Richards and
timulus level but after sustained attention has occurred Hunter, 1998). This nonspecific arousal system sustains
(preattention termination). attention and maintains a vigilant state. The heart rate
Sustained attention and attention termination affect changes occurring during sustained attention (sus-
a wide range of cognitive functions in infants. The heart tained heart rate slowing) index the onset and contin-
rate slows down and remains below prestimulus levels uing presence of this arousal. The heart rate changes
during sustained attention. Cognitively, this phase of at- during attention termination (return of heart rate to
tention involves subject-controlled processing of stim- its prestimulus level) index the lack of activation of this
ulus information. Sustained attention is accompanied arousal system. These two phases of attention therefore
behaviorally by maintaining fixation on a focal stimulus reflect the nonspecific arousal that may affect a number
in the presence of a peripheral distracting stimulus of sensory and brain systems. Incidentally, these phases
(Hicks and Richards, 1998; Hunter and Richards, 1997; and the "automatic interrupt" and "stimulus orienting"
Lansink and Richards, 1997; Richards, 1987, 1997a), attention phases also may be used to measure specific
acquiring stimulus information (Richards, 1997b) and attentional systems in the young infant (e.g., Balaban,
exhibiting recognition memory (Richards and Casey, 1996; Berg and Richards, 1997; Richards, 1998, 2000a).
1990), and enhancement of responses in a selected The heart rate changes occurring during attention
stimulus modality and inhibition of responses in a non- and their indexing of arousal brain systems are impor-
selected stimulus modality (Richards, 1998, 2000a). Al- tant for developmental cognitive neuroscience because
ternatively, at the end of sustained attention the heart these heart rate changes show important developmen-
rate returns to its prestimulus level and the phase of tal changes in the first six months of infancy. A consis-
attention termination occurs. Attention termination is tent pattern of developmental changes has been shown
accompanied by inattentiveness toward the stimulus in to occur in the sustained attention phase. The level of
the presence of continued fixation on the stimulus, i.e., the heart rate deceleration during sustained attention
heightened levels of distractibility, lack of acquisition of increases from 14 to 26 weeks of age (3 to 6 months)
(Casey and Richards, 1988; Richards, 1985, 1987,
stimulus information, and lack of selective modality
1989a,b, 1994). The level of heart rate during sustained
effects.
attention is thought to reflect the depth of the arousal.
The arousal system of the brain controls the heart
Thus, the changes in the evoked heart rate response
rate changes that occur during sustained attention. The
during sustained attention imply that the arousal con-
neural control of this heart rate change originates from
trolled by the brain is increasing over this age range.
cardioinhibitory centers in the orbitofrontal cortex.
The age change in heart rate during sustained atten-
This area has reciprocal connections with the limbic tion parallels some of the behavioral manifestations of
system and through these connections is involved in attention. This includes an increasing ability of infants
modulating activity within the mesencephalic reticular to acquire familiarity with stimulus characteristics in a
formation arousal system (Heilman et al, 1987; Mesu- fixed period of time (Frick and Richards, 2001; Rich-
lam, 1983) and probably the dopaminergic and cholin- ards, 1997b), enhanced tracking of moving stimuli
ergic neurotransmitter systems (Robbins and Everitt, (Richards and Holley, 1999), and the selective modality
1995). The cardioinhibitory centers act through the enhancement effect found in selective attention (Rich-
parasympathetic nervous system to slow heart rate when ards, 1998, 2000a). The behavioral and heart rate in-
the arousal system is engaged. This slowing of heart rate dices of attention are not as well synchronized at
occurs as "vagus nerve" (10th cranial nerve) activity in- younger ages (e.g., 8 weeks) as they are at older ages
creases, leading to a slowing of the cardiac pacemaker (Hicks and Richards, 1998; Hunter and Richards, 1999,
firing, increases in interbeat intervals, and heart rate 2000; Richards, 1989b). The age-related changes in
slowing. The "arousal" system does not result in more heart rate during sustained attention, and these age
activity in heart rate, but inhibited activity. Similarly, changes in the tasks corresponding to sustained atten-
some other peripheral physiological processes (e.g., tion, imply that this general arousal system develops in
body movement) are inhibited during conditions of at- the first few months of infancy and that it increasingly
tentive arousal. Thus, the arousal brain system when op- affects infants' cognitive behavior.
erating in this arousal-attention manner selectively
enhances some brain systems and functions while in- OTHER PSYCHOPHYSIOLOGICAL MEASURES There are
hibiting others. other psychophysiological measures that have been
326 COGNITION
of brain function in some contexts. For example, au- velopment of the arousal attention system in young in-
ditory brainstem-evoked potentials are derived from fants modulates this selective attention effect. The
highly filtered, event-averaged EEG. These brainstem- startle reflex is a response to high-intensity short-
evoked potentials are generated at specific points in the duration stimuli, and includes widespread flexor jerk
neural pathway from the peripheral auditory apparatus and whole body startle. One aspect of the startle reflex
through the auditory nerves to the brainstem. They is the startle blink reflex. The blink reflex occurs in
measure functioning of these pathways in a direct sense. response to visual, auditory, tactile, and other stimuli.
Similarly, some cortical ERP measures may be similarly The acoustic startle blink reflex and the visual startle
interpreted using high-density recording (Tucker, 1993; blink reflex are based upon short-latency reflex path-
Tucker et al., 1994), and cortical sources may be hy- ways involving first-order neurons in the sensory path-
pothesized and compared with the scalp distribution of ways and the brainstem cranial nerves that move the
the ERP components (Nunez, 1990; Scherg, 1990; muscles for the blink (Balaban, 1996; Davis, 1997; Hack-
Scherg and Picton, 1991). Functioning of the cortical ley and Boelhouwer, 1997). The blink reflex represents
areas may be inferred from these cortical source local- an "automatic interrupt" system that interrupts ongo-
ization procedures in a direct fashion. The use of the ing information processing and shifts the organism's
ERP and cortical source localization procedures as a goals for other activities (Graham, 1979, 1992).
direct measure in the study of attention is only begin- One characteristic of the blink reflex that has been
ning with infant participants. Such use of the EEG and of interest to cognitive psychophysiologists is its modi-
ERP should lead to a higher quality of information fiability by selective attention. Directing attention to
about the relation between the brain and attention in one stimulus modality enhances the blink reflex to a
infant psychological development. stimulus of that modality and attenuates the blink reflex
In the rest of the chapter we review studies that show to stimuli in other modalities (Anthony, 1991; Anthony
the developmental changes that occur in the arousal and Graham, 1983, 1985; Balaban, Anthony, and Gra-
form of attention. The first section reviews two studies ham, 1989; Hackley and Graham, 1983; Haerich, 1994;
(Richards, 1998, 2000a) that use modification of the Richards, 1998, 2000a). This attentional modulation
blink reflex with selective attention. These studies show shows that a higher-order cognitive process, selective
how a specific attention system interacts with the devel-
attention to a specific modality, may modify reflexes
opment of the general arousal system. Next we review
controlled by simple reflex arcs in the central nervous
a study (Richards and Holley, 1999) showing the effect
system. This attenuation/modification of the blink re-
of the developing arousal system on eye movements that
flex also may be used as an index of the amount of
themselves show development over the first six months
higher-order selective attention. The modality-selective
in infancy. The final section of the reviews presents
effect of attention on the blink reflex has been shown
some studies that show developmental changes in sus-
in young infants (Anthony and Graham, 1983; Balaban,
tained attention that are related to a "higher cognitive
function," infants' recognition of briefly presented vi- Anthony, and Graham, 1989; Richards, 1998, 2000a).
sual stimuli (Frick and Richards, 2001; Richards, 1997b, Two recent studies used heart rate changes associated
2000b; Richards and Casey, 1990). These studies show with the arousal aspect of attention to show develop-
that familiarization of patterns presented for only a few mental changes in the selective modality effect on the
seconds during sustained attention results in recogni- blink reflex (Richards, 1998, 2000a). In these studies,
tion memory (Frick and Richards, 2001; Richards, infants at 8, 14, 20, or 26 weeks of age were presented
1997b). This section also presents some new, as yet un- in the foreground with interesting visual or auditory
published data which show that during attentive states stimuli (Richards, 1998) or a multimodal auditory-
infants will recognize stimuli very quickly, exhibiting visual stimulus (Richards, 2000a). The infant's heart
appropriate EEG and ERP changes associated with rec- rate was recorded and delays were defined according to
ognition memory (Richards, 2000b). These studies heart rate changes associated with phases of attention.
should be considered examples of how developmental For example, "sustained attention" was defined as a sig-
psychophysiology may contribute to developmental nificant heart rate deceleration beyond the prestimulus
cognitive neuroscience of attention. level, and "attention termination" was defined as a re-
turn of heart rate to its prestimulus level following a
Selective attentional modification of blink reflexes period of sustained attention. When it was thought that
sustained attention was engaged, or attention was unen-
Here we review studies showing the effect of selective gaged, an auditory or visual stimulus that elicits a blink
attention on the blink reflex and examine how the de- reflex was presented. The amplitude of the blink reflex
FIGURE 22.4 Blink reflex magnitude (rms,mV)as a function FIGURE 22.5 Blink reflex magnitude (rms, mV) difference
of the sustained attention or attention termination and the from the prestimulus to the sustained attention conditions
match/mismatch between the foreground and the blink stim- and the match/mismatch between the foreground and the
ulus. The solid black bars represent trials on which the fore- blink stimulus, separately for the four testing ages. The solid
ground and the blink reflex stimuli were in the same modality, black bars represent trials on which the foreground and the
and the light bars represent trials on which the foreground blink reflex stimuli were in the same modality, and the light
and blink reflex stimuli were in different modalities. Also bars represent trials on which the foreground and blink reflex
shown is the average blink magnitude on prestimulus trials stimuli were in different modalities. The SE for the presti-
with no foreground stimulus (solid line; SE ranges as dashed mulus trials is shown as a dashed line around the 0 rms mV
line). (Adapted from Richards, 1998.) point. (Adapted from Richards, 1998.)
328 COGNITION
tive cognitive processes (e.g., "automatic interrupt"; see affecting eye movement development, including mod-
Graham, 1979, 1992) and is controlled by subcortical els by Bronson (1974, 1997), Maurer and Lewis (1979,
pathways and brain mechanisms. The spinal motor neu- 1991, 1998), Johnson and colleagues (Johnson, 1990,
rons controlling the blink muscles and the brainstem 1995; Johnson, Gilmore, and Csibra, 1991; Johnson,
afferent pathways involved in the blink reflexes are rela- Posner, and Rothbart, 1998), Hood (Hood, 1995;
tively mature at birth (Balaban, 1996). Thus, the reflex Hood, Atkinson, and Braddick, 1998), and Richards
itself shows little developmental change over the testing (Richards and Casey, 1992; Richards and Hunter, 1998).
ages such as those used in this study. The development A model proposed by Johnson (1990, 1995; Johnson,
of sustained attention over this age range thus influ- Gilmore, and Csibra, 1991; Johnson, Posner, and Roth-
ences the extent to which selective attention will affect bart, 1998) describes the developmental changes in
this low-level cognitive process. Over this age range these three eye movement systems. This model hypoth-
there is an increasing influence of the general arousal esizes that layers of the primary visual area develop at
system over this very specific attention system, both con- different rates and become mature at different ages.
trolled by brain processes. The primary visual area layers containing brain path-
ways that control reflexive eye movement are relatively
Visual smooth pursuit eye movements mature at birth, hence reflexive .saccadic eye move-
and attention ments dominate the infant's behavior in the first 2 post-
natal months. The primary visual area layers that
This section reviews the relation between the develop- contain brain pathways which control voluntary sac-
ment of the arousal attention system in young infants cadic eye movements develop rapidly from the first to
and three eye movement control systems that show de- the sixth postnatal months. In conjunction with this de-
velopment in the same period of time. There are three velopment, attention-directed voluntary saccades show
types of eye movements that may be made when track- developmental changes over the first six months. Fi-
ing visual stimuli. Each eye movement type is controlled nally, the primary visual area layers that contain brain
by separate areas of the brain. Reflexive saccadic eye pathways which control smooth pursuit eye movements
movements occur in response to the sudden onset of a develop more slowly than the other layers. Several parts
peripheral stimulus. These eye movements are con- of the brain pathways that control smooth pursuit eye
trolled by a brain pathway involving the retina, lateral movements show protracted developmental changes
geniculate nucleus, superior colliculus, and perhaps the over the first two years (Richards and Hunter, 1998).
primary visual area (Schiller, 1985, 1998). Voluntary sac-
Thus, smooth pursuit eye movements are the latest to
cadic eye movements occur under voluntary or planned
begin development and show changes over a longer pe-
control. These eye movements often involve attention-
riod than just the first six months of infancy. Figure 22.6
directed targeted eye movements. The voluntary sac-
(Richards and Hunter, 1998) shows a hypothetical de-
cadic eye movements are controlled by a brain pathway
involving several parts of the cortex, visual areas 1, 2, velopmental trend for these three eye movement
and 4, the parietal cortex area PG, and the frontal eye systems.
fields (Schiller, 1985, 1998). Smooth pursuit eye move- Richards and Holley (1999) examined infants' track-
ments represent a third type of eye movements used in ing behavior over this age range under conditions of
tracking visual stimuli. These eye movements occur only attention and inattention. Their study shows how the
in the presence of smoothly moving visual stimuli, and development of the general arousal system affects the
smoothly track visual stimuli over a wide range of visual exhibition of eye movements in the first six months of
space. Smooth pursuit eye movements also are con- infancy. Infants at 8, 14, 20, and 26 weeks of age were
trolled by brain pathways involving the cortex, includ- presented with stimuli that moved at varying speeds (8-
ing areas MT (medial temporal) and MST (middle 24 deg/s) on a television monitor. The infants' heart
superior temporal), and perhaps the parietal cortex rate was recorded and periods of visual tracking were
(Schiller, 1985, 1998). The voluntary saccadic and separated into attentive and inattentive states using the
smooth pursuit eye movements are affected by attention heart-rate-defined attention phases described earlier.
whereas reflexive saccadic eye movements are relatively The infants' eye movements were tracked with an elec-
independent of attention control. trooculogram (EOG) by recording electrical potential
Unlike the blink reflex, the brain areas involved in changes due to shifts in the eyes. The eye movements
the control of these three eye movement systems un- were separated into smooth pursuit and saccadic eye
dergo developmental changes in the first six months. movements and related to the attentiveness of the
There have been several models of the brain changes infant.
330 COGNITION
FIGURE 22.7 The smooth pursuit EOG gain and saccade fre- period when sustained heart rate deceleration was occurring,
quency (saccades per second) as a function of stimulus track- and the bottom two plots were taken from the period after
ing speed and testing age (8 and 14 weeks combined, 20 and heart rate had returned to its prestimulus level. (Adapted
26 weeks combined). The top two plots were taken from the from Richards and Holley, 1999; Richards and Hunter, 1998.)
erence, i.e., look longer at the novel stimulus than the heart rate changes that are related to the attention
familiar stimulus during the paired-comparison test phases. On separate trials, at a delay defined by the de-
phase. celeration of heart rate, a delay defined by the return
Two studies using heart-rate-defined attention of heart rate to its prestimulus level, or time-defined
phases have shown that exposure to the familiar stim- delays, a familiarization stimulus was presented for 5 or
ulus during sustained attention results in recognition 6 seconds. One condition with the Sesame Street movie
memory for stimuli presented for just 5 or 6 seconds alone was provided (no familiarization stimulus, i.e., no-
(Frick and Richards, 2001; Richards, 1997b). In these exposure control) and one condition with a 20-second
studies infants at 14, 20, or 26 weeks of age were pre- exposure to the familiar stimulus was presented. Follow-
sented with a Sesame Street movie, "Follow That Bird," ing each familiar stimulus presentation, a paired-
on a television monitor. This movie is very interesting comparison recogniton memory test was done. The
to young infants and reliably elicits the full range of infants' duration of fixation on the novel and the
332 COGNITION
presented relatively frequently and a second stimulus is ages (e.g., checkerboard pattern, circles, squares). We
presented infrequently. These studies report a large have data from 6-month-old infants and are currently
negative ERP component occurring about 400-800 ms testing 4.5- and 7.5-month-old infants to extend these
after stimulus onset located primarily in the frontal and findings to other ages and to test developmental
central EEG leads. This has been labeled the Nc (neg- changes occurring in these memory processes.
ative central) component (Courchesne, 1977, 1978). In Figure 22.9 shows ERP changes from the central (Cz)
most studies the Nc component is larger to the infre- and parietal (Pz) leads in 6-month-old infants in re-
quently presented stimuli and is thought to represent a sponse to the visual stimuli. The top two graphs show
general attentive state or alerting to the presence of a the ERP changes for the first stimulus on each trial
novel stimulus. If the frequently presented and infre- (stimulus orienting), during sustained attention, and
quently presented stimuli are already familiar to the in- during inattentiveness. The ERP changes in these two
fant, the Nc component does not differ (Nelson and graphs show a significantly larger Nc component (neg-
Collins, 1991, 1992). This distinction does not occur in ative component about 400-700 ms) for the "attentive"
4-month-old infants (Karrer and Ackles, 1987; Nelson phases (stimulus orienting, sustained attention, dashed
and Collins, 1991, 1992) but does occur in infants 6 lines) compared to the "inattentive" phase (attention
months old and older. termination, solid line). This confirms the idea that the
These oddball procedure studies also report later Nc component represents an "attention-alerting" mech-
components in the ERP. These components are slowly anism that occurs in response to any visual stimulus.
changing positive or negative potential shifts from The bottom two graphs of figure 22.9 show the ERP
about 800 to 1500 ms following stimulus presentation. changes for the three presentation procedures (FF, IF,
Nelson and Collins (1991) reported that in 6-month- and IN) only for the presentations occurring during
old infants there are three distinctions that may be sustained attention. The Nc is not different for these
made. First, a familiar stimulus that was presented in- three procedures, but the slow wave portion of the
frequently (infrequent familiar, IF) resulted in an in- graphs (750-1500 ms) shows a large positive slow wave
creased slow-wave positivity in this later period relative for the infrequent familiar presentation and a smaller
to the ERP observed when the familiar stimulus was pre- negative slow wave for the infrequent novel stimulus.
sented frequently (frequent familiar, FF). Second, a se- The stimulus presentations occurring during inatten-
ries of stimuli that were never previously presented to tion did not show the slow wave differences between the
the infant and were presented infrequently (infrequent three presentation procedures (not shown in figure
novel, IN) were presented. These stimuli resulted in a 22.9).
negative slow wave component in this later interval. Figure 22.10 (see color plate 15) shows topographical
Thus, these novel stimuli show that infants were sensi- maps of the Nc and a later slow wave for the IF and FF
tive to novelty per se (IN versus IF) as well as the relative stimuli during sustained attention. These show the Nc
probability of stimulus occurrence (IF versus FF). As component as a widespread negativity in the central
with the Nc component, at 4 months of age there was area of the scalp occurring for the frequent and infre-
no difference between the later occurring slow waves, quent familiar stimuli, but the later slow wave compo-
whereas by 6 months of age (Nelson and Collins, 1991) nent occurred primarily in the frontal-central regions
or 8 or 12 months (Nelson and Collins, 1992; Nelson only for the infrequently presented stimuli.
and deRegnier, 1992) these three stimulus presentation The relation between sustained attention and infants'
procedures resulted in differing ERP potential shifts. recognition of briefly presented visual stimuli shows
We are currently conducting a study using this pro- that the arousal form of attention is related to complex
cedure, measuring ERPs, and presenting the FF, IF, and infant cognition. Recognition memory is accomplished
IN stimuli in different phases of attention (Richards, by several brain areas and cognitive functions. It re-
2000b). As with the other studies of infant recognition quires the acquisition of stimulus information and
memory (Frick and Richards, 2001; Richards, 1997b) memory storage over some period of time. The mea-
Ithe infant's attention is elicited with a Sesame Street surement of recognition memory also requires perfor-
movie, "Follow That Bird," that elicits the heart-rate- mance on a task exhibiting the existence of the stored
defined attention phases. Then, during stimulus ori- memory. The results of these studies show that the
enting, sustained attention, or attention termination, arousal aspect of attention may "invigorate" each of
the brief visual stimuli are presented overlaying (re- these cognitive processes. This enhances familiarization
placing) the attention-eliciting stimulus. These briefly when information acquisition is occurring, may facili-
presented stimuli consist of static computer-generated tate memory consolidation during the waiting period,
patterns that are easily discriminable by infants at these and enhances the processes involved in the exhibition
of recognition memory. The effect on recognition changes in specific attentional systems or that corre-
memory is true for the overall responses to the stimulus sponded to developments occurring in other brain-
in the paired-comparison recognition-memory test based attention-directed infant behavior.
phase (Richards and Casey, 1990) and for the individual There are three ways in which future research and
cognitive processes occurring for transient responses to progress in the study of the development of attention-
the stimulus (Richards, 2000b). arousal in infants could progress. First, this review was
limited to studies using heart rate as a measure of the
Conclusions general arousal system in the brain. There are other
measures that may be useful in this regard. For exam-
Attention shows dramatic development in the early pe- ple, continuous levels of EEG activity are thought to be
riod of infancy, from birth to 12 months. This chapter influenced by general arousal mechanisms in the brain.
has emphasized an attention system that represents a Since the EEG represents the summed activity of large
general arousal of cognitive functions. The system in groups of neurons, one might expect that the brain ar-
the brain controlling this arousal develops in the first eas controlling arousal or the neurochemical systems
few months of life, and this brain development is re- should have an influence on overall neural activity (ex-
sponsible for the behavioral/attentional development tent, duration, and localization). Thus, measures of
seen in young infants. This chapter reviewed several EEG such as spectral power and coherence may give
studies that showed the effect of this arousal system, information about arousal. Such measures also may
indexed by heart rate changes showing sustained atten- show relatively localized CNS arousal.
tion. There were developmental changes in infant sus- A second area in which research on the development
tained attention that were reflected in developmental of the brain systems controlling arousal may benefit is
334 COGNITION
FIGURE 22.10 A topographical mapping of the ERP compo- Nc (centered at 560 ms) and the slow wave (centered at 1120
nents occurring during sustained attention. The ERP com- ms) components for the 20 recording electrodes. The data
ponents were the Nc (400-700 ms; top figures) and the later are plotted with a cubic spline interpolation algorithm, with
slow wave component (700-1500 ms; bottom figures) taken an averaged electrode reference, and represents absolute am-
during the presentation of the frequent familiar (left figures) plitude of the ERP for the recorded data rather than differ-
and the infrequent familiar (right figures) stimuli. The data ence ERPs.
in each figure represent an 80-ms average of the ERP for the
direct measures of the brain. Such measures in animal ever, noninvasive measurements from psychophysiolog-
models have included invasive chemical manipulations ical measures that are tuned to specific neurochemical
and measurement as well as destruction of the areas systems might be found. Perhaps one type of quantita-
controlling arousal through lesions or neurochemical tive activity in the EEG may be linked to a specific neu-
inhibitors. These measures cannot be applied in infant rochemical system and another type linked to another
participants because of ethical considerations. How- system. The simple recording of EEG, ERP, or heart rate
336 COGNITION
DESIMONE, R., and J. DUNCAN, 1995. Neural mechanisms of JOHNSON, M. H., R. O. GILMORE, and G. CSIBRA, 1998. Toward
selective visual attention. Annu. Rev. Neurosci. 18:193-222 a computational model of the development of saccade plan-
FAGAN, J. F., 1974. Infant recognition memory: The effects of ning. In Cognitive Neuroscience of Attention: A Developmental
length of familiarization and type of discrimination task. Perspective,]. E. Richards, ed. Hillsdale, N.J.: Lawrence Erl-
Child Deo. 59:1198-1210. baum, pp. 103-130.
FRICK, J., and J. E. RICHARDS, 2001. Individual differences in JOHNSON, M. H., M. I. POSNER, and M. K. ROTHBART, 1991.
recognition of briefly presented visual stimuli. Infancy 2 (in Components of visual orienting in early infancy: Contin-
press). gency learning, anticipatory looking and disengaging. J
GRAHAM, F. K., 1979. Distinguishing among orienting, de- Cogn. Neurosci. 3:335-344.
fense, and startle reflexes. In The Orienting Reflex in Humans, KARRER, R., and P. K. ACKLES, 1987. Visual event-related po-
H. D. Kimmel, E. H. van Olst, and J. F. Orlebeke, eds. Hills- tentials of infants during a modified oddball procedure. In
dale, N.J.: Lawrence Erlbaum, pp. 137-167. Current Trends in Event-Related Potential Research, R.Johnson,
GRAHAM, F. K., 1992. Attention: The heartbeat, the blink, and J. W. Rohrbaugh, and R. Parasuraman, eds. Amsterdam: El-
the brain. In Attention and Information Processing in Infants sevier Science, pp. 603-608.
and Adults: Perspectives from Human and Animal Research, B. A. KARRER, R., and P. K. ACKLES, 1988. Brain organization and
Campbell, H. Hayne, and R. Richardson, eds. Hillsdale, N.J.: perceptual/cognitive development in normal and Down
Lawrence Erlbaum, pp. 3-29. syndrome infants: A research program. In The Early Identi-
HACKLEY, S. A., and F. K. GRAHAM, 1983. Early selective atten- fication of Infants with Developmental Disabilities, P. Vietze and
tion effects on cutaneous and acoustic blink reflexes. Phys- H. G. Vaughan, Jr., eds. Philadelphia: Grune & Stratton.
iol.Psychol. 11:235-242. KARRER, R., and L. A. MONTI, 1995. Event-related potentials
HACKLEY, S. A., and A. J. W. BOELHOUWER, 1997. The more of 4-7-week-old infants in a visual recognition memory task.
or less startling effects of weak prestimulation—revisited: Electroencephalogr. Clin. Neurophysiol. 94:414-424.
Prepulse modulation of multicomponent blink reflexes. In LANSINK, J. M., and J. E. RICHARDS, 1997. Heart rate and be-
Attention and Orienting: Sensory and Motivational Processes, P.J. havioral measures of attention in six-, nine-, and twelve-
Lang, R. F. Simons, and M. T. Balaban, eds. Mahway, N.J.: month-old infants during object exploration. Child Dev.
Erlbaum, pp. 205-228. 68:610-620.
HAERICH, P., 1994. Startle reflex modification: Effects of at- MAUNSELL, J. H. R., and V. B. FERRERA, 1995. Attentional
tention vary with emotional valence. Psychol. Set. 5:407-410. mechanisms in visual cortex. In The Cognitive Neurosciences,
HEILMAN, K. M., R. T. WATSON, E. VALENSTEIN, and M. E. M. S. Gazzaniga, ed. Cambridge, Mass.: MIT Press, pp. 451-
GOLDBERG, 1987. Attention: Behavior and neural mecha- 461.
nisms. In Handbook of Physiology, V. B. Mountcastle, F. Plum, MAURER, D., and T. L. LEWIS, 1979. A physiological explana-
and S. R. Geiger, eds. Bethesda, Md.: American Physiologi- tion of infants' early visual development. Can. J. Psychol.
cal Society, pp. 461-481. 33:232-252.
HICKS, J. M., andj. E. RICHARDS, 1998. The effects of stimulus MAURER, D., and T. L. LEWIS, 1991. The development of pe-
movement and attention on peripheral stimulus localiza- ripheral vision and its physiological underpinnings. In New-
tion by 8- to 26-week-old infants. Infant Behav. Dev. 21:571- born Attention: Biological Constraints and the Influence of
589. Experience, M. J. S. Weiss and P. R. Zelazo, eds. Norwood,
HILLYARD, S. A., G. R. MANGUN, M. G. WOLDROFF, and S. J. N.J.: Ablex, pp. 218-255.
LUCK, 1995. Neural systems mediating selective attention. MAURER, D., and T. L. LEWIS, 1998. Overt orienting toward
In The Cognitive Neurosciences, M. S. Gazzaniga, ed. Cam- peripheral stimuli: Normal development and underlying
bridge, Mass.: MIT Press, pp. 665-682. mechanisms. In Cognitive Neuroscience of Attention: A Develop-
HOOD, B. M., 1995. Shifts of visual attention in the human mental Perspective, J. E. Richards, ed. Hillsdale, N.J.: Lawrence
infant: A neuroscientific approach. Adv. Infancy Res. 10:163— Erlbaum, pp. 51-102.
216. MESULAM, M. M., 1983. The functional anatomy and hemi-
HOOD, B. M.,J. ATKINSON, and O. J. BRADDICK, 1998. Selec- spheric specialization for directed attention. Trends Neurosci.
tion-for-action and the development of orienting and visual 6:384-387.
attention. In Cognitive Neuroscience of Attention: A Developmen- NELSON, C. A., 1994. Neural correlates of recognition memory
tal Perspective, J. E. Richards, ed. Hillsdale, N.J.: Lawrence in the first postnatal year. In Human Behavior and the Devel-
Erlbaum, pp. 219-250. oping Brain, G. Dawson and K. W. Fischer, eds. New York:
HUNTER, S. K, and J. E. RICHARDS, 1999. Peripheral stimulus Guilford Press, pp. 269-313.
localization by 5- to 14-week-old infants during phases of NELSON, C. A., and P. F. COLLINS, 1991. Event-related potential
attention, (submitted). and looking-time analysis of infants' responses to familiar
HUNTER, S. K, and J. E. RICHARDS, 2000. Postural and stimulus and novel events: Implications for visual recognition mem-
effects on eye movement in young infants, (submitted). ory. Dev. Psychol. 27:50-58.
JOHNSON, M. H., 1990. Cortical maturation and the develop- NELSON, C. A., and P. F. COLLINS, 1992. Neural and behavioral
ment of visual attention in early infancy. J. Cogn. Neurosci. correlates of visual recognition memory in 4- and 8-month-
2:81-95. old infants. Brain Cogn. 19:105-121.
JOHNSON, M. H., 1995. The development of visual attention: NELSON, C. A., and D. DUKETTE, 1998. In Cognitive Neuroscience
A cognitive neuroscience perspective. In The Cognitive Neu- of Attention: A Developmental Perspective, J. E. Richards, ed.
rosciences, M. S. Gazzaniga, ed. Cambridge, Mass.: MIT Press, Hillsdale, N.J.: Lawrence Erlbaum, pp. 327-362.
pp. 735-747. NELSON, C. A., and R. A. DEREGNIER, 1992. Neural correlates
JOHNSON, M. H., 1997. Developmental Cognitive Neuroscience. of attention and memory in the first year of life. Dev. Neu-
London: Blackwell. ropsychol. 8:119-134.
338 COGNITION
23 The Functional Development
and Integration of the Dorsal
and Ventral Visual Pathways:
A Neurocomputational Approach
MARK H.JOHNSON, DENIS
MARESCHAL, AND GERGELYCSIBRA
ABSTRACT In this chapter we discuss evidence from com- our knowledge of the functional development of the
putational, brain imaging, and behavioral studies relating to two main cortical visual pathways in the human brain.
the development of the dorsal and ventral visual pathways. There is now a substantive body of evidence that vi-
This evidence indicates that (1) for both pathways some as-
pects of function are delayed in development relative to others sual information processing in the primate cortex is di-
and (2) the two pathways may be less integrated during in- vided into two relatively distinct streams (see Stiles, this
fancy than in later life. We discuss some of the factors that volume). However, only recently has the dual route vi-
contribute to these differential timetables of development and sual processing paradigm been applied to the study of
some of the possible consequences thereof. infant perceptual and cognitive development (Atkin-
son, 1998; Berthenthal, 1996; Mareschal, Plunkett, and
In the emerging field of developmental cognitive
Harris, 1999), and many fundamental questions re-
neuroscience, we approach our understanding of the
main. One question is whether it is the dorsal or the
functional development of the human brain by inte-
ventral route that functionally develops first during in-
grating evidence from neural development with that
fancy, while another concerns whether there is increas-
from cognitive and behavioral studies. Whereas it is
relatively straightforward to correlate changes at the neu- ing separation, or increasing integration, between the
ral level with changes in behavior, it is challenging to two pathways with development. In this chapter we re-
integrate information from these sources into a single, view evidence from neuroimaging, computational mod-
cohesive account of developmental change. We believe eling, and behavior, indicating that these questions are
that recent advances in two areas, one theoretical and unlikely to have simple answers. Although there is some
one technological, will allow more rapid progress in evidence that dorsal stream eye movement control is
integrating neural and behavioral evidence in devel- relatively delayed, other aspects of dorsal route function
opment. First, the advent of neural network and con- (such as reaching) may be more precocial. Computa-
nectionist modeling will provide the appropriate level tional and behavioral evidence indicates that even when
of theoretical framework in which evidence from neural the dorsal route is functioning, there may be an initial
development and behavior can be modeled simulta- lack of integration between the two pathways resulting
neously (see Munakata and Stedron, this volume). Sec- in specific patterns of behavioral deficits. We conclude
ond, recent advances in neuroimaging now allow the with speculations as to the causes and consequences of
noninvasive measurement of brain activity in healthy different developmental timetables in the dorsal and
infants with reasonable spatial and temporal accuracy. ventral cortical streams.
In this chapter we discuss evidence gathered through
both of these new techniques in an attempt to advance The dorsal and ventral routes of visual processing
340 COGNITION
object. These authors conditioned monkeys to respond mation is at a premium down this pathway. Motor
with a reach for a desired object and not to respond for actions involve localizing targets within a three-
an undesired object. They found that the cells in the dimensional spatial temporal world. In contrast, rec-
parietal cortex fired (i.e., computed the appropriate ognition or identification of objects requires that
spatial-temporal properties) even if the object was an spatial-temporal variability be minimized. Early work in
undesired object and the monkey did not reach. Hence, machine vision found that view-invariant recognition
processing in this channel appears to be independent (i.e., the ability to recognize an object as the same in-
of superficial feature recognition and any of the re- dependently of orientation and location) was a very dif-
sponses associated with the target object. ficult computational problem (Boden, 1989). One of
the most efficient ways of achieving view-invariant rec-
PROCESSING IN THE VENTRAL PATHWAY The properties ognition is to factor out spatial variability. However,
of cells in the ventral stream seem to complement those removing spatial information from the object represen-
in the dorsal stream. There appears to be some kind of tation is completely at odds with the requirements of
hierarchical representation in the ventral stream. As one the motor system. Hence the need for two distinct
progresses down the stream, cells respond to more and classes of object representations.
more complex clusters of features. At the higher levels, Finally, Jeannerod (1999) has argued that recent neu-
the complex cells show remarkable selectivity in their fir- roimaging evidence indicates that the dissociation be-
ing (e.g., face recognition cells). These neurons are all tween the two pathways is less clear than presented by
selective to the figural and surface properties of objects, Milner and Goodale (1995). Specifically, structures on
and have very large receptive fields on the retina. Al- the dorsal pathway are often activated following the pre-
though they can process feature information, they lose sentation of objects, without manual response being re-
much of their spatial resolution on the retina. In effect, quired. Leaving aside the question of whether eye
these cells develop spatially invariant representations of movements are elicited or planned in these passive view-
objects by responding to the presence of a consistent fea- ing paradigms, it is hard to rule out the possibility that
ture cluster independently of its position. Some cells reaching actions are automatically planned, even if not
seem to respond maximally to a preferred object orien- executed, on the presentation of graspable objects. Our
tation (independently of position), thereby computing view is that the coactivation of the two pathways is en-
a "view-centered representation." Other cells respond tirely consistent with two streams of information pro-
equally to an object in any orientation. These cells have cessing in which the type of processing that occurs
developed a transformation-invariant representation. within each pathway is incompatible.
Transformation-invariant representations could pro-
vide the basic raw material for recognition memory and The development of the dorsal and ventral streams
other long-term representations of the visual world.
There is evidence that the responsiveness of cells in the Identifying the presence of two visual processing
ventral stream can be modulated by the prior occur- streams begs the question of how these streams develop
rence of a stimulus. Moreover, there is evidence indi- and how they interact with each other during develop-
cating that some cells continue to fire for several ment. Over the past few years there have been some
seconds after the object has disappeared (Ungerleider, speculations about the developmental sequence of the
1985), suggesting that some kind of memory trace two pathways. For example, Atkinson (1998) has ar-
remains. gued, on the basis that infants and children are delayed
In this chapter, we are less concerned with whether on their judgments of motion coherence compared to
these pathways are called "what" and "where" (Unger- their thresholds for perceiving coherent forms, that the
leider and Mishkin, 1982) or "perception" and "action" dorsal pathway develops later than the ventral. In con-
(Milner and Goodale, 1995) and more concerned with trast, studies of developmental neuroanatomy in 1-
the fact that they process different types of object in- week-old macaque monkeys led Webster and colleagues
formation, carry out different computations, and de- (1995) to the view that the patterns of connectivity of
velop different object representations with distinct temporal lobe structures on the ventral pathway were
properties. Nevertheless, one can speculate as to the still relatively immature, while the connectivity of the
computational reasons for why two streams of process- parietal lobe (on the dorsal route) was already adult-
ing may have evolved. If we agree with Milner and like. Evidence from developmental neuroanatomy in
Goodale (1995) that the representations in the dorsal human infants is not compelling in this regard. For ex-
stream are closely linked to the functions of the motor ample, in resting PET studies of glucose uptake, virtu-
system, it is not surprising that spatial-temporal infor- ally identical overall patterns of developmental change
342 COGNITION
(e.g., Johnson, 1990), we were surprised to find no evi- leads consistent with frontal eye field disinhibition of
dence of these components (figure 23.1) in our infant subcortical (collicular) circuits when a central foveated
subjects (see also Kurtzberg and Vaughan, 1981; Rich- stimulus is removed (Csibra, Tucker, and Johnson,
ards, in press). This finding suggests that the target- 1998; in press). In brief, we interpret these findings in
driven saccades performed by 6-month-olds in our terms of the frontal eye fields maintaining fixation onto
study were controlled solely by subcortical routes for foveated stimuli by inhibiting collicular circuits (see
visually guided responses mediated by the superior Johnson, 1990). However, when saccades to peripheral
colliculus. stimuli are made, we believe these are largely initiated
Because this result was surprising, we conducted two by collicular circuits, sometimes as a consequence of
follow-up studies. In one, we tested 12-month-olds with inhibition being released by the frontal eye fields.
the same procedure. Preliminary results from this ex- In the saccadic ERP data from 6-month-olds there was
periment indicate that these older infants do show a strong evidence for a postsaccadic component known
spike potential like that observed in adults, though as the lambda wave. The lambda wave is a sharp poten-
smaller in amplitude (figure 23.2). The other study ex- tial appearing over visual cortical areas that is generated
plored whether the dorsal pathway could be activated when a peripheral target stimulus is foveated (Kurtz-
in very young infants through a more demanding sac- berg and Vaughan, 1977). While lambda waves can be
cade task. Specifically, we compared ERPs before reac- observed in adults, they were markedly enhanced in our
tive (target-elicited) and anticipatory (endogenous) infant subjects. One conclusion from this work is that
saccades in 4-month-old infants (Csibra, Tucker, and while dorsal pathway control of eye movements is not
Johnson, in press). We were not able to record any re- evident at 6 months, early visual cortical responses to
liable posterior activity prior to either reactive or antic- foveal stimuli, believed to originate from structures at
ipatory eye movements. Thus, even when the saccade is the gateway to the ventral pathway (V2 and V4), are
generated by cortical computation of the likely location present. Thus, within a single ERP trace, there is evi-
of the next stimulus, as in the case of anticipatory eye dence for the relatively delayed development of the dor-
movements, the dorsal pathway does not seem to be sal pathway with respect to the ventral pathway. But why
involved in the planning of this action. should the lambda wave be enhanced in infants relative
As an aside, we should note that the lack of evidence to adults? One possibility may lie in the fact that in
for parietal (dorsal pathway) control over eye move- order to successfully integrate visual input with eye
ments in our experiments with 6-month-olds does not movements, adults "forward-map" the expected visual
lead us to conclude that there is no cortical influence input at the end of their saccade (see Johnson et al.,
over saccades at this age. In all our studies with this age 1998). If infants are unable to integrate information
group we have observed effects recorded over frontal about visual input and eye movements due to a lack of
344 COGNITION
FIGURE 23.3 The two-dimensional version of the double-step
saccade task. (From Gilmore and Johnson, 1997.)
ities to develop a motion representation over different conjunction with size or volume information (Mares-
feature transformations. In one experiment, these au- chal, 1997). Both size and volume are spatial dimen-
thors habituated infants to an object moving in differ- sions that are likely to be encoded by the dorsal route.
ent ways. They then tested the infant with either the Thus, these tasks only provide evidence of dorsal route
same object moving in a novel way, or a novel object processing.
moving in the same novel way. They found that infants Further data supporting the suggestion of a disso-
looked reliably longer at the novel object. The impli- ciation between ventral and dorsal processing have
cation is that the infants had encoded the shapes of the recently become available. Infants fail to use surface-
objects and based their responses on the novel shape feature information to individuate and enumerate ob-
rather than the novel movement: Infants could encode jects that move behind and out from a screen (Leslie et
shape independently of movement. In a second exper- al., 1998; Simon, Hespos, and Rochat, 1995; Xu and
iment, the authors habituated infants to different ob- Carey, 1996). In these studies, infants watched two dif-
jects moving in the same way. They then tested the ferent objects move in and out (one at a time) from
infants with a novel object moving either the same way behind an occluder. The screen was subsequently re-
moved to reveal either one or two objects. Young infants
or in a novel way. They found that infants looked reli-
consistently ignored surface-feature information and
ably longer at the novel movement. The implication is
relied on spatial-temporal cues when assessing the num-
that the infants had encoded the movements of the ob- ber of objects behind the occluder as indexed by fixa-
jects independently of shape and based their responses tion time. Using a paradigm similar to that of Xu and
on the novel movement rather than the novel shape: Carey (1996), Wilcox (1999) systematically varied (one
Infants could encode movement independently of at a time) the features by which pairs of objects differed
shape. when they appeared from behind the occluding screen.
Further behavioral evidence comes from studies of She found that at 4.5 months, infants will use shape and
infant responses to temporarily occluded objects. Early size information to individuate objects, but only at 7.5
competence in preferential looking tasks that claimed months will they use surface texture information, and
to show the coordination of position and feature infor- not until 11.5 months do they use color to individuate
mation in reasoning about hidden objects (e.g., Bail- objects. Note that shape is not a cortically separable ob-
largeon, 1993) have—on close scrutiny—provided ject feature as it is processed in both the dorsal and
evidence only for the use of positional information in ventral routes. Thus, an infant relying on the dorsal
346 COGNITION
Thus, two successive images will probably be derived well as trajectory information (to localize the object).
from the same object. At the end of learning each com- Related arguments about the need to coordinate "what"
plex cell becomes associated with a particular feature and "where" information in object-directed tasks have
combination wherever it appears on the retina. been presented elsewhere (e.g., Leslie et al., 1998;
The trajectory prediction module uses a partially recur- Prazdny, 1980).
rent, feed-forward network trained with the backpro- The model embodies the basic architectural con-
pagation learning algorithm. This network learns to straints on visual cortical pathways revealed by contem-
predict the next instantaneous, retinal position of the porary neuroscience: an object-recognition network
object, and is designed to capture properties associated that develops spatially invariant feature representations
with the dorsal route of visual processing. The internal of objects, a trajectory-prediction network that is blind
representations it develops are constrained both by the to surface features and computes appropriate spatial-
computational properties of the associative mecha- temporal properties even if no actions are undertaken
nisms in the network and by the spatial-temporal pre- toward the object, and a response module that inte-
diction task it is engaged in. grates information from the two latter networks for use
All units in the visual memory layer have a self- in voluntary actions. In this model surprise is caused by
recurrent connection. The resulting spatial distribution a mismatch between the information stored in an in-
of activation across the visual memory layer takes the ternal representation and the new information arriving
form of a comet with a tail that tapers off in the direc- from the external world. More specifically, in the
tion from which the object has come. The length and trajectory-prediction module, surprise occurs when
distinctiveness of this tail depend on the velocity of the there is a discrepancy between the predicted reappear-
object. The information in this layer is then forced ance of an object from behind an occluder and its
through a bottleneck of 75 hidden units to generate a actual reappearance on the retina. In the object-
more compact, internal rerepresentation of the object's recognition module, surprise occurs when there is a dis-
spatial-temporal history. As there are no direct connec- crepancy between the feature representation stored
tions from the input to the output, the network's ability across the complex units and the new representation
to predict the next position is a direct measure of the produced by the new image.
reliability of its internal object representation. We in- The model learns to track both visible and occluded
terpret the response of the trajectory prediction net- objects. Each pathway develops specific task-appropriate
work as a measure of its sensitivity to spatial-temporal representations that persist beyond direct perception.
information about the object. It is successful in demonstrating how the requirement
The output of the response integration network corre- to integrate information across two object representa-
sponds to the infant's ability to coordinate and use the tions in a voluntary retrieval task can lead to a devel-
information it has about object position and object opment lag relative to performance on surprise tasks
identity. This network integrates the internal represen- that only require access to either spatial-temporal in-
tations generated by other modules (i.e., the feature formation concerning an occluded object or surface-
representation at the complex cell level and spatial- feature information accessed separately. Moreover, this
temporal representation in the hidden unit layer) as lag appears only when the network is required to deal
required by a retrieval response task. It consists of a with hidden objects, as is the case with infants.
single layered perception whose task is to output the Note that early surprise responses can arise from fea-
same next position as the prediction network for two of ture violations, from spatial-temporal violations, and
the objects, and to inhibit any response (all units set to even from both types of violation arising concurrently
0.0) for the other two objects. This reflects the fact that and independently, but not from a violation involving
infants do not retrieve (e.g., reach for) all objects. In the integration of feature and spatial-temporal informa-
general, infants are not asked or rewarded for search. tion concerning an occluded object. The model pre-
The experimental set-up relies on spontaneous search dicts that infants will show a developmental lag not just
by the infant. Some objects are desired (e.g., sweet) on manual search tasks but also on surprise tasks that
whereas others are not desired (e.g., sour). Heighten- involve such integration. Conversely, the model sug-
ing the desirability of an object (e.g., by providing the gests that infants will show early mastery of response
infant with a prior opportunity to play with the object) tasks that do not require the integration of information
has been shown to elicit more search in manual re- across cortical representations.
trieval tasks (Harris, 1986). Any voluntary retrieval re- As noted earlier, the developmental lag in the
sponse will necessarily require the processing of feature model is not caused simply by the need to integrate
information (to identify the object as a desired one) as spatial-temporal and featural information. The same
348 COGNITION
1998). Furthermore, as the infant becomes more mo- We have presented some elements of a future re-
bile toward the end of the first year of life, the nature search program into the development of the dorsal and
of his or her experience of the world changes. From his ventral visual pathways. In these future studies we antic-
or her initial stance as a relatively passive viewer of the ipate a tighter integration between neuroimaging, mod-
world, the infant becomes able to explore objects and eling, and behavior, which will lead us to a more
spatial environments manually. This new "infant- cohesive account of the development of visually guided
generated" experience no doubt helps shape appropri- behavior during infancy.
ate dorsal pathway cortical circuitry.
Why might there be a dissociation within the dorsal ACKNOWLEDGMENTS We acknowledge financial support
pathway between eye movement control and reaching? from MRC Programme grant G97 15587, EU Biomed grant
BMH4-CT97-2032, and Birkbeck College.
It is established that there is an effective subcortical
(collicular) route for eye movement control. As we sug-
gested earlier, this route may be adequate for many sit- REFERENCES
uations in which saccades are required, and is sufficient ANDERSON, R. A., L. H. SYNDER, C. S. Li, and B. STRICANNE,
to produce an adaptive response in most circumstances. 1993. Coordinate transformations in the representation of
It is only when more challenging saccade paradigms are spatial information. Curr. Opin. Neurobiol. 3:171-176.
tried, such as the double-step paradigm, that maladap- ATKINSON, J., 1998. The 'where and what' or 'who and how'
tive behavior is revealed. Thus, while there are clearly of visual development. In The Development of Sensory, Motor
and Cognitive Capacities in Early Infancy: From Perception to Cog-
benefits to dorsal pathway control over eye movements, nition, F. Simion and G. Butterworth, eds. Hove, East Sussex:
the gains to be made may be less than those for the Psychology Press, pp. 3-20.
cortical control of reaching. BAILLARGEON, R., 1993. The object concept revisited: New di-
By delaying aspects of the development of the dorsal rections in the investigation of infant's physical knowledge.
pathway, plasticity can be retained for longer periods of In Visual Perception and Cognition in Infancy, C. E. Granrud,
ed. Hillsdale, N.J.: Lawrence Erlbaum, pp. 265-315.
time. However, there is a potential cost to this in that
BALABAN, C. D., and J. M. WEINSTEIN, 1985. The human pre-
later developing systems tend to be more vulnerable to saccadic spike potential: Influences of a visual target, sac-
disruptions in development (see Johnson, 1997). Thus, cade direction, electrode laterality and instruction to
we anticipate that the dorsal visual stream should show perform saccades. Brain Res. 347:49—57.
more evidence of plastic changes following differences BENTIN, S., T. ALLISON, A. PUCE, E. PEREZ, and G. MCCARTHY,
in experience in otherwise healthy children, but that 1996. Electrophysiological studies of face perception in hu-
mans.J. Cogn. Neurosci. 8(6):551-565.
the dorsal pathway will also show more evidence of def- BERTENHAL, B. I., 1996. Origins and early development of per-
icits in some developmental disorders. There is evi- ception, action and representation. Annu. Rev. Psychol.
dence consistent with both of these predictions. 47:431-459.
In a series of experiments involving comparisons be- BODEN, M. A., 1989. Computer Models of Mind. Cambridge: Cam-
tween congenitally deaf and hearing subjects, Neville bridge University Press.
BOWER, T. G. R., and J. G. PATTERSON, 1973. The separation
and colleagues have found results indicating that the of place, movement and object in the world of the infant.
dorsal stream may be more modifiable in response to J. Exp. Child Psychol. 15:161-168.
alterations in afferent input than the ventral pathway BOWERING, E., D. MAURER, T. L. LEWIS, H. P. BRENT, and
(Neville, 1995). For example, while visual responses re- P. RIEDEL, 1996. The visual field in childhood: Normal de-
corded at the scalp to peripheral and transient signals velopment and the influence of deprivation. Developmen-
tal Cognitive Neuroscience Technical Report No. 96.1.
are greatly enhanced in the congenitally deaf as com- BURNHAM, D. E, 1987. The role of movement in object per-
pared to hearing, there are no differences between ception by infants. In Perceptual Development in Early Infancy:
these groups in measured responses to foveal stimuli. Problems and Issues, B. McKennie, and R. H. Day, eds. Hills-
In a parallel series of studies, Maurer and colleagues dale, N.J.: Lawrence Erlbaum.
have investigated the effect of early visual deprivation CHUGANI, H. T, M. E. PHELPS, and J. C. MAZZIOTTA, 1987.
Positron emission tomography study of human brain func-
on subsequent visual acuity (Bowering et al., 1996).
tional development. Ann. Neural. 22:487-497.
Children who had cataracts earlier in life subsequently CSIBRA, G., M. H.JOHNSON, and L. TUCKER, 1997. Attention
showed deficits in their sensitivity to peripheral visual and oculomotor control: A high density ERP study of the
stimuli several years later. Even children who had had gap effect. Neuropsychologia 35:855-865.
unilateral cataracts subsequently showed prolonged def- CSIBRA, G., L. A. TUCKER, and M. H.JOHNSON, 1998. Neural
correlates of saccade planning in infants: A high-density
icits in their sensitivity to peripheral visual stimuli, lead-
ERP study. Intl. J. Psychophysiol 29:201-215.
ing the authors to conclude that cortical systems CSIBRA, G., L. A. TUCKER, and M. H. JOHNSON, in press. An-
subserving peripheral vision appear to be more sensi- ticipatory and reactive saccades in infants: A high-density
tive to experience early in life. ERP study. Infancy.
350 COGNITION
SIMON, T. J., S. J. HESPOS, and P. ROCHAT, 1995. Do infants VAN ESSEN, D. C., C. H. ANDERSON, and D. J. FELLEMAN, 1992.
understand simple arithmetic? A replication of Wynn Information processing in the primate visual system: An in-
(1992). Cogn.Dev. 10:253-269. tegrated system perspective. Science 255:419-422.
UNGERLEIDER, L. G., and M. MISHKIN, 1982. Two cortical vi- WEBSTER, M. J., J. BACHEVALIER, and L. G. UNGERLEIDER,
sual systems: Separation of appearance and location of ob- 1995. Development and plasticity of visual memory circuits.
jects. In Analysis of Visual Behavior, D. L. Ingle, M. A. In Maturational Windows and Adult Cortical Plasticity, B.Julesz
Goodale, and R. J. Mansfield, eds. Cambridge, Mass.: MIT and I. Kovacs, eds. Reading, Mass.: Addison-Wesley, pp. 73-
Press. 86.
UNGERLEIDER, L. G., 1985. The cortical pathways for object WILCOX, T, 1999. Object individuation: Infants' use of shape,
recognition and spatial perception. In Pattern Recognition size, pattern and colour. Cognition 72:125-166.
Mechanisms, C. Changas and C. Gross, eds. Berlin: Springer- Xu, F., and S. CAREY, 1996. Infants' metaphysics: The case of
Verlag, pp. 21-37. numerical identity. Cogn. Psychol. 30:11-153.
ABSTRACT Two major approaches to understanding devel- theme. We then discuss how temperament is defined
opment are the delineation of common mechanisms of emo- and measured and what is currently known about di-
tion and cognition and the study of individual differences mensions of attentional focusing and shifting and ef-
among these mechanisms. In this chapter we attempt to in-
tegrate these two approaches within the domain of attentional fortful control. Finally, we briefly consider the role of
reactivity and self-regulation. We examine brain networks in- temperament in empathy and conscience.
volved in orienting and executive control and discuss how they
relate to temperamental differences in attention. We believe Attention and self-regulation
consideration of individual variations in how common mech-
anisms of attention function can greatly enlarge our under- In 1958, Donald Broadbent summarized British work in
standing of normal and pathological development.
the field of attention in his volume Perception and Com-
This chapter describes efforts to develop a common munication. He proposed the existence of a filter that
framework for studies arising from two quite different held back signals from an unattended sensory channel
traditions within psychology. The first involves the de- to keep them from interfering with selected input. His
velopment of attention as a mechanism common to all beautiful studies, summarized in nearly every textbook
humans and related to regulation of other types of cog- in psychology, provided a basis for studying how indi-
nitive and emotional behavior (Posner and Petersen, viduals select relevant information from masses of po-
1990; Posner and Rothbart, 1998). We argue that atte- tential input. The studies reviewed in Broadbent's book
tion operates through a series of networks to allow peo- were conducted largely in Britain during and immedi-
ple to regulate selection of sensory input, motor output, ately following World War II. The viewpoint expressed
and emotions in the service of internal goals. The sec- in this body of work is that attention is a high-level skill
ond tradition involves the study of individual differ- devoted to the selection of relevant information. Most
ences in the ability to orient attention and to regulate attention research has been carried out within a cog-
activity through effort (Rothbart, Ahadi, and Evans, nitive framework following the Broadbent model. It has
2000). The temperamental dimensions of orienting and involved adult subjects, who are instructed to carry out
effortful control summarize differences among people a specific task. Attempts have been made to develop
in their orienting reactivity and in their ability to exer- models of information flow within the nervous system
cise attentional regulation. that could be generalized across all subjects.
We begin by developing the theme that attention can An important link between studies of attention in
be considered a vehicle for self-regulation. This is not human cognition and those using the methods of neu-
a common view, so we focus on how this perspective rophysiology was provided by Y N. Sokolov (1963). So-
arose. We then briefly outline what is known from re- kolov's description of the orienting reflex provided a
search on attention networks and their development. physical basis for filtering sensory input and presaged
Because there are extensive reviews of this material the intense interest within neurophysiology in how at-
(e.g., Richards, 1998; Ruff and Rothbart, 1996), we em- tention might modulate activity within sensory specific
phasize only those aspects that deal with our general areas (Hillyard and Anllo-Vento, 1998). Luria (1973)
recognized the importance of orienting and distin-
guished between an early developing (largely involun-
MARY K. ROTHBART Department of Psychology, University of tary) , biological attention system and a later developing
Oregon, Eugene, Oregon.
MICHAEL i. POSNER Sackler Institute for Developmental Psy- (more voluntary) social attention system. We believe
chobiology, Weill Medical College of Cornell University, New that Luria was roughly correct in making a distinction
York, New York. between voluntary and involuntary attention systems,
354 COGNITION
FIGURE 24.1 Alerting and orienting networks. The upper during attention shifts (c). The lower panel indicates the role
panel indicates the frontal and parietal areas involved in alert- of the anterior cingulate in interacting with cortical areas in-
ing (a, b) and the superior parietal areas shown to be active volved in a wide variety of cognitive processes.
mentary motor area (SMA), and portions of the basal terns of activation when recorded with depth electrodes
ganglia. Neuroimaging studies have shown activity in during tasks such as lexical decisions and semantic cat-
these areas during tasks that require mental effort, such egorizations that were not due to motor or premotor
as in dealing with conflict (Posner and DiGirolamo, programming (Abdullaev, Bechtereva, and Melnichuk,
1998). It has also been argued that tasks involving both 1998). Many studies have shown behavioral deficits on
cognitive and emotional controls, both of which require these tasks in animals following experimental lesions
the selection of relevant information, produce activa- of the anterior dorsolateral frontal cortex and the
tion in separate areas of the anterior cingulate (Bush et caudate.
al., 1998; Bush, Luu, and Posner, 2000). Figure 24.2 in- The basal ganglia have been considered to be par-
dicates the separation between areas activated during ticularly important in mediating the connection be-
emotional control tasks and those activated during cog- tween executive attention and other attentional
nitive tasks involving neutral valence. networks (LaBerge, 1995). Lateral areas of the frontal
Accumulating evidence suggests involvement of the cortex have also often been identified with executive
basal ganglia, and more specifically the caudate nu- attention. However, there is also much literature sug-
cleus, in cognitive functioning (Hayes et al., 1998). The gesting that these areas involve representation of spe-
caudate and the prefrontal cortex showed similar pat- cific kinds of spatial, verbal, or form information
356 COGNITION
cient to cause the infant to disengage from fixation to Studies with patient populations showed that lesions
this central attractor and shift to these new objects. This of a midbrain eye movement structure, the superior col-
pattern changes very dramatically in the period be- liculus, but not of various cortical structures, interfered
tween 2 and 4 months of age, when infants will disen- with this tendency to visually explore a novel location.
gage from fixation and orient to newly presented Since inhibition of return appears to depend upon a
peripheral stimuli, a finding that might partly reflect collicular mechanism, we expected inhibition of return
the improved visual acuity that co-occurs during this to be present at the youngest ages we studied (Posner
period (Gwiazda, Bauer and Held, 1989). However, in- et al., 1985). There is now widespread agreement that
fants of 4 months are also able to learn to use central inhibition of return is found in 4-month-old infants,
cues to reorient attention. When distinct attractors were but not infants of 2-3 months (Richards and Hunter,
used to cue right-sided and left-sided targets, respec- 1998). However, studies show that inhibition of return
tively, 4-month-olds learned to anticipate the target by can be present in newborn infants, presumably reflect-
moving their eyes in the direction indicated by each ing the midbrain mechanisms that are dominant at
cue. This ability to direct the eyes toward a target based birth (Valenza, Simion, and Umilta, 1994). It is not un-
on an arbitrary association was present at 4 months, but usual to find functions that are present very early in
not at 2 months (Johnson, Posner, and Rothbart, 1991). infancy, based on a subcortical system, that seem to dis-
By 4 months, it has been demonstrated that the shift appear for a period, only to return later with cortical
of attention can be entirely covert, that is, without any maturation. For example, infants at birth favor a real
eye movement. In one such study, a brief flash was pre- face over a scrambled face (Johnson and Morton,
sented at a peripheral location where it captured the 1991). This difference in responding to real versus
infant's attention but did not cause an eye movement. scrambled faces is thought to be based on a primitive
Covert orienting was confirmed when infants demon- midbrain mechanism, but the preference disappears
strated a high probability and a reduced latency in mov- for a few months, only to return at 4 months, but now
ing their eyes to a subsequent strong target at the with more adult-like properties.
previously cued location, rather than to a location at a Much infant research has used duration of looking as
similar distance on the opposite side (Hood, 1993). a measure of interest or attention (Fantz, 1961). During
Hence, like adults, infants at 4 months of age can have the first 6 months of life, the average time that an infant
attention captured by a cue without actually executing looks at an object declines markedly (Colombo, 1995).
There is evidence that during sustained fixation, there
an eye movement.
are periods of close attention, but also periods in which
The similar development of the control of eye move-
interest, as measured, for example, by heart rate or la-
ment and covert orienting by peripheral and central
tency to disengage, declines (Richards, 1998; Richards
cues seems to reflect the common parietal mechanisms
and Hunter, 1998). Infants with long fixation duration
that have been shown to be involved in such attention
also tend to be slow to disengage (Frick, Colombo, and
shifts in adults by both lesion (Posner and Dehaene, Saxon, 1999).
1994) and PET studies (Corbetta et al., 1993). Between Infants exhibit a strong preference for novel objects
2 and 4 months, just at the time PET studies suggest a over familiar ones (Fantz, 1961; Rose et al., 1982). This
strong increase in metabolic processes in the parietal is particularly true when familiar objects have been ex-
lobe (Chugani, 1994), we find improvement in the in- posed many times (habituation) or when infants have a
fant's ability to shift attention to a peripheral target long period of exposure before being given a choice
even when strong attractors are present on the fovea. between objects (Rose et al., 1982). There has been
some dispute as to whether newborns favor novel stim-
NOVELTY One principle of visual system development uli over familiar ones, although some evidence suggests
is that subcortical visual mechanisms tend to develop that they, as well as older children, may look longer at
prior to cortical mechanisms (Bronson, 1974). In familiar objects under some circumstances (Barille,
adults, orienting away from a just-attended location re- Armstrong, and Bower, 1999; Ruff and Rothbart, 1996).
duces the probability of returning attention to that lo- In adults, mechanisms underlying preference for lo-
cation. This phenomenon, called inhibition of return cation versus object novelty seem to be based on quite
(IOR), was first demonstrated in reaction time studies different systems (Egly, Driver, and Rafal, 1994). We
with normal adults. It seems to reflect one of many in- compared the same forty 6-month-old infants in single
herent responses that favor novelty. In vision, IOR fa- trial novelty preference studies for locations and for ob-
vors searching in new locations rather than ones already jects (Posner et al., 1998), finding both forms of novelty
examined. preference in these infants. Because of the number of
358 COGNITION
For infants, the control of orienting is partially deter- in their learning abilities, and we found that the ability
mined by the extent to which caregivers present rele- to learn context-sensitive cues was positively related to
vant information. However, infants are not entirely caregivers' reports of their children's vocabulary devel-
helpless, and they are clearly also capable of soliciting opment and to other aspects of executive control (dis-
attention from adults (Stern, 1985). During the first cussed following). Hence, we believe that this type of
years of life, more direct control of attention is trans- context-dependent learning reflects functioning within
ferred from caregivers to their children. It seems pos- higher level attentional networks involved in executive
sible, then, that the coping mechanisms underlying control.
emotional self-regulation in early development extend
to encompass the control of cognition during later in- EXECUTIVE CONTROL Many psychologists agree with
fancy and childhood. Denckla (1996) that "the difference between the child
and adult resides in the unfolding of executive func-
LEARNING As described previously, there is consider- tions." In an early study of cognitive control, Diamond
able evidence that 3-4-month-old infants can learn to (1991) found evidence that 9-12-month-old children
anticipate the locations of visual stimuli by moving their proceed through a series of stages in resolving conflict
eyes to them in advance of their presentation (Canfield between reaching directly and reaching along the line
et al., 1997; Haith, Hazan, and Goodman, 1988; John- of sight in order to retrieve an object in a box. At 9
son, Posner, and Rothbart, 1991). In our studies (Clo- months of age, the line of sight completely dominates
hessy, Posner, and Rothbart, in press), 4-month-olds the nature of the child's response. Even if the child's
learned a sequence of three successive spatial locations hand touches the toy through the open side of the box,
arranged in a triangle; presentation of a stimulus at one if it is not in line with the side where the child is looking,
location predicted unambiguously the stimulus presen- the infant will withdraw the hand and reach along the
tation at the next location and so on. This skill may line of sight, striking the closed side. Three months
depend upon maturation of basal ganglia and/or pa- later, infants are able to look at a closed side but reach
rietal structures, both of which are involved in the through the open end to retrieve the toy (see Diamond,
implicit learning of sequences by adults (Grafton, this volume).
Hazeltine, and Ivry, 1995). Like infants, adults can learn However, being able to reach for a target indepen-
sequences of spatial locations implicitly when each lo- dent of the line of sight is a very limited form of conflict
cation is invariably associated with another location resolution. Gerstadt, Hong, and Diamond (1994) stud-
(e.g., locations 13241324). This learning occurs even ied verbal conflict using a modification of the Stroop
when the adult is distracted with a secondary task paradigm in children as young as 3Vs years. Two cards
known to occupy focal attention (Curran and Keele, were prepared to suggest day and night to the children.
1993). In adults, the implicit form of skill learning Children in the conflict condition were instructed to
seems also to rely mainly upon subcortical structures. say "day" to the moon and "night" to the sun card. Chil-
Children at about 18 months of age, but not earlier, dren in the control condition said either day or night
also show the ability to learn context-sensitive se- to a checkerboard or ribbon card. At every age, accu-
quences (Clohessy, Posner, and Rothbart, in press). racy scores were significantly lower for children in the
Context-sensitive sequences (e.g., 1213) are ones in conflict versus control conditions. Other efforts have
which an association (e.g., where to go after location been made (for example, using the Wisconsin Card
1), is ambiguous unless one considers information Sort task; Zelazo, Reznick, and Pinon, 1995) to study
other than the immediately preceding item. In adults, children as young as 31 months on inhibitory control
when all the associations are ambiguous (e.g., 123213), tasks; little evidence of successful inhibitory control be-
learning appears to require access to the kind of higher fore 3 years of age has been found.
level "executive" attention that is needed to resolve con- According to this analysis, a more direct measure of
flict (Curran and Keele, 1993). Although adults can the development of executive attention might be re-
learn unambiguous sequences in the presence of dis- flected in the ability to resolve conflict between simul-
tractors, adults are not similarly able to learn context- taneous stimulus events as in the Stroop effect. Since 2-
sensitive sequences (e.g., locations 123213), in which and 3-year-old children do not read, we reasoned that
each association is ambiguous in that a given location the use of basic visual dimensions of location and iden-
predicts more than one subsequent location, in the tity might be the most appropriate way to study the early
presence of distractors. Consistent with the increased resolution of conflict. The variant of the Stroop effect
difficulty that adults have in learning context-sensitive we designed to be appropriate for ages 2-3 years in-
sequences, individual children showed wide differences volved presenting a picture depicting a simple object
360 COGNITION
1982; Strelau, 1983). According to many of these the- decrease the influence of negative affectivity by shifting
orists, individuals were at the mercy of their dispositions attention away from the negative cues related to anger.
to approach or avoid a situation or stimulus, given re- Empathy is also strongly related to effortful control,
ward or punishment cues. More extraverted individuals with children high in effortful control showing greater
were expected to be sensitive to reward and to show empathy (Rothbart, Ahadi, and Hershey, 1994). Effort-
tendencies to rapid approach; more fearful or intro- ful control may support empathy by allowing the indi-
verted individuals, sensitive to punishment, were ex- vidual to attend to the thoughts and feelings of another
pected to show inhibition or withdrawal from without becoming overwhelmed by their own distress.
excitement (Gray, 1987). Similarly, guilt/shame in 6- and 7-year-olds is positively
These frameworks may neglect important modera- related to effortful control and negative affectivity
tors of approach and avoidance behaviors. For example, (Rothbart, Ahadi, and Hershey, 1994). Negative affec-
systems of effortful control allow approach to situations tivity may contribute to guilt by providing the individual
in the face of immediate cues for punishment, and with strong internal cues of discomfort, thereby increas-
avoidance of situations in the face of immediate cues ing the probability that the cause of these feelings is
for reward. In this sense, the programming of effortful attributed to an internal rather than external cause
control is critical to socialization, which often demands (Dienstbier, 1984). Effortful control may contribute fur-
that behavior proceed in a manner that is counter to ther by allowing the flexibility needed to relate these
reactive tendencies. Indeed, the ability to rise above negative feelings of responsibility to one's own specific
prepotent emotional drives may underlie human con- actions and to negative consequences for another per-
scientiousness and empathy. Consistent with this hy- son (Derryberry and Reed, 1994, 1996).
pothesis, the laboratory work of Kochanska and her
CONCLUSIONS To summarize, the neural system re-
associates indicates that the development of conscience
lated to effortful control shows an important develop-
is related to temperamental individual differences in
ment between 2 and 4 years, but it continues to develop
effortful control (Kochanska, Murray, and Coy, 1997;
during childhood and into adolescence, allowing more
Kochanska, Murray, and Harlan, 1999). Other research
sophisticated forms of self-regulation based on verbal
suggests some stability of executive attention during information, representations of the self, and projec-
childhood. In Mischel's work, for example, the number tions concerning the future. These findings suggest the
of seconds delayed by preschool children while waiting importance of effortful control to the child's emotional,
for physically present rewards predicted parent- cognitive, and social development. Without under-
reported attendveness and ability to concentrate as well estimating environmental influences, underlying tem-
as regulation of negative affect when the children were perament systems may serve a central role in the
adolescents (Mischel, 1983; Shoda, Mischel, and Peake, self-organization of personality (Derryberry and Reed,
1990). 1994, 1996). This is particularly evident in the functions
Questionnaire studies of 6- and 7-year-olds have of attention, which serve to select and coordinate the
found an effortful control factor to be defined in terms most important information, contribute to the storage
of scales measuring attentional focusing, inhibitory of this information in memory, and provide an impor-
control, low-intensity pleasure, and perceptual sensitiv- tant regulatory function.
ity (Rothbart et al., submitted). Effortful control is con-
sistently negatively related to a negative affectivity ACKNOWLEDGMENT This research was supported by NIMH
temperament factor. This negative relation is in keeping Grant RO1-43361 to the University of Oregon.
with the notion that attentional skill may help attenuate
negative affect. An interesting example involves the
negative relation between effortful control and aggres- REFERENCES
sion. Aggression relates positively to surgency and to ABDULLAEV, Y. G., N. P. BECHTEREVA, and K. V. MELNICHUK,
negative affectivity, especially anger (Rothbart, Ahadi, 1998. Neuronal activity of caudate nucleus and prefrontal
cortex in cognitive tasks. Behav. Brain Res. 97:159—177.
and Hershey, 1994). Effortful control is negatively re- BADDELEY, A., 1992. Is working memory working? The fif-
lated to aggression, but it makes no unique contribu- teenth Bartlett lecture. Quart. J. Exp. Psychol. 44A:1-31.
tion to it, and effortful control may regulate aggression BARILLE, M., E. S. ARMSTRONG, and T. G. R. BOWER, 1999.
indirectly by controlling reactive tendencies underlying Novelty and frequency as determinants of newborn prefer-
ence. Dev. Sci. 2:47-52.
surgency and negative affectivity. For example, children BLASS, E., 1992. Linking developmental and psychobiological
high in effortful control may be able to direct attention research. SRCD Newsletter. Chicago: University of Chicago
away from the rewarding aspects of aggression, or to Press.
362 COGNITION
KOCHANSKA, G., K. T. MURRAY, and E. HARLAN, 2000. Effort- ROSE, S. A., A. W. GOTTFRIED, P. MELLY-CARMINAR, and W. H.
ful control in early childhood: Continuity and change, BRIDGER, 1982. Familiarity and novelty preferences in in-
antecedents, and implications for social development. Dev. fant recognition memory: Implications for information pro-
Psychol. 36:220-232. cessing. Dev. Psychol 18:704-713.
LABERGE, D. H., 1995. AttentionalProcessing. Cambridge Mass.: ROTHBART, M. K., 1989. Temperament and development. In
Harvard University Press. Temperament in Childhood, G. A. Kohnstamm, J. E. Bates, and
LURIA, A. R., 1973. The Working Brain. New York: Basic Books. M. K. Rothbart, eds. New York: John Wiley and Sons, pp.
MARROCCO, R. T., and M. C. DAVIDSON, 1998. Neurochemistry 187-247.
of attention. In The Attentive Brain, R. Parasuraman, ed. ROTHBART, M. K., S. A. AHADI, and D. E. EVANS, 2000. Tem-
Cambridge, Mass.: MIT Press, pp. 35-50. perament and personality: Origins and outcomes.J. Person-
MISCHEL, W., 1983. Delay of gratification as process and as ality Soc. Psychol. 78:122-135.
person variable in development. In Human Development: An ROTHBART, M. K., S. A. AHADI, and K. L. HERSHEY, 1994. Tem-
Interactional Perspective, D. Magnusson and V. P. Allen, eds. perament and social behavior in childhood. Merrill-Palmer
New York: Academic Press, pp. 149-165. Quart. 40:21-39.
NORMAN, D. A., and T. SHALLICE, 1986. Attention to action: ROTHBART, M. K., S. A. AHADI, K. L. HERSHEY, and P. FISHER,
Willed and automatic control of behavior. In Consciousness submitted. Investigations of temperament at 3-7 Years: The
and Self-Regulation, R. J. Davidson, G. E. Schwartz, and Children's Behavior Questionnaire.
D. Shapiro, eds. New York: Plenum Press, pp. 1-18. ROTHBART, M. K., and J. E. BATES, 1998. Temperament. In
POSNER, M. I., 1980. Orienting of attention. The 7th Sir F. C. Handbook of Child Psychology, Vol. 3, W. S. E. Damon and
Bartlett Lecture. Quart.J. Exp. Psychol. 32:3-25. N. V. E. Eisenberg, eds. Social, Emotional and Personality De-
POSNER, M. I., and S. DEHAENE, 1994. Attentional Networks. velopment, 5th ed. New York: Wiley, pp. 105-176.
Trends Neurosci. 7:75-79. ROTHBART, M. K., D. DERRYBERRY, and K. HERSHEY, 2000.
POSNER, M. I., and G. J. DIGIROLAMO, 1998. Executive atten- Stability of temperament in childhood: Laboratory infant
tion: Conflict, target detection and cognitive control. In The assessment to parent report at seven years. In Temperament
Attentive Brain, R. Parasuraman, ed. Cambridge: MIT Press, and Personality Development Across the Life Span, V. J. Molfese
pp. 401-423. and D. L. Molfese, eds. Mahwah, N. J.: Lawrence Erlbaum,
POSNER, M. I., and C. D. GILBERT, 1999. Attention and pri- pp. 85-119.
mary visual cortex. Proc. Natl. Acad. Sci. USA 96(6):2585- RUFF, H. A., and M. K. ROTHBART, 1996. Attention in Early De-
2587. velopment: Themes and Variations. New York: Oxford Univer-
POSNER, M. I., and S. E. PETERSEN, 1990. The attention system sity Press.
of the human brain. Annu. Rev. Neurosci. 13:25-42. SANDER, L. W., 1962. Issues in early mother-child interaction.
POSNER, M. I., R. D. RAFAL, L. CHOATE, and J. VAUGHAN, 1985. J Amer. Acad. Child Psychiatry 1:141-166.
Inhibition of return: Neural mechanisms and function. SHODA, Y., W. MISCHEL, and P. K, PEAKE, 1990. Predicting ad-
Cogn. Neuropsychol. 2:211-228. olescent cognitive and self-regulatory competencies from
POSNER, M. L, and M. E. RAICHLE, 1994. Images of Mind. New preschool delay of gratification: Identifying diagnostic con-
York: Scientific American Library. ditions. Dev. Psychol. 26:978-986.
POSNER, M. L, and M. K. ROTHBART, 1998. Attention, self- SMITH, E. E., and J. JONIDES, 1995. Working memory in hu-
regulation and consciousness. Phil. Trans. R. Soc. Lond. [B] mans: Neuropsychological evidence. In The Cognitive Neu-
353:1915-1927. rosciences, M. S. Gazzaniga, ed. Cambridge, Mass.: MIT Press,
POSNER, M. L, M. K. ROTHBART, L. THOMAS-THRAPP, and pp. 1009-1020.
G. GERARDI, 1998. Development of orienting to locations SOKOLOV, Y. N., 1963. Perception and the Conditioned Reflex. New
and objects. In Visual Attention, R. Wright, ed. New York: York: Macmillan.
Oxford University Press, pp. 269-288. STERN, D. N., 1985. The Interpersonal World of the Infant. New
RICHARDS, J. E. (ED.), 1998. Cognitive Neuroscience of Attention: York: Basic Books.
A Development Perspective. Mahway, N.J.: Erlbaum. STRELAU, J., 1983. Temperament, Personality, Activity. New York:
RICHARDS, J. E., and S. K. HUNTER, 1998. Attention and eye Academic Press.
movements in young infants: Neural control and develop- VALENZA, E., F. SIMION, and C. UMILTA, 1994. Inhibition of
ment. In Cognitive Neuroscience of Attention: A Developmental return in newborn infants. Infant Behav. Dev. 17:293-302.
Perspective, J. E. Richards, ed. Mahway, N.J.: Erlbaum, pp. WELLS, A., and G. MATTHEWS, 1994. Attention and Emotion: A
131-162. Clinical Perspective. Hillsdale, N.J.: Erlbaum.
ROBERTSON, I. H., M. JEHKONEN, L. PIZZAMIGLIO, A. SHIEL, WOLFF, P. H., 1987. The Development of Behavioral States and the
and E. WEBER, 1977. Auditory sustained attention is a Expressions of Emotion in Early Infancy. Chicago: University of
marker of unilateral spatial neglect. Neuropsychologia Chicago Press.
35:1527-1532. ZELAZO, P. D., J. S. REZNICK, and D. E. PINON, 1995. Response
ROBSON, K. S., and H. A. Moss, 1970. Patterns and determi- control and the execution of verbal rules. Dev. Psychol.
nants of maternal attachment.J. Pediatr. 77:976-985. 31:508-517.
ABSTRACT Despite considerable information on the capacity Our current understanding of the organization of
of infants to learn and remember under a variety of condi- memory has gradually evolved from reports of clinical
tions, little is known about the neural basis of memory pro- cases with circumscribed lesions and from psychological
cesses early in life. This chapter reviews studies in nonhuman
primates demonstrating that memory comprises multiple sys- testing, neurorecording, and neuroimaging of normal
tems that are subserved by different neural circuits developing and diseased brains. Together with neurobiological ex-
at different time points in early infancy. The procedural mem- periments in nonhuman primates, these studies have
ory system mediated by the striatum and cerebellum is present led to the notion that memory can be divided into
at birth and reaches functional maturity in the first postnatal
broad categories.
months in monkeys. In contrast, the declarative memory sys-
tem, mediated by the hippocampal formation, and the work- The distinction between procedural and declarative
ing memory system, mediated by the dorsolateral prefrontal memory emerged from the study of patients with dam-
cortex, have a more protracted development, reaching func- age to the medial temporal region (Squire, 1992). This
tional maturity around one year of age for the former and damage yields deficits in retention at long delays but
three years of age for the latter. Given the many similarities
not at short ones. In addition, the long-term memory
in the developmental time course of memory processes and
neural systems in humans and monkeys, the experimental loss affects the ability to recollect past events or knowl-
findings have important implication for our understanding of edge, generally labeled declarative memory, while an
the development of memory functions and their neural sub- array of learning and retention abilities, labeled pro-
strates in humans. cedural memory, remain intact (Squire and Knowlton,
1995). A similar dissociation was found through lesion
Studies aimed at investigating the neural bases of in-
studies in nonhuman primates. That is, damage to the
fant memory have burgeoned during the last two de-
medial temporal lobe produces a selective loss in rec-
cades and several excellent reviews have recently
ognizing objects when long, but not short, delays are
appeared in the literature (Cowan, 1997; Herschkowitz,
tested, while the acquisition and retention of visual dis-
Kagan, and Zilles, 1997; Nelson, 1995, 1997; Rovee-
Collier, 1997). This chapter focus on studies of non- crimination skills even for relatively long periods of
human primates that have provided insights into the 24 hours are spared (Mishkin and Appenzeller, 1987).
neurobiological bases of early memory development. Thus, long-term memory appears to include two inde-
Two areas of memory have been more investigated by pendent systems mediated by different neural sub-
developmental neurobehavioral studies in primates. strates. The declarative (Squire, 1992) or prepositional
One relates to the development of long-term memory, (Tulving, 1995) memory systems permit the acquisition
involving procedural and declarative memory pro- of facts and specific events and are indexed by memory
cesses, and the other is the development of short-term tests, such as recognition and recall, that require ex-
memory, involving working memory processes. plicit recollection. These memory systems are depen-
dent on the medial temporal lobe/diencephalic
structures and their interactions with cortical areas. In
JOCELYNE BACHEVALIER Department of Neurobiology and contrast, the procedural memory system mediates the
Anatomy, University of Texas Health Science Center, Houston, acquisition and retention of skilled performance in-
Texas. dexed by tasks in which memory is expressed implicitly
366 COGNITION
cessively over several trials until the animal selects the gressively from the occipital areas to the most anterior
rewarded cue on almost every trial. Monkeys 9-15 days temporal areas and appears to be completed around
old rapidly master black/white and left/right discrimi- 3-6 months postnatally in monkeys.
nations (Harlow, 1959) and at 15-25 days can discrim- In the neostriatum, the cellular components are in
inate between two patterns or forms (Harlow, 1959; place in early gestation (165 days in monkeys), and by
Harlow, Harlow, Rueping, and Mason, 1960; Zimmer- the end of gestation, cortical inputs begin to invade the
man, 1961; Zimmerman and Torrey, 1965). Further- neostriatum (Brand and Rakic, 1979; Goldman-Rakic,
more, the ability to solve more complex discrimination 1981). In the caudate nucleus and putamen, synaptic
tasks in which the animal has to learn a series of object- density continues to increase until the end of the first
discriminations presented concurrently also emerges postnatal month (Brand and Rakic, 1984) and changes
relatively early in life. Thus, by 3-4 months of age (the in neuronal and neuropil morphology are observed un-
earliest age tested), infant monkeys can learn, as effi- til 2-4 months postnatally (Cano, Pasik, and Pasik,
ciently as adult monkeys, a short list of object pairs with 1989; Difiglia, Pasik, and Pasik, 1980). Refinement
short intertrial intervals (Mahut and Zola, 1977) and within the striatum proceeds until the end of the first
even long lists of object pairs with extremely long inter- postnatal year when the striatal neurochemical mosaic
trial intervals (Bachevalier and Mishkin, 1984). These attains an adult pattern (Martin, Spicer, and Cork,
data suggest that the ability to acquire and retain 1992). Furthermore, histoanatomical studies indicate
perceptual-motor associative responses, and hence to that motor cortical areas are predominantly connected
form some types of procedural memory, is present in with subcortical structures at birth, and thereafter, in-
the first few postnatal months in monkeys. Conse- tercortical connections progressively develop between
quently, some of the neural components of this memory the motor cortex and other cortical areas (Kemper,
system are clearly functional early in life to enable adult Caveness, and Yakovlev, 1973). In fact, the corticospinal
performance on procedural memory tasks. projections in the macaque monkey mature gradually
over a period of at least 11 months, long after simple
MATURATION OF NEURAL CIRCUITS MEDIATING PROCE- measures of dexterity show functional maturity. This re-
DURAL MEMORY The ability to acquire conditioned re- sult suggests that the later changes may contribute to
sponses has been shown to involve the cerebellum the improved speed and coordination of skilled motor
and associated brainstem structures (for a review, see tasks (Armand et al., 1994; Flament, Hall, and Lemon,
Woodruff-Pak and Thompson, 1988). In contrast, the 1992; Galea and Darian-Smith, 1995; Olivier et al.,
neural circuitry and structures that mediate visual dis- 1997).
crimination learning are still a matter of intense debate. Finally, in the cerebellum, neurogenesis in the cortex
The evidence gathered to date suggests that perceptual- and deep nuclei occurs early during gestation and neu-
motor learning (underlying discrimination tasks) re- ron differentiation continues through midgestation
quires the interactions between sensory and motor (Kornguth, Anderson, and Scott, 1967). At birth, the
cortical areas, with the participation of the striatum and morphology of the Purkinje cells and their axodendritic
cerebellum (Gaffan, 1996; Mishkin, Malamut, and synapses is similar in structure to the adult pattern
Bachevalier, 1984; Ungerleider, 1995; Wise, 1996). (Kornguth, Anderson, and Scott, 1967). Migration of
Thus, impairment in visual discrimination tasks follows granule cells continues during the first few postnatal
damage to the inferior temporal cortex, where object months (Rakic, 1971). Synapses from the Purkinje cells
features are processed (Buffalo et al., 1999; Mishkin, are already present prenatally on the soma and axons
1954; Phillips et al., 1988); some of the striatal regions, of deep nuclear neurons, and synapses from climbing
to which this visual cortex projects (Buerger, Gross, and fibers of the inferior olive onto Purkinje cells and those
Rocha-Miranda, 1974; Divac, Rosvold, and Szwarcbart, from mossy fibers onto granule cells occur approxi-
1967; Wang, Aigner, and Mishkin, 1990); or the white mately at midgestation (Kornguth, Anderson, and
matter of the temporal stem, which presumably inter- Scott, 1968). Finally, at birth, all Purkinje cells have re-
rupts connections between the visual cortical areas and ceived synaptic contacts (Levitt et al., 1984). Neuro-
the striatum (Horel, 1978; Zola-Morgan, Squire, and chemically, abundant NGF receptors are present on
Mishkin, 1982). Purkinje cells, granule cells, and neurons of the deep
Both metabolic (Bachevalier, Hagger, and Mishkin, nuclei in fetal cerebellum, but they decrease in the neo-
1991) and electrophysiological (Rodman, 1994) studies natal period (Schatteman et al., 1988). Similarly, many
have indicated that functional maturity of the ventral cells in the cerebellum are immunoreactive to neuro-
visual cortical areas, which provide visual inputs to mo- peptides early in gestation, but this immunoreactivity
tor cortical areas and to the striatum, proceeds pro- decreases significantly during the first postnatal months
368 COGNITION
(Gunderson and Swartz, 1985). Furthermore, prefer- pocampal formation emerges very early in life. These
ence for novelty is abolished in monkeys that have re- findings are supported by anatomical studies indicating
ceived early or late lesions to either the medial temporal that almost all anatomical components of the hippo-
lobe, including the amygdala, hippocampus, and adja- campal formation appear to be present at birth in mon-
cent cortical areas (Bachevalier, Brickson, and Hagger, keys. Thus, neurogenesis in the monkey hippocampus
1993) or to the hippocampal formation (Clark et al., begins early in gestation (Rakic and Nowakowski, 1981;
1996; Alvarado and Bachevalier, 2000; Pascalis and Seress and Ribak, 1995a,b), and at birth, all pyramidal
Bachevalier, 1999). These data imply not only that rec- cells of the Ammon fields and almost all granule cells
ognition memory is an early developing process, but of the dentate gyrus have completed their migration.
also that the hippocampal formation is functional early Despite this early maturation, refinement of hippocam-
in life to support this ability. pal circuitry continues well into postnatal life. Thus, the
The other behavioral task that has been used to assess adult-like ultrastructural features of the pyramidal cells
recognition memory in infant monkeys is a problem- is not achieved until the end of the third postnatal
solving task, the delayed nonmatching-to-sample month in the monkey, and the dentate cells continue
(DNMS) task. In the DNMS task, the animal must first to establish synaptic contacts until the end of the first
learn that the rule is to choose on every trial a novel postnatal year (Seress and Ribak, 1995a,b). Although
object paired with another object seen just a few sec- there have been no metabolic or electrophysiological
onds earlier (for additional details, see Overman and studies to assess the functional status of the hippocam-
Bachevalier, this volume). After learning the DNMS pal formation in early infancy, the anatomical findings
rule with a single object and a short delay, the animal suggest that at birth the hippocampus has reached a
is given a memory test in which its recognition ability is state of functional maturation sufficient to support
assessed further with longer delays and also with longer spontaneous recognition memory processes. However,
lists of to-be-remembered objects. The results indicate this state of functional maturity may not be sufficient to
that it is not before 4 months of age that monkeys begin support adult-level visual recognition performance.
to learn the DNMS rule, and it is not before 2 years of The hippocampus needs to receive visual information
age that they learn it with adult proficiency. In addition, about objects, which, in the adult, is provided by higher-
as compared to adult monkeys, immature monkeys are order visual temporal cortical areas. Anatomical con-
deficient in performing the memory test with extended nections between area TE, the cortical areas on the
delays and lists (Bachevalier and Mishkin, 1984). Be- parahippocampal gyrus, and the hippocampus do in-
cause recognition memory is an early developing pro- deed exist in the first week of life in monkeys (Webster,
cess, it is apparent that the difficulty young normal Ungerleider, and Bachevalier, 1991a). However, area
infant monkeys show in learning DNMS cannot be due TE is not fully functional before the fourth postnatal
to an absence of the adult preference for novelty or month (Bachevalier, Hagger, and Mishkin, 1991; Rod-
even to a deficiency in recognizing stimuli, which the man, 1994) and neurogenesis in this cortical area pro-
DNMS task exploits. Instead, the results suggest that the ceeds well into adulthood (Gould et al., 1999). Hence,
difficulty is due to a deficiency in rule learning, and to support visual recognition, the hippocampus must
particularly a difficulty in associating the reward in the interact with visual areas other than area TE. One pos-
well with the abstract quality (e.g., novelty) of the object sible pathway that could support visual recognition pro-
(Bachevalier, 1990; Bachevalier, Brickson, and Hagger, cesses at birth is a pathway linking visual cortical areas
1993; Diamond, 1990a; Malkova et al., in press; see also located earlier in the ventral visual pathway with the
Overman and Bachevalier, this volume). Because per- hippocampal formation. The existence of such tran-
formance on the DNMS task is not impaired by neo- sient connections has already been demonstrated, and
natal or late damage to the hippocampal formation these connections are known to retract progressively be-
(Bachevalier, Beauregard, and Alvarado, 1999; Murray tween the third and sixth postnatal months (Webster,
and Mishkin, 1998), the data indicate that the difficulty Ungerleider, and Bachevalier, 1991a,b). An alternative
in DNMS rule learning is due to immaturity within possibility is that early visual recognition processes are
structures of a neural circuit lying outside of the mediated by subcortical visual systems. Based on data
hippocampus. gathered in humans, it has been proposed that the new-
born visual system, unlike the adult visual system, is
MATURATION OF NEURAL CIRCUITS MEDIATING REC- largely controlled by subcortical neural structures, and
OGNITION MEMORY Taken together, the findings with that the visual cortical areas begin to play a more im-
the visual recognition tasks indicate that recognition portant role in visual functions after the third postnatal
memory taxed by the VPC and mediated by the hip- month (Atkinson, 1979; Bronson, 1974; Held, 1985;
370 COGNITION
old and 1-year-old monkeys are impaired on bicondi- monkey (Eckenhoff and Rakic, 1988; Rakic and Nowa-
tional discrimination learning as compared to adults kowski, 1981). Finally, synaptogenesis in the hippocam-
(Killiany and Mahut, 1990). Furthermore, the ability to pal formation increases, and a remodeling of the
learn transverse patterning emerges around 1 year of synaptic contacts continues, until the end of the first
age, although complete adult proficiency is not reached year of life (for review, see Alvarado and Bachevalier,
before 2 years of age (Malkova et al., 1999). Finally, 2000; Seress and Ribak, 1995a,b). This remodeling and
adult-like performance on the oddity task is not at- refinement of synaptic interactions within the hippo-
tained before 3-4 years of age (Harlow, 1959). There campal formation also implies that the functional cross-
have been no studies that have investigated the devel- talk between the hippocampal formation and other
opment of spatial relational memory abilities in neocortical areas may not reach functional maturity un-
monkeys. til the end of the first postnatal year in the monkey,
corresponding to the fourth year in humans.
MATURATION OF THE NEURAL CIRCUIT MEDIATING RE- Given all the evidence gathered to date, it is likely
LATIONAL LEARNING Very little is known about the that while interactions between the hippocampal for-
neural circuitry subserving relational learning. One mation and subcortical and allocortical areas (peri-
neural structure that seems crucial for this learning pro- rhinal and entorhinal cortex) may be functional early
cess is the hippocampal formation. Thus, selective dam- in life to support some form of recognition memory,
age to the hippocampal region in adult monkeys is the interactions between the hippocampal formation
known to impair performance on the transverse pat- and the neocortical areas on the inferior temporal cor-
terning (Alvarado, Bachevalier, and Mishkin, 1998) and tex and prefrontal cortex may develop more progres-
spatial memory tasks (Mahut, 1972; Mahut and Zola, sively during the first year postnatally to account for the
1973; Murray, Baxter, and Gaffan, 1998; Parkinson, protracted appearance of adult proficiency on the re-
Murray, and Mishkin, 1988). Although there are, as yet, lational memory tasks (Alvarado and Bachevalier, 2000;
no data on the effects of hippocampal lesions on bicon- see also Nelson, 1995, 1997). To summarize, as for pro-
ditional discrimination or oddity tasks in monkeys, the cedural memory, the development of declarative mem-
findings indicate a critical role of the hippocampal for- ory appears to comprise several subsystems (see
mation for normal performance on relational tasks. In Mishkin, Vargha-Khadem, and Gadian, 1998 for a simi-
contrast, lesions of the ventrolateral portions of the pre- lar conclusion), some of which emerge earlier than oth-
frontal cortex (i.e., inferior convexity and orbitofrontal ers during ontogenesis.
cortex), which produce a severe deficit in recognition
memory, do not affect relational learning (Alvarado et Development of working memory
al., 1997). These data from lesion studies, together with
the evidence regarding the protracted development of The active maintenance of short-term memories is com-
relational learning in monkeys, imply that the hippo- monly termed "working memory" (Baddeley, 1987;
campal formation is not fully functional during approx- Goldman-Rakic, 1987). This form of memory has typi-
imately the first year of age in monkeys to efficiently cally been studied in monkeys with the use of delayed
support this form of memory. This conclusion seems to response (DR) or delayed alternation (DA) tasks (Ja-
contradict the early functional maturation of the hip- cobsen, 1936). In both tasks, the monkey is required to
pocampus to support recognition memory. One possi- maintain a sensory cue in memory for a short delay,
bility is that mediation of relational learning may then use it to guide a behavioral response. In the classic
require the participation of the hippocampus and its DR task, two lateral wells of a test tray are covered with
interactions with higher-association cortical areas, such identical plaques, and the experimenter shows the
as the temporal and prefrontal cortical areas. There is monkey that a reward is being hidden under one of the
already evidence indicating that the maturation of the plaques. After a delay of a few seconds, the animal must
hippocampus progresses until adolescence in monkeys. select the plaque under which the reward has been hid-
Thus, although neurogenesis within the hippocampus den. In the DA task, the reward alternates from trial to
is almost complete in all the Ammon fields at birth trial from the right well to left well, both covered with
(Rakic and Nowakowski, 1981), the ultrastructural fea- identical plaques. On any given trial, the monkey must
tures of the neurons do not achieve the adult-like pat- remember under which plaque the reward was hidden
tern until the end of the first 3 postnatal months of life during the previous trial in order to subsequently select
(Seress and Ribak, 1995a,b). In addition, neurogenesis the correct one.
in the dentate gyrus continues well into postnatal life The role of the dorsolateral prefrontal cortex in the
and tapers off between 4 and 6 months of age in the mediation of working memory is now well established
372 COGNITION
member which object he had picked on the previous reaches a maximum rate of accumulation around birth
trial and avoid it in favor of the other two objects. There and is maintained at a constant level between 2 months
have been no developmental studies to assess the mat- to 3 years of age (Bourgeois, Goldman-Rakic, and Rakic,
uration of abilities to solve these two working memory 1994). Finally, synapses are eliminated until 20 years of
tasks in monkeys, although there are studies in human age, a process referred to as "synaptic pruning." Inter-
populations (Luciana and Nelson, 1998). However, in estingly, during synaptic pruning most of the dendrites
a similar task, the Hamilton Perseverance Test, Harlow eliminated are from the dendritic spines of the pyra-
(1959) showed a protracted maturation in monkeys. In midal neurons and not those of the dendritic shafts.
the Hamilton Perseverance Test, the animal is pre- Similarly, dopamine receptors appear to reach a peak
sented on each trial with four identical boxes mounted around 2-4 months of age and then decline to reach
with a spring-loaded lid that automatically closes as soon adult levels by 3 years of age (Lidow, Goldman-Rakic,
as the lid is opened. Only one of the four boxes, aligned Rakic, 1991; Lidow and Rakic, 1992). Thus, although
in a row in front of the monkey, contains the food re- the gross morphological development of the prefrontal
ward. The rewarded box changes in a random manner cortex progresses rapidly prenatally and postnatally,
from trial to trial, with the provision that the same box substantial changes in synaptic connectivity can be ob-
is never rewarded twice in succession. On each trial, the served well into adolescence (for review, see Lewis,
monkey is allowed four responses to find the reward, 1997). Similar findings were obtained with electrophys-
and an error is defined as any return to an unrewarded iological recordings of prefrontal neurons in infant
box after it had been visited once. Thus, in this task too, monkeys. Thus, the percentage of prefrontal neurons
the monkey needs to keep track of the boxes visited that exhibit delay period activity doubles between 12
during a given trial to avoid making unrewarded and 36 months of age, suggesting that developmental
choices. Adult monkeys 4-5 years old rapidly develop a changes in prefrontal circuitry facilitate the recruit-
strategy to avoid making unrewarded choices during ment of these neurons to support working memory (Al-
the task. However, 1- and 2.5-year-old monkeys make exander, 1982).
many more errors on this task and do not appear to use This progressive maturation of the prefrontal cortex
a strategy. Although the effects of lesions of the dorso- has also been demonstrated by studying the effects of
lateral prefrontal cortex on that task have never been early damage to this cortical area. Longitudinal studies
investigated, the protracted maturation on this task sug- investigating the consequences of early damage to the
gests that adult performance for some working memory prefrontal cortex (for review, see Goldman-Rakic et al.,
tasks (search task, self-ordered, and Hamilton Persever- 1983) indicated that DR performance of monkeys op-
ance tasks) may emerge later in life than others (de- erated on around 1.5-2.5 months of age did not differ
layed response and A-not B), a finding similar to that from that of control animals when tested at 1 year of
found for the development of hippocampal-dependent age, but was significantly worse when the same monkeys
memory functions. were retested at 2.5 years. These findings indicate that
the dorsolateral prefrontal cortex is not functionally
MATURATION OF DORSOLATERAL PREFRONTAL CORTEX committed to DR performance before 2-3 years of age
Neurogenesis of pyramidal neurons in the prefrontal in the monkey. Comparable results were found with re-
cortex occurs prenatally in the monkey, and, by the end versible cooling of the prefrontal cortex (Alexander
of gestation, all neurons in the prefrontal cortex have and Goldman, 1978). Cooling of the dorsolateral pre-
reached their final laminar distribution (Schwartz, frontal cortex at a temperature that causes 21-25% dec-
Rakic, and Goldman-Rakic, 1991). However, new neu- rements in DR performance in monkeys of 3 years of
rons reach this cortical area even in adulthood (Gould age, produces a decrement of only 7—8% in monkeys
et al., 1999). Local circuitry GABA neurons are also gen- of 2 years of age and no detectable changes in monkeys
erated before birth; but subsets of GABAergic neurons of 1 year of age.
become functionally mature at different times during For the DA task, performance was reduced by local
development, and for some of them, maturation con- hypothermia as early as 8.5 months of age, but the re-
tinues during the first few months after birth (for re- duction in performance was maximal only when hypo-
view, see Lewis, 1997). Afferent projection systems also thermia was induced at 3 years of age (Alexander and
arrive prenatally in the prefrontal cortex, although cal- Goldman, 1978). These findings suggest that DR and
losal fibers reach a peak at birth and are eliminated DA have a different developmental time course and that
during the first 3 postnatal months (LaMantia and performance of prepubertal monkeys on the DR task
Rakic, 1990; Schwartz and Goldman-Rakic, 1984). Fur- before complete maturation of the dorsolateral pre-
thermore synaptogenesis within the prefrontal cortex frontal cortex could be mediated by other subcortical
374 COGNITION
TABLE 25.1 ALVARADO, M. C., J. BACHEVALIER, and M. MISHKIN, 1998.
Functional development of multiple memory systems in Neurotoxic lesions of the hippocampal formation impair
monkeys and humans monkeys' acquisition of the transverse patterning problem.
Soc. Neurosci. Abstr. 24:928.
Memory Systems/Tasks Monkeys Humans ALVARADO, M. C., L. MALKOVA, C. K. LEX, M. MISHKIN, and
J. BACHEVALIER, 1997. Effects of early lesions of the inferior
Procedural Memory convexity and orbital prefrontal cortices on relational mem-
Conditioned responses 1 week 1-2 daysa ory in rhesus monkeys. Soc. Neurosci. Abstr. 23:498.
Visual discrimination tasks: ARMAND, J., S. A. EDGLEY, R. N. LEMON, and E. OLIVIER, 1994.
Two pairs 2—3 weeks Protracted postnatal development of corticospinal projec-
Concurrent multiple pairs 3-4 months 1—1.5 yearsa,b tions from the primary motor cortex to hand motoneurones
Declarative Memory in the macaque monkey. Exp. Brain Res. 101:178-182.
Recognition tasks: ATKINSON, J., 1979. Development of optokinetic nystagmus in
Visual paired — Comparison 2 weeks 3 daysa the human infant and monkey infant: An analog to devel-
Delayed nonmatching 2 years 4-5 yearsa,b opment in kittens. In NATO Advanced Study Institute Series,
Relational tasks: R. D. Freeman, ed. New York: Plenum Press.
Biconditional discrimination 2-3 years BACHEVALIER, J., 1990. Ontogenetic development of habit
and memory formation in primates. Ann. N.Y. Acad. Sci.
Transverse patterning 2-3 years 4-5 yearsc
Oddity 3-4 years 7 yearsb 608:457-477.
BACHEVALIER, J., M. BEAUREGARD, and M. C. ALVARADO, 1999.
Working Memory Long-term effects of neonatal damage to the hippocampal
A-not-B 2-4 months 8-12 monthsa formation and amygdaloid complex on object discrimina-
Spatial delayed response 2-4 months 8-12 months3 tion and object recognition in rhesus monkeys (Macaca mu-
Spatial delayed alternation 1-2 years latto). Behav. Neurosci. 113:1-25.
BACHEVALIER, J., M. BRICKSON, and C. HAGGER, 1993. Limbic-
Note: For visual functions 1 week of development in monkeys
dependent recognition memory in monkeys develops early
corresponds roughly to 1 month of development in humans.
in infancy. NeuroReport 4:77-80.
"Nelson, 1995; ''Overman, this volume;cRudy, Keith, and Geor-
BACHEVALIER, J., M. BRICKSON, C. HAGGER, and M. MISHKIN,
gen, 1993.
1990. Age and sex differences in the effects of selective tem-
poral lobe lesion on the formation of visual discrimination
habits in rhesus monkeys. Behav. Neurosci. 104:885-899.
the similarities or differences between the early and the BACHEVALIER, J., C. HAGGER, and M. MISHKIN, 1991. Func-
adult memory systems? Does the early memory system tional maturation of the occipitotemporal pathway in infant
change progressively with age, is it maintained while rhesus monkeys. In Alfred Benzon Symposium No. 31: Brain
other processes develop, or does it disappear to be re- Work and Mental Activity, Quantitative studies with radioactive
placed by new ones? Answers to these questions will not tracers, N. A. Lassen, D. H. Ingvar, M. E. Raichle, and L.
Friberg, eds. Copenhagen, Munksgaard, pp. 231-240.
only guide our search for neurobiological correlates un- BACHEVALIER, J., and M. MISHKIN, 1984. An early and late
derlying memory processes early in infancy, but will also developing systems for learning and retention in infant
shed light on the neurobiological correlates underlying monkeys. Behav. Neurosci. 98:770-778.
adult memory processes. BACHEVALIER, J., and M. MISHKIN, 1994. Effects of selective
neonatal temporal lobe lesions on visual recognition mem-
ory in rhesus monkeys.J. Neurosci. 14:2128-2139.
ACKNOWLEDGMENTS The work discussed here was sup- BACHEVALIER, J., L. G. UNGERLEIDER, B. O'NEILL, and D. P.
ported by NIMH grants (MH-54167 and MH-58846). FRIEDMAN, 1986. Regional distribution of [3H]naloxone
binding in the brain of a newborn rhesus monkey. Dev. Brain
Res. 25:302-308.
BADDELEY, A., 1987. Working Memory, Oxford, England: Clar-
REFERENCES endon Press.
ALEXANDER, G. E., 1982. Functional development of frontal BOURGEOIS, J. P., P. S. GOLDMAN-RAKIC, and P. RAKIC, 1994.
association cortex in monkeys: Behavioral and electrophys- Synaptogenesis in the prefrontal cortex of rhesus monkeys.
iological studies. Neurosci. Res. Prog. Bull. 20:471-479. Cereb. Cortex 4:78-96.
ALEXANDER, G. E., and P. S. GOLDMAN, 1978. Functional de- BRAND, S., and P. RAKIC, 1979. Genesis of the primate neo-
velopment of the dorsolateral prefrontal cortex: An analysis striatum: [3H]Thymidine autoradiographic analysis of the
utilizing reversible cryogenic depression. Brain Res. time of neuron origin in the rhesus monkey. Neuroscience
143:233-249. 4:767-778.
ALVAREZ, P., S. ZOLA-MORGAN, and L. R., SQUIRE, 1995. Dam- BRAND, S., and P. RAKIC, 1984. Cytodifferentiation and syn-
age limited to the hippocampal region produces long- aptogenesis in the neostriatum of fetal and neonatal rhesus
lasting memory impairment in monkeys. J. Neurosci. monkeys. Anat. Embryol. 169:21-34.
15:3796-3807. BRONSON, G., 1974. The postnatal growth of visual capacity.
ALVARADO, M. C., and J. BACHEVALIER, 2000. A protracted Child Dev. 45:873-890.
maturation of hippocampal memory functions in primates. BROZOVSKI, T, R. M. BROWN, H. E. ROSVOLD, and P. S. GOLD-
Learning Memory 7:244-256. MAN, 1979. Cognitive deficit caused by regional depletion
376 COGNITION
GUNDERSON, V. M., and K B. SWARTZ, 1985. Visual recognition LEVITT, P., P. RAKIC, P. DE CAMILLI, and P. GREENGARD, 1984.
in infant pigtailed macaques after 24-hour delay. Am.J. Pri- Emergence of cyclic guanosine 3':5'-monophosphate-
mal. 8:259-264. dependent protein kinase immunoreactivity in developing
GUNDERSON, V. M., and K. B. SWARTZ, 1986. Effects of famil- rhesus monkey cerebellum: Correlative immunocytochem-
iarization time on visual recognition memory in infant pig- ical and electron microscopic analysis. J. Neurosci. 4:2553-
tailed macaques (Macaco, nemestrina). Dev. Psychol. 22:477- 2564.
480. LEWIS, D. A., 1997. Development of the primate prefrontal
HARLOW, H. E, 1959. The development of learning in the cortex. In Neurodevelopment and Adult Psychopathology, M. S.
rhesus monkey. Am. Scientist Winter A59-479. Keshavan and R. M. Murray, eds. Cambridge, UK: Cam-
HARLOW, H. E, M. K. HARLOW, R. R. RUEPING, and W. A. MA- bridge University Press, pp. 12-30.
SON, 1960. Performance of infant rhesus monkeys on dis- LIDOW, M. S., P. S. GOLDMAN-RAKIC, and P. RAKIC, 1991. Syn-
crimination learning, delayed response, and discrimination chronized overproduction of neurotransmitter receptors in
learning set.J Comp. Physiol. Psychol. 53:113-121. diverse regions of the primate cerebral cortex. Proc. Natl.
HAYASHI, M., 1987. Ontogeny of glutamic acid decarboxylase, Acad. Sci. USA 88:10218-10221.
tyrosine hydroxylase, choline acetyltransferase, somato- LIDOW, M. S., and P. RAKIC, 1992. Scheduling of monoamin-
statin and substance P in monkey cerebellum. Dev. Brain Res. ergic neurotransmitter receptor expression in the primate
32:181-186. neocortex during postnatal development. Cereb. Cortex
HELD, R., 1985. Binocular vision: Behavioral and neuronal de- 2:401-416.
velopment. In Neonates Cognition: Beyond the Booming, Buzz- LUCIANA, M., and C. A. NELSON, 1998. The functional emer-
ing, Confusion, J. Mehler and R. Fox, eds. Hillsdale, N.J.: gence of prefrontally-guided working memory systems in
Lawrence Erlbaum, pp. 37-44. four- to eight-year-old children. Neuropsychologia 36:273-
HERSCHKOWITZ, N., J. KAGAN, and K. ZILLES, 1997. Neuro- 293.
biological bases of behavioral development in the first year. MAHUT, H., 1972. A selective spatial deficit in monkeys after
Neuropediatrics 28:296-306. transection of the fornix. Neuropsychologia 10:65-74.
HOREL, J. A., 1978. The neuroanatomy of amnesia: A critique MAHUT, H., and S. ZOLA, 1973. A non-modality specific im-
of the hippocampal memory hypothesis. Brain 101:403- pairment in spatial learning after fornix lesions in monkeys.
445. Neuropsychologia 11:255-269.
HOREL, J. A., D. E. PYTKO-JOYNER, M. L. VOYTKO, and K. SALS- MAHUT, H., and S. ZOLA, 1977. Ontogenetic time-table for the
BURY, 1987. The performance of visual tasks while segments development of three functions in infant macaques and the
of the inferotemporal cortex are suppressed by cold. Behav. effect of early hippocampal damage upon them. Soc. Neu-
Brain Res. 23:29-42. rosci. Abstr. 3:428.
JACOBSEN, C. E, 1936. Studies of cerebral function in primates. MALAMUT, B. L., R. C. SAUNDERS, and M. MISHKIN, 1984. Mon-
Comp. Psychol. Monogr. 13:1-68. keys with combined amygdalo-hippocampal lesions succeed
in object discrimination learning despite 24-hour intertrial
JOHNSON, M. H., 1993. Cortical maturation and perceptual de-
intervals. Behav. Neurosci. 98:759-769.
velopment of visual attention in early infancy. In Brain De-
MALKOVA, L., M. C. ALVARADO, E. L. PIXLEY, A. M. BELCHER,
velopment and Cognition: A Reader, M. H. Johnson, ed.
M. MISHKIN, andj. BACHEVALIER, 1999. Maturation of re-
Cambridge, Mass.: Blackwell, pp. 167-194.
lational memory processes in monkeys. Soc. Neurosci. Abstr.
KEMPER, T. L., W. E CAVENESS, and P. I. YAKOVLEV, 1973. The
25:88.
neuronographic and metric study of the dendritic arbours
MALKOVA, L., J. BACHEVALIER, M. WEBSTER, and M. MISHKIN,
of neurons in the motor cortex of Macaca mulatto at birth
in press. Effects of neonatal inferior prefrontal and medial
and at 24 months of age. Brain 96:765-782. temporal lesions on learning the rule for delayed non-
KILLIANY, R., and H. MAHUT, 1990. Hippocampectomy in in- matching-to-sample. Dev. Neuropsychol.
fant monkeys facilitates object reward association learning MALKOVA, L., M. MISHKIN, and J. BACHEVALIER, 1995. Long-
but not for conditional object-object associations. Soc. Neu- term effects of selective neonatal temporal lobe lesions on
rosci. Abstr. 16:847. learning and memory in monkeys. Behav. Neurosci. 109:212-
KORNGUTH, S. E., J. W. ANDERSON, and G. SCOTT, 1967. Ob- 226.
servations on the ultrastructure of the developing cerebel- MALKOVA, L., G. I. PIXLEY, M. J. WEBSTER, M. MISHKIN, and
lum of the Macaca mulatta.J. Comp. Neurol. 130:1-24. J. BACHEVALIER, 1998. The effects of early rhinal lesions on
KORNGUTH, S. E., J. W. ANDERSON, and G. SCOTT, 1968. The visual recognition memory in rhesus monkeys. Soc. Neurosci.
development of synaptic contacts in the cerebellum of Ma- Abstr. 24:1906, 1998.
caca mulatta. J. Comp. Neurol. 132:531-546. MARTIN, L. J., D. M. SPICER, and L. C. CORK, 1992. Postnatal
KOWALSKA, D., J. BACHEVALIER, and M. MISHKIN, 1991. The development of the striatal mosaic in rhesus monkey. Soc.
role of the inferior prefrontal convexity in performance of Neurosci. Abstr. 18:957.
delayed nonmatching-to-sample. Neuropsychologia 29:583- MEUNIER, M., J. BACHEVALIER, and M. MISHKIN, 1997. Effects
600. of orbital frontal and anterior cingulate lesions on object
KUBOTA, K., and H. NIKI, 1971. Prefrontal cortical activity and and spatial memory in rhesus monkeys. Neuropsychologia
delayed alternation performance in monkeys. /. Neurophy- 35:999-1015.
siol. 34:337-347. MEUNIER, M., J. BACHEVALIER, M. MISHKIN, and E. A. MURRAY,
LAMANTIA, A. S., and P. RAKIC, 1990. Axon overproduction 1993. Effects on visual recognition of combined and sepa-
and elimination in the corpus callosum of the developing rate ablations of the entorhinal and perirhinal cortex in
rhesus monkey.J. Neurosci. 10:2156-2175. rhesus monkeys.J Neurosci. 13:5418—5432.
378 COGNITION
SPENCE, K. W., 1952. The nature of response in discrimination WEISKRANTZ, L., L. MIHAILOVIC, and C. G. GROSS, 1962. Ef-
learning. Psychol. Rev. 59:89-93. fects of stimulation of frontal cortex and hippocampus on
SQUIRE, L. R., 1992. Declarative and nondeclarative memory: behavior in the monkey. Brain 85:487-504.
Multiple brain systems supporting learning and memory.J WILSON, F. A. W., S. P. O'SCALAIDHE, and P. S. GOLDMAN-
Cogn. Neurosci. 4:232-243. RAKIC, 1993. Dissociation of object and spatial processing
SQUIRE, L. R., and B. J. KNOWLTON, 1995. Memory, hippocam- domains in primate prefrontal cortex. Science 260:1955-
pus, and brain systems. In The CognitiveNeuwsciences, M. Gaz- 1957.
zaniga, ed. Cambridge, Mass.: MIT Press, pp. 825-837. WISE, K L., L. A. WISE, and R. R. ZIMMERMANN, 1973. Piage-
SQUIRE, L. R., and S. M. ZOLA, 1996. Structure and function tian object permanence in the infant rhesus monkey. Dev.
of declarative and nondeclarative memory systems. Proc. Psychol. 3:429-437.
Natl. Acad. Sd. USA 93:13515-13522. WISE, S. P., 1996. The role of the basal ganglia in procedural
SQUIRE, L. R., and S. M. ZOLA, 1998. Episodic memory, se- memory. Sem. Neurosci. 8:39-46.
mantic memory, and amnesia. Hippocampus 8:205-211. WOODRUFF-PAK, D. S., and R. F. THOMPSON, 1988. Cerebellar
STAMM, J. S., and S. C. ROSEN, 1969. Electrical stimulation and correlates of classical conditioning across the life span. In
steady potential shifts in prefrontal cortex during delayed Life-Span Development and Behavior, P. B. Baltes, R. M. Lerner,
response performance by monkeys. Acta Biol. Exp. 29:385- and D. M. Featherman, eds. Hillsdale, N.J.: Erlbaum, pp. 1-
399. 37.
SUZUKI, W. A., S. ZOLA-MORGAN, L. R. SQUIRE, and D. G. YAMASHITA, A., M. HAYASHI, K. SHIMIZU, and K. OSHIMA,
AMARAL, 1993. Lesions of the perirhinal and parahippocam- 1990. Neuropeptide-immunoreactive cells and fibers in the
pal cortices in the monkey produce long lasting memory developing primate cerebellum. Dev. Brain Res. 51:19-25.
impairments in the visual and tactual modalities.J. Neurosci. ZIMMERMAN, R. R., 1961. Analysis of discrimination learning
13:2430-2451. capacities in the infant rhesus monkey.J Comp. Physiol. Psy-
TULVING, E., 1995. Organization of memory: Quo vadis? In chol. 54:1-10.
The Cognitive Neuwsciences, M. S. Gazzaniga, ed. Cambridge, ZIMMERMAN, R. R., and C. C. TORREY, 1965. Ontogeny of
Mass.: MIT Press, pp. 839-847. learning. In Behavior of Nonhuman Primates: Modern Research
UNGERLEIDER, L. G., 1995. Functional brain imaging studies Trends, Vol. 2, A. M. Schrier, H. F. Harlow, and F. Stollnitz,
of cortical mechanisms for memory. Science 270:769-775. eds. New York: Academic Press, pp. 405-447.
WANG, J., T. AIGNER, and M. MISHKIN, 1990. Effects of neo- ZOLA-MORGAN, S., and L. R. SQUIRE, 1984. Preserved learning
striatal lesions on visual habit formation in rhesus monkeys. in monkeys with medial temporal lesions: Sparing of motor
Soc. Neurosci. Abstr. 16:617. and cognitive skills.J Neurosci. 4:1072-1085.
WEBSTER, M. J., L. G. UNGERLEIDER, and J. BACHEVALIER, ZOLA-MORGAN, S., L. R. SQUIRE, D. G. AMARAL, and W. A.
199la. Connections of inferior temporal areas TE and TEO SUZUKI, 1989. Lesions of the perirhinal and parahippocam-
with medial temporal-lobe structures in infant and adult pal cortex that spare the amygdala and hippocampal for-
monkeys./ Neurosci. 11:1095-1116. mation produce severe memory impairment. J Neurosci.
WEBSTER, M. J., L. G. UNGERLEIDER, and J. BACHEVALIER, 9:4355-4370.
1991b. Lesions of inferior temporal area TE in infant mon- ZOLA-MORGAN, S., L. R. SQUIRE, R. P. GLOWER, and N. L. REM-
keys alter cortico-amygdalar projections, NeuroReport2:769- PEL, 1993. Damage to the perirhinal cortex exacerbates
772. memory impairment following lesions of the hippocampal
WEBSTER, M. J., J. BACHEVALIER, and L. G. UNGERLEIDER, formation./ Neurosci. 13:251-265.
1995. Development and plasticity of visual memory circuits. ZOLA-MORGAN, S., L. R. SQUIRE, and M. MISHKIN, 1982. The
In Maturational Windows and Adult Cortical Plasticity, B.Julesz neuroanatomy of amnesia: Amygdala-hippocampus versus
and I. Kovacs, eds. Redwood City, Calif.: Addison-Wesley, pp. temporal stem. Sconce 218:1337-1339.
73-86. ZOLA-MORGAN, S., L. SQUIRE, and S. J. RAMUS, 1994. Severity
WEINSTEIN, J., R. C. SAUNDERS, and M. MISHKIN, 1988. of memory impairment in monkeys as a function of locus
Temporo-prefrontal interaction in rule learning by ma- and extent of damage within the medial temporal lobe
caques. Soc. Neurosci. Abstr. 14:1230. memory system. Hippocampus 4:483-495.
ABSTRACT The human face is typically one of babies' first generally for understanding brain and behavioral
sights following birth, and from this time they show an interest development.
in orienting to facelike patterns. Yet in spite of this impressive This chapter provides a description of developmental
early competency, face processing follows a protracted devel-
opmental course before becoming adultlike. This chapter de- changes in face processing together with a critical eval-
scribes the development of face processing abilities and uation of hypotheses regarding the neural and cogni-
outlines theories of the neural and cognitive mechanisms un- tive mechanisms underlying these changes. It is divided
derlying developmental change. It is concluded that the cor- into three sections, the first addressing general issues
tical specialization for face processing observed in adults is in the study efface processing, the second providing an
achieved through a gradual specialization of an initially more
general-purpose processing system.
overview of normative development during infancy and
childhood, and the third considering atypical develop-
The human face holds special significance as a pow- ment of face processing.
erful visual source of social information. Normally,
adults can, with seemingly little effort, quickly and ac- What's special about face recognition?
curately perceive the complex array of characteristics
encoded in the face, such as identity, emotion, and di- A fundamental question in the study of face processing
is whether, within the ventral visual object-processing
rection of gaze. Even newborns appear to know some-
pathway, there is further division into face and object
thing about the structure of the face; shortly after birth,
processing pathways. In other words, is the cortical face
they will move their eyes further to keep a moving pat-
processing system different, in terms of (1) anatomical
tern in sight if its elements are positioned in a facelike
localization and/or (2) perceptual-cognitive speciali-
arrangement (Johnson et al., 1991). These impressive
zation, from the cortical object processing system? It is
abilities make it easy to forget both the highly sophis-
no exaggeration to say that the majority of studies of
ticated mechanisms underlying adults' face processing
face processing in adults address this question, either
abilities and the protracted developmental pathway by
directly or indirectly. More studies than not have found
which the basic abilities of the newborn are trans-
differences between face and object processing on vari-
formed into the mature state. Studying this develop- ous neural and perceptual-cognitive measures (re-
mental process can provide a unique perspective on viewed in Farah et al., 1998; Nachson, 1995). Some
debates that are central to the study efface processing, investigators accept these results as evidence in favor of
such as the domain-specificity of the neural and cogni- cortical localization and specialization of face process-
tive mechanisms mediating face processing and the ing, while others argue that the differences can be at-
relative contributions of innate cortical organization tributed to confounding factors (e.g., differences in
and visual experience in the development of face pro- level of categorization for face and object tasks; see
cessing. Moreover, as these questions regarding cortical Damasio, Tranel, and Damasio, 1990; Gauthier et al.,
specialization are not unique to the topic of face pro- 1999). It is beyond the scope of this chapter to provide
cessing, this line of study also has implications more a comprehensive review of the adult literature on this
point; the purpose here is simply to briefly review as-
pects of face processing by adults that are proposed to
MICHELLE DE HAAN Cognitive Neuroscience Unit, Institute
of Child Health and Great Ormond Street Hospital, University be unique in order to provide a general background
College London Medical School, The Wolfson Centre, Lon- and to show how developmental studies can address
don, England. these issues.
382 COGNITION
mental approach can provide information regarding 1996) or equally (Kleiner and Banks, 1987; Mondloch
the role of experience in the emergence of the cortical et al., 1999) visible support the social hypothesis and
face processing system. For example, is a distinction be- cannot be explained by the sensory hypothesis. Since
tween face and object processing present at birth, or infants show this preferential orienting as little as hours
does it only develop following extensive visual experi- after birth, it is either a very quickly learned or a con-
ence with faces? This information is critical not only for genital representational bias.
understanding normative development of face process- What are the neural bases of infants' preferential ori-
ing skills but also for identifying the factors underlying enting to facelike stimuli in the visual environment?
developmental impairments in face processing skills. One hypothesis is that it is mediated by a subcortical
retinotectal pathway (Morton and Johnson, 1991). For
Normative development of face processing example, newborns only show the preference in con-
ditions to which the subcortical systems are sensitive
THE NEWBORN Studying the newborn's ability to per- (when stimuli are moving and are in the peripheral vi-
ceive and remember faces is of considerable interest sual field but not when they are in the central visual
because it allows one to assess the representational bi- field). Further experimental evidence in support of this
ases the infant is born with that might guide subsequent view is that infants orient more toward facelike patterns
learning/experience. Since visual experience begins at than inverted face patterns in the temporal, but not the
birth, these biases are innate in the sense that they rely nasal, visual field (Simion et al., 1998). Since the retino-
on little or no visual experience with faces to emerge. tectal (subcortical) pathway is thought to have greater
input from the temporal hemifield and less input from
Facedness Newborn babies move their eyes, and some- the nasal hemifield than the geniculostriate (cortical)
times their heads, longer to keep a moving facelike pat- pathway, this asymmetry in the preferential orienting to
tern in view than several other comparison patterns faces is consistent with subcortical, but not cortical, in-
(Johnson et al., 1991). It seems that all that is needed volvement. The lack of preferential orienting to faces
to elicit a response is a very schematic version of the in the nasal field is not simply due to a general lack of
face: Just a triangular arrangement with three blobs for sensitivity, since in this same visual field there is in-
eyes and a mouth is sufficient (Johnson and Morton, creased responding to an optimal spatial frequency
1991). Some authors have argued that these data can nonface stimulus over another nonface stimulus (Sim-
be explained by a sensory hypothesis. In this hypothesis, ion et al., 1998). This evidence is consistent with the
there is no aspect of the system that is responding spe- hypothesis that the preferential orienting to faces is me-
cifically to faces; instead, the preferential orienting of diated by a subcortical mechanism; however, as the vi-
the newborn to faces is just a consequence of more gen- sual cortex of the newborn is at least partially functional
eral mechanisms guiding visual attention. For example, (Maurer and Lewis, 1979), it might also contribute to
in the linear systems model of infant visual attention, the response.
visual preferences in the first months of life are based
on the amplitude spectra of the stimuli (the amount of Identity Not only do newborns preferentially orient to
energy in a pattern, defined by the amplitudes and ori- faces, but they also process information about facial
entations of the component spatial frequencies; Banks identity. Newborns look longer at the mother's face
and Ginsburg, 1985; Banks and Salapatek, 1981). Thus, than a stranger's face just hours to days after birth (even
it is argued that infants will prefer facelike patterns be- when cues from her smell and her voice are eliminated;
cause they tend to have more energy at the spatial fre- Bushnell et al., 1989; Field et al., 1984; Pascalis et al.,
quencies to which they are more sensitive. In contrast, 1995). This observation represents a difficulty both for
others explain newborns' preferential orienting to faces the sensory and for the social hypothesis, since it dem-
with a social hypothesis. In this view, infants have an onstrates that from the beginning of life there is a
innate preference for facelike stimuli that is based not mechanism involved in face processing that is sensitive
only on the visibility of the stimuli but on a more spe- to experience with individual faces. Both views must
cific knowledge of the configuration of the face. For postulate the existence of an additional mechanism ca-
example, in the Conspec/Conlern hypothesis, an in- pable of learning about individual faces based on ex-
nate, subcortical mechanism, Conspec, causes new- perience. This early learning may be mediated by the
borns to orient to patterns with elements in a facelike hippocampal complex (Johnson and de Haan, in press;
arrangement (Morton and Johnson, 1991). Recent Nelson, 1995), since this area is known be involved
studies demonstrating that babies will still prefer a face- in memory, to functionally mature early relative to
like pattern even to one that is more (Valenza et al., memory-related neocortical areas, and to mediate
384 COGNITION
erence emerges for fixating faces compared to other ine the neural correlates efface processing have shown
patterns presented in the central visual field (Maurer that face-responsive ERP components are more specific
and Barrera, 1981). This behavioral change is thought to human faces in adults than in infants (de Haan, John-
to be due to the functional development of visual cor- son, and Oliver, 1998; de Haan et al., 1999). In adults,
tical pathways that inhibit the preferential following re- the N170, a negative deflection over occipitotemporal
sponse and mediate the new preferential fixation electrodes that peaks approximately 120-200 ms after
response (Johnson and Morton, 1991). stimulus onset, is thought to reflect the initial stage of
Although 2-month-old infants do look longer at cen- the structural encoding of the face (Bentin et al., 1996).
trally presented faces than at other patterns, there are It is of larger amplitude and longer latency for inverted
factors other than facedness that can be equally or more than for upright faces (Bentin et al., 1996; de Haan,
salient in attracting their attention, such as stimulus Johnson, and Oliver, 1998; see figure 26.1), a pattern
complexity. For example, in one series of studies (Haaf, that parallels behavioral studies showing that adults are
1974, 1977) the looking times of infants 1-5-months slower at recognizing inverted than upright faces
old were measured to a set of four stimuli that varied (Carey and Diamond, 1994). In adults the effect of stim-
independently in complexity (number of elements) ulus inversion on the N170 is specific for human faces
and facedness (number of facial features in the proper and does not occur for monkey faces or sheep faces (de
place). The looking of infants 1-2 months old was de- Haan et al., 1999; see figure 26.1; see color plate 16).
termined solely by the complexity of the pattern (they In 6-month-old infants, the P400, a positive deflection
looked longer at more complex patterns), while older over occipitotemporal electrodes that peaks approxi-
infants' looking was affected both by the complexity mately 350-450 ms after stimulus onset, is, like the
and the facedness of the pattern. N170, affected by stimulus inversion. However, the in-
Moreover, at this young age infants' representation fant P400 differs from the adult N170 on a number of
of facedness appears quite different from that of adults parameters. First, the effect of stimulus inversion is evi-
in that it is based primarily on the eyes rather than on dent in the ERPs approximately 200 ms later for 6-
the entire constellation of facial features. For example, month-olds than for adults (see figure 26.2; see color
when 2-month-olds are shown schematic faces paired plate 17). This difference in timing is not necessarily
with faces that are scrambled or are missing various due to an overall slower rate of information processing
in infants since the PI component, reflecting earlier
parts, they will look equally long at the full face and a
stages of visual processing, occurs only slightly later for
face with only eyes present (nose and mouth missing),
infants than for adults. This suggests that the inversion
and will look at both of these longer than at a face with
effect may occur at a different and later stage in face
only nose and mouth present (eyes missing; Maurer,
processing for infants than adults (Johnson and
1985). Furthermore, the position of the eyes in the face
de Haan, in press). Second, infants, like adults, show an
does not make a difference to the infants' preference,
inversion effect bilaterally for human faces but, unlike
although they are able to discriminate among faces with adults, they also show an inversion effect over left hemi-
eyes in varying positions following habituation (Maurer, sphere electrodes for monkey faces. The fact that the
1985). Thus, at 2 months of age the eyes are a more inversion effect was specific to human faces in adults
salient feature of the face than the nose or mouth, but but was not in infants suggests that at 6 months of age
where the eyes are located is immaterial to the babies' the infant's face processing system is more broadly
preference for facelike drawings. The importance of responsive to the category of faces than is the adult's.
the eyes is also noticeable in infants' social interactions: The fact that the inversion effect for monkey faces oc-
Hains, Muir, and Franke (1994) found that 3-5-month- curred only over the left-hemisphere, and not the right-
olds' smiling decreased during a social interaction hemisphere electrodes, might be because the right
when their adult partner's eyes were averted and recov- hemisphere develops specificity to human face process-
ered when mutual gaze was reestablished. It may not be ing earlier than does the left.
until 4-6 months of age that infants begin to have a In adults, the N170 is thought to reflect activation of
more adultlike representation of facedness and re- the fusiform gyms and/or cortical regions lateral to it,
spond to faces independently of complexity and with a as these areas have been identified as face-responsive in
greater appreciation of facial features in addition to the studies using fMRI (Bentin et al., 1996; Kanwisher,
eyes (see Maurer, 1985). McDermott, and Chun, 1997) and intracranial record-
Even by 6 months of age, babies' mental representa- ings (Allison et al., 1999). It is not known whether the
tion of facedness may differ from that of adults in being infant P400 also reflects activation of these areas. In
broader and less tuned to the human face. For example, adults, there is evidence to show that fusiform involve-
studies using event-related potentials (ERPs)1 to exam- ment in visual processing is related to visual expertise.
For example, "face-selective" areas of the fusiform gyrus from activation only to faces to activation in addition to
are also activated when individuals are tested with cate- other expert categories). One factor that may contrib-
gories of visual stimuli in which they are either naturally ute to this difference is that, during development, the
experts or are trained to be experts (Gauthier et al., infant must handle both the task of discovering the cate-
2000). This suggests the possibility that the process gory itself ("facedness") and the task of discovering
of acquiring visual expertise in adulthood parallels mechanisms to encode individual examples within the
the development of face processing, and that both in- category; in contrast, when adults acquire expertise
volve recruitment of fusiform areas. However, there is with a visual category, either the structure of the exper-
some indication that the two processes may not be com- imental situation or their own previous knowledge
pletely similar. One indication is that the developmental means that they are faced with only the second of these
process appears to be characterized by a narrowing, or two tasks. Further studies using ERPs (which are easily
limiting, of the specialized processing (e.g., from all measurable in both infant and adults) to compare the
faces to just human faces), while the adult's learning development of face processing with the development
process is characterized by an "expanding" of the use of visual expertise in adults would be of interest in this
of the fusiform area to other object categories (e.g., context.
386 COGNITION
FIGURE 26.2 Amplitude distributions of activity at the peak peak of the positivity of the P400 is seen in red. The white/
of the P400 created using spherical spline interpolation. The light areas indicate positivities. The black/dark areas indicate
view is looking straight on from the back of the head. The negativities.
Identity At approximately 2 months of age several eyes (Maurer and Salapatek, 1976). Several prominent
changes in infants' visual processing occur that affect theories of adult face processing postulate that adults
their processing of facial identity. One such change is encode facial identity using information about the spa-
that infants become more sensitive to the internal facial tial relations among the features of the face (e.g., Dia-
features of static faces. For example, 2-3-month-old in- mond and Carey, 1986; Rhodes et al., 1998). This
fants are able to remember faces from only the internal enhanced sensitivity to internal facial features might al-
features both after a 2- and a 24-hour delay (Pascalis et low infants to encode faces in a manner more similar
al., 1998), they show preferential looking to the mother to how adults encode faces. However, this increased sen-
even if only her internal facial features are showing sitivity to internal features is not specific to faces and
(Morton, 1993), and they can recognize a familiar face also occurs for other nonface patterns.
from a novel viewpoint based on only internal features A second change in face processing that occurs at
(Pascalis et al., 1998). Studies of infants' scanning pat- approximately 2-3 months of age is that infants begin
terns show that at this age there is an increase in scan- to relate information between individual faces. For
ning of the internal features of faces, particularly the adults, one important aspect of encoding facial identity
388 COGNITION
category than pairs that are equally physically different Conclusions The evidence suggests that by at least 3-6
but are both within the same category. In the domain months of age cortical areas are involved in face pro-
of speech and color perception, the categorical percep- cessing. This is evident in some characteristics of corti-
tion paradigm has been extensively used to argue for cal processing of faces that are quite adultlike by 6
basic perception categories (Repp, 1984). To study cat- months: (1) the hemispheric difference in the nature
egorical perception of emotional expressions, realistic of information encoded from the face, (2) the right
drawings (Etcoff and Magee, 1992) and/or photo- hemisphere bias for recognizing facial identity, and
graphs were changed in equal steps so that a face show- (3) the effect of inversion on electrocortical measures
ing one expression gradually transformed into another of processing faces. However, for the infant these char-
(de Gelder, Teunisse, and Benson, 1997; Young et al., acteristics may not be unique to faces. For example,
1997). With these types of stimuli, adults show the clas- there is also a hemispheric difference in the nature of
sic pattern of categorical perception, with enhanced information encoded from geometric patterns; the
discrimination of faces that cross an emotion category right hemisphere is also activated during object pro-
compared to pairs that are equally physically different cessing; and the inversion effect is not specific to human
but are both within the same emotion category. A re- faces (and possibly not specific to faces at all, as this has
cent study using looking times measures showed that by yet to be tested). Moreover, important changes that oc-
7 months of age infants also show categorical percep- cur in facial identity and emotion processing, such as
tion of a happy-fear continuum (Kotsoni, de Haan, and increased processing of internal features and the rela-
Johnson, 1999). Whether this categorical perception tion of individual examples of faces and ability to
represents a broad positive-negative dimension or form categories, appear to be more general cognitive
more specific emotion categories remains to be deter- changes that affect not only processing of faces but also
mined. However, the results support the idea that in- processing of other visual stimuli. Thus, it may be that
fants perceive at least some emotions categorically from the cortical mechanisms involved in face processing are
very early in life, and this ability may, like other cases of not yet highly unique to faces themselves.
categorical perception such as color vision, have a hard- In spite of this, it should be noted that the abilities
wired basis. of the young infant are sufficient to allow them to re-
Recent studies with adults suggest that perception of spond to faces in social situations. By 3-6 months, in-
different types of emotional expressions may have dif- fants appear to have some expectations regarding the
ferent neural bases. For example, the amygdala appears nature of social interactions in the presence of a per-
to be involved during processing of fearful and sad faces son's face (Ellsworth, Muir, and Hains, 1993; Tronick et
but not of angry faces (Blair et al., 1999). Moreover, al., 1978). They appear to be particularly attentive to
even within a category of expression there may be dif- the emotional aspects of the face, and show this in their
ferent, parallel pathways of processing. For example, in own emotional reactions by becoming distressed when
one neuroimaging study it was found that the right their adult partner ceases interacting normally and
amygdala, pulvinar, and superior colliculus were acti- smiling less if their partner is a puppet or an upside-
vated only in response to masked ("unseen") fear- down human rather than a normal upright human
related stimuli but not those that were seen, while the (D'Entremont, 1994; Ellsworth, Muir, and Hains, 1993;
left amygdala was activated to both seen and unseen Gusella, Muir, and Tronick, 1988; Muir and Rach-
fearful stimuli (Morris, Ohman, and Dolan, 1999). Longman, 1989).
These results were interpreted to suggest that there is a
subcortical pathway involving the right amygdala via
midbrain and thalamus that processes unseen fearful CHILDHOOD
stimuli in parallel to the cortical pathway for conscious Facedness There is evidence to suggest that even beyond
identification. At present, very little is known about the infancy the child's mental representation of facedness
development of these neural pathways mediating face differs from that of the adult. One way this has been
recognition during infancy (reviewed in Nelson and de demonstrated is by studies of the development of the
Haan, 1997). There is some evidence that the amygdala "other race" and "other species" effects. These effects
may be involved in processing of facial expressions dur- refer to the fact that adults are worse at recognizing the
ing infancy (Balaban, 1995), but it is not clear whether faces of individuals of another race or species than
processing of different expressions is subserved by dif- those of their own race and species. This presumably
ferent pathways or whether there are parallel pathways occurs because the latter are seen most frequently and
in processing of expressions during infancy. can thus more easily and effectively be related to
390 COGNITION
All ages tested showed a composite effect: composites it may be that developmental changes in recognizing
were recognized less well than noncomposites, suggest- unfamiliar faces are due to their increased employment
ing that configural encoding was used at all ages. How- of this strategy.
ever, adults and 10-year-old children did show an Yet another account of the mechanism underlying
additional inversion effect in that upright faces were developmental changes in face recognition during
generally recognized more quickly than inverted ones; childhood is the right hemisphere development hypothesis.
in contrast, 6-year-olds responded equally quickly to the Carey and Diamond (1977) hypothesized that devel-
two (Carey and Diamond, 1994). opmental changes in face recognition are due to devel-
In order to account for this pattern of findings, the opment of the right hemisphere. In support of their
authors argued that there is more than one type of con- argument, they pointed out that young children's per-
figural encoding: (1) a type of configural encoding that formance on face processing tasks was similar to that
is measured by the composite/noncomposite effect, observed in adult patients with right hemisphere brain
that causes an inversion effect, and that is presented at damage. However, studies using the divided visual field
least from 6 years; (2) a type of configural encoding that technique in children have yielded mixed results. Some
causes an age-dependent increase in the inversion ef- studies report that a left visual field advantage in face
fect. They argue that this second type of encoding that processing emerges only after age 8 (Reynolds and
"switches" with development is norm-based encoding (i.e., Jeeves, 1978) while others report that even at age 7
encoding relative to a prototype or common face space) there is a left visual field advantage and the degree of
and causes age-related changes in face processing and advantage does not change with further development
inversion effects during childhood. This two-process ex- (Young and Bion, 1981).
planation also can account for the results of infancy
studies: The type of configural encoding assessed in the Conclusions During childhood a major change in face
composite effect might be the form that is present and processing may occur that relates to emergence of
causes inversion effects and right hemisphere bias in norm-based encoding effaces. As the child's prototype
infant face processing. of the face becomes more adultlike, it may become
A different account of developmental changes in face more tuned to the types of faces that the child most
processing during childhood is the depth of processing hy- often sees, and with this may emerge the distinctiveness
pothesis. Adults' memory for faces is better if during effects and other-race and other-species effects seen in
learning they make a "deep" judgment regarding a per- adult face processing. Since norm-based encoding is
sonality trait (e.g., honesty) of a face than if they make thought to reflect development of visual expertise with
a more "shallow" judgment based on a physical char- a category (Rhodes, 1993), it may be that it is this type
acteristic (e.g., hair color, gender; Coin and Tiberghien, of processing that is related to fusiform face-area acti-
1997; McKelvie, 1985; Mueller and Wherry, 1980; vation, as recent studies suggest that activation of this
Yoshikawa, 1995). This deep encoding does not neces- area is related to expertise with a particular category of
sarily lead to superior memory than that observed after visual stimulus (Gauthier et al., 1999). The fusiform gy-
normally employed strategies, suggesting that adults rus may mediate this expert processing itself or it may
normally spontaneously engage in deep rather than select the input to channel into a special type of pro-
shallow types efface processing (Sporer, 1991). The ex- cessing. Neurally, changes in children's face processing
act reason for the benefit from deep encoding is not might be due in part to functional development of the
known, but it has been proposed to be due to encoding fusiform areas and left frontal areas known to be in-
of a semantic representation of the face (McKelvie, volved in adults' memories for faces, and these changes
1985) and/or to an increased likelihood of encoding may be in part driven by experience due to increased
distinctive features (Winograd, 1981). It is possible that exposure to faces and the number and type of features
children, unlike adults, only gradually come to auto- children can or do attend to and encode. In addition,
matically use a deep or optimal strategy for encoding children may tend to automatically process faces deeply
new faces, and therefore their recognition of unfamiliar as they grow older, resulting in better face recognition.
faces gradually improves throughout childhood. These
changes might be linked to functional development of Developmental disorders of face processing
left frontal cortical areas as left prefrontal cortex is ac-
tivated during encoding effaces by adults (Haxby et al., Specific developmental disorders efface processing are
1994, 1995) and during deep encoding (Kapur et al., relatively rare. However, studying these cases can be of
1994). Since children do show superior memory for interest because it can allow one to determine whether
deeply versus shallowly encoded items (Schultz, 1983), there exists already early in life a domain-specific face
392 COGNITION
Gelder, Vroomen, and van der Heide, 1991). Together, pairments in face processing following unilateral focal
these results suggest that children with autism might cortical lesions are uncommon.
normally spontaneously process different aspects of fa- What factors underlie this slow developmental spe-
cial information than do children without autism and cialization of face processing? In adults this specializa-
that they have difficulty in perceiving cross-modal tion is thought to reflect in part extensive experience
information. in encoding faces and, more specifically, even in encod-
The fact that many characteristics of face processing ing particular types of faces. Thus, part of the devel-
that normally begin to emerge already in the first half- opmental process may be driven by increases in the
year of life (e.g., inversion effects, categorical percep- number of faces seen during development together
tion) are atypical in children with autism suggests that with general increases in speed and capacity of process-
these abnormalities begin at an early age. However ing that allow children to process and remember
there is also evidence that children can develop strate- greater numbers of faces. If such experience does play
gies that compensate for these abnormalities. For ex- a role, it might be particularly interesting to study
ample, the absence of categorical perception of facial changes in face processing at times when the exposure
expression in children with autism occurs at all IQ lev- to faces and demands on the system naturally increase
els, even though only those with a lower IQ show an (e.g., when children first enter school or large play-
impairment in expression recognition. The fact that in- groups) because these might be times at which transi-
dividuals with the same perceptual deficit could yet tions in the mechanisms of face processing can be
show different levels of ability in expression recognition observed. These changes in input may themselves play
suggests that some individuals are able to use compen- a role in reorganization of the cerebral mechanisms un-
satory strategies. Thus there appears to be a degree of derlying face processing at different ages.
plasticity in the developing system that allows for devel- With respect to brain mechanisms, one view is that
opment of alternative strategies/mechanisms in face there may be special "windows of opportunity" during
processing. which neural systems "expect" a particular input or ex-
perience to shape their normal development. The neu-
Conclusions: How does face processing develop? ral system remains open to this input only for a limited
amount of time, and if it is absent or atypical, devel-
Infants are born with many tendencies that work to- opment will follow an abnormal trajectory. In this
gether to make the human face a frequent focus of in- "experience-expectant" view (Greenough and Black,
fant visual attention. However impressive, these 1992), structure within the environment that is typically
tendencies are only the beginning step in a long devel- common to all members of the species (e.g., seeing pat-
opmental process. They may best be considered not as terned light, seeing faces, seeing emotional expres-
the immediate precursors to the adult face processing sions) is allowed to shape development so that not all
system, but as more general orienting tendencies that aspects have to be prespecified. It has been proposed
serve a number of developmental purposes. One might that, at a more microscopic level, these windows of op-
be that they cause faces to be a frequent input to the portunity are reflected by the period of synaptic over-
developing face processing system; another, that they production and subsequent "pruning" of synapses to
assist the relatively helpless infant to elicit care from adult levels in the cerebral cortex (Greenough and
their caretakers, and so on. Black, 1992). In this view, development of specialization
The evidence suggests that by at least 3-6 months, for face processing may not be the same as development
when electrocortical responses to familiar faces and of expertise in adulthood because the "neural environ-
hemisphere differences in recognition of faces are seen, ment" may differ at the two ages. The visual recognition
cortical areas are involved in face processing. However, system may have a window of opportunity when envi-
in spite of the fact that several adultlike characteristics ronmental inputs can "tune" the system to be maximally
of face processing are present by this time, such as in- efficient at processing the types of stimuli typically seen
version effects and the right hemisphere bias, these in the visual environment. For the developing infant,
characteristics are not unique to face processing. In this faces will provide a frequent input so some aspects of
way, the infant's face processing system appears to be the system may become specially tuned to the types of
much broader in nature than the adult's. To a certain processing most useful for encoding faces; however,
extent, this may be true even into childhood. The idea faces are not the only input, so other aspects of the
that during development the face processing system is system may be driven by other, nonface inputs. Later in
less specialized is consistent with results of studies of life these systems in their "steady state" may still be best
early cortical lesions, which suggest that specific im- tuned to the input received during development, but
394 COGNITION
BRAVERMAN, M., D. FEIN, D. Lucci, and L. WATERHOUSE, 1989. European Developmental Psychology Conference, Spetses,
Affect comprehension in children with pervasive develop- Greece.
mental disorders.J. AutismDev. Disorders 19:301-315. DE HAAN, M., M. H.JOHNSON, D. MAURER, and D. PERRETT,
BUSHNELL, I. W., 1979. Modification of the externality effect 1999. Recognition of individual faces and average face pro-
in young infants.J. Exp. Child Psychol. 28:211-229. totypes by 1- and 3-month-old infants. Submitted
BUSHNELL, I. W. R., F. SAI, and J. T. MULLIN, 1989. Neonatal DE HAAN, M., and C. A. NELSON, 1997. Recognition of the
recognition of the mother's face. Br.J. Dev. Psychol. 7:3-15. mother's face by six-month-old infants: A neurobehavioral
CAMPBELL, R., B. BROOKS, E. DE HAAN, and T. ROBERTS, 1996. study. Child Dev. 68:187-210.
Dissociating face processing skills: Decision about lip-read DE HAAN, M., and C. A. NELSON, 1998. Discrimination and
speech, emotion and identity. Quart. J. Exp. Psychol. categorisation of facial expressions of emotion during in-
A49:295-314. fancy. In Perceptual Development, A. Slater, ed. Hove, UK: Psy-
CAMPBELL, R., and M. TUCK, 1995. Recognition of parts of chology Press, pp. 287-309.
famous-face photographs by children: An experimental DE HAAN, M., and C. A. NELSON, 1999. Brain activity differ-
note. Perception 24:451 -456. entiates face and object processing in 6-month-old infants.
CAPPS, L., N. YIRMIYA, and M. D. SIGMAN, 1992. Understand- Dev. Psychol. 35:1113-1121.
ing of simple and complex emotions in non-retarded chil- D'ENTREMONT, B., 1994. Young infants' responding to static
dren with autism.J. Child Psychol. Psychiatry 33:1169-1182. and dynamic happy and sad expressions during a social in-
CAREY, S., and R. DIAMOND, 1977. From piecemeal to config- teraction. Infant Behav. Dev. 17:600.
urational representation effaces. Science 195:312-314. DE RENZI, E., D. PERANI, G. A. CARLESIMO, M. C. SILVERI, and
CAREY, S., and R. DIAMOND, 1994. Are faces perceived as con- F. FAZIO, 1994. Prosopagnosia can be associated with dam-
figurations more by adults than by children? Vis. Cogn. age confined to the right hemisphere—An MRI and PET
1:253-274. study and a review of the literature. Neuropsychologia 32:893-
CAREY, S., R. DIAMOND, and B. WOODS, 1980. Development of 902.
face recognition—A maturational component? Dev. Psychol. DERUELLE, C., and S. DE SCHONEN, 1991. Hemispheric asym-
16:257-269. metries in visual pattern processing in infancy. Brain Cog-
nition 16:151 -179.
CARON, A., R. CARON, C. MUSTELIN, and J. ROBERTS, 1992.
DERUELLE, C., and S. DE SCHONEN, 1995. Pattern processing
Infant responding to aberrant social stimuli. Infant Behav.
in infancy: Hemispheric differences in the processing of
Dev. 15:335.
shape and location of visual components. Infant Behav. Dev.
CELANI, G., M. W. BATTACCHI, and L. ARCIDIACONO, 1999.
18:123-132.
The understanding of emotional meaning of facial expres-
DE SCHONEN, S., M. GIL DE DIAZ, and E. MATHIVET, 1986.
sions in people with autism. J Autism Dev. Disorders 29:
Hemispheric asymmetry in face processing in infancy. In
57-66.
Aspects of Face Processing, H. D. Ellis, M. A.Jeeves, F. New-
CHANCE, J. E., A. L. TURNER, and A. G. GOLDSTEIN, 1982. De- combe, and A. Young, eds. Dordrecht: Martinus Nijhoff
velopment of differential recognition for own- and other- Publishers, pp. 199-209.
race faces.J. Psychol. 112:29-37. DE SCHONEN, S., and E. MATHIVET, 1989. First come, first
CICONE, M., W. WAPNER, and H. GARDNER, 1980. Sensitivity to served: A scenario about the development of hemispheric
emotional expressions and situations in organic patients. specialization in face recognition during infancy. Eur. Bull.
Cortex 16:145-158. Cogn. Psychol. 9:3-44.
COIN, C., and G. TIBERGHIEN, 1997. Encoding activity and face DIAMOND, R., and S. CAREY, 1977. Developmental changes in
recognition. Memory 5:545-568. the representation effaces.J. Exp. Child Psychol. 23:1-22.
DAMASIO, A. R., D. TRANEL, and H. DAMASIO, 1990. Face ag- DIAMOND, R., and S. CAREY, 1986. Why faces are and are not
nosia and the neural substrates of memory. Annu. Rev. Neu- special: An effect of expertise.J. Exp. Psychol. Gen. 115:107—
rosci. 13:89-109. 117.
DAVIES, S., D. BISHOP, A. MANSTEAD, and D. TANATAM, 1994. ELLIS, H. D., 1992. The development of face processing skills.
Face perception in autistic children. J Child Psychol. Psychi- Philos. Trans. R. Soc. London B. Biol. Sci. 335:105-111.
atry 35:1033-1058. ELLIS, H. D., and J. W. SHEPHERD, 1975. Recognition of up-
DE GELDER, B., J.-P. TEUNISSE, and P. J. BENSON, 1997. Percep- right and inverted faces presented in the left and right visual
tion of facial expressions: Categories and their internal fields. Cortex 11:3-7.
structure. Cogn. Emotion. 11:1-23. ELLIS, H. D., J. W. SHEPHERD, and G. M. DAVIES, 1979. Iden-
DE GELDER, B., J. VROOMEN, and L. VAN DER HEIDE, 1991. Face tification of familiar and unfamiliar faces from internal and
recognition and lipreading in autism. Eur. J. Cogn. Psychol. external features: Some implications for theories of face rec-
3:69-86. ognition. Perception 8:431-439.
DE HAAN, E., and R. CAMPBELL, 1991. A fifteen-year follow-up ELLSWORTH, C., D. W. MUIR, and S. HAINS, 1993. Social com-
of a case of developmental prosopagnoia. Cortex 27:489- petence and person-object differentiation: An analysis of
509. the still-face effect. Dev. Psychol. 29:63-73.
DE HAAN, M., M. H. JOHNSON, and A. OLIVER, 1998. A high- ETCOFF, N., andj.J. MAGEE, 1992. Categorical perception of
density ERP study of infant and adult face processing. Tech- facial expressions. Cognition 44:227-240.
nical Report of the Human Frontiers Scientific Foundation FARAH, M. J., 1990. Visual Agnosia. Cambridge, Mass.: MIT
Developmental Cognitive Neuroscience Team. Press.
DE HAAN, M., M. H. JOHNSON, H. A. HATZAKIS, and O. PAS- FARAH, M. J., C. RABINOWITZ, G. E. QUINN, and G.T. Liu, 2000.
CALIS, 1999. Does cerebral specialization for face processing Early commitment of neural substrates for face recognition.
develop? A high-density ERP study. Poster presented at the Cogn. Neuropsychol. 17:117-124.
396 COGNITION
MILEWSKI, A. E., 1976. Infants' discrimination of internal and RHODES, G., S. BRAKE, and A. P. ATKINSON, 1993. What's lost
external pattern elements.J. Exp. ChildPsychol. 22:229-246. in inverted faces? Cognition 47:25-57.
MONDLOCH, C. J., T. L. LEWIS, D. R. BUDREAU, D. MAURER, RHODES, G., S. CAREY, G. BYATT, and F. PROFFITT, 1998. Cod-
J. L. DANNEMILLER, B. R. STEPHENS, and K. A. KLEINER- ing spatial variations in faces and simple shapes: A test of
GATHERCOAL, 1999. Face perception during early infancy. two models. Vis. Res. 38:2307-2321.
Psychol Sci. 10:419-422. RHODES, G., S. TAN, S. BRAKE, and K. TAYLOR, 1989. Race sen-
MORRIS, J. S., C. D. FRITH, D. I. PERRETT, D. ROWLAND, A. W. sitivity in face recognition: An effect of different encoding
YOUNG, A. J. CALDER, and R. J. DOLAN, 1996. A differential processes. In Cognition in Individual and Social Contexts, A. F.
neural response in the human amygdala to fearful and Bennett and K. M. McConkey, eds. Amsterdam: Elsevier,
happy facial expressions. Nature 383:812-815. pp. 83-90.
MORRIS, J. S., A. OHMAN, and R. J. DOLAN, 1999. A subcortical RUBENSTEIN, A. J., L. KALAKANis, and J. H. LANGLOIS, 1999.
pathway to the right amygdala mediating 'unseen' fear. Proc. Infant preferences for attractive faces: A cognitive expla-
Natl. Acad. Sci. USA 96:1680-1685. nation. Dev. Psych. 35:848-855.
MORTON, J., 1993. Mechanisms of infant face processing. In SAMS, M., J. K. HIETANEN, R. HARI, R. J. ILMONIEMI, and O. V.
Developmental Neurocognition: Speech and Face Processing in the LOUNASMAA, 1997. Face-specific response from the human
First Year of Life, B. S. de Boysson-Bardies, S. de Schonen, inferior occipito-temporal cortex. Neuroscience 77:49-55.
P. Jusczyk, P. McNeilage, andj. Morton, eds. London: Klu- SCHULTZ, E. E., JR., 1983. Depth of processing by mentally
wer Academic Publishers, pp. 93-102. retarded and MA-matched nonretarded individuals. Am. J.
MORTON, J., and M. JOHNSON, 1991. CONSPEC and CON- Mental Deficiency 88:307-313.
LERN: A two-process theory of infant face recognition. Psy- SERGENT, J., S. OHTA, and B. MACDONALD, 1992. Functional
chol. Rev. 2:164-181. neuroanatomy of face and object processing: A positron
MUELLER, J. H., and K. L. WHERRY, 1980. Orienting strategies emission tomography study. Brain 115:15-36.
at study and test in facial recognition. Am.J. Psychol. 93:107- SIMION, F., E. VALENZA, C. UMILTA, and B. DALLA BARBA, 1998.
117. Preferential orienting to faces in newborns: A temporal-
MUIR, D. W., S. M. J. HAINS, and L. A. SYMONS, 1994. Baby nasal asymmetry.J Exp. Psychol: Hum. Percept. Perform. 24:
and me: Infants need minds to read! Eur. Bull. Cogn. Psychol. 1399-1405.
13:669-682. SLATER, A., V. MORISON, and D. ROSE, 1983. The perception
MUIR, D. W., and K. RACH-LONGMAN, 1989. Once more with of shape by the new-born baby. Br.J. Psychol. 1:135-142.
expression: On de Schonen and Mathivet's (1989) model SPORER, S. L., 1991. Deep—deeper—deepest? Encoding strat-
for the development of face perception in human infants. egies and the recognition of human faces. J Exp. Child Psy-
Eur. Bull. Cogn. Psychol 9:103-109. chol.: Learn. Mem. Cogn. 17:323-333.
NACHSON, I., 1995. On the modularity of face recognition: TANTAM, D., L. MONAGHAN, H. NICHOLSON, andj. STIRLING,
The riddle of domain specificity. J. Clin. Exp. Neuropsychol. 1989. Autistic children's ability to interpret faces: A research
17:256-275. note.J Child Psychol. Psychiatry 30:623-630.
NELSON, C. A., 1995. The ontogeny of human memory: A cog- TAYLOR, M. J., G. MCCARTHY, E. SALIBA, and E. DEGIOVANNI,
nitive neuroscience perspective. Dev. Psychol. 31:723-738. 1999. ERP evidence of developmental changes in process-
NELSON, C. A., and M. DE HAAN, 1997. A neurobehavioral ap- ing effaces. Clin. Neurophysiol 110:910-915.
proach to the recognition of facial expressions in infancy. TEMPLE, C. M., 1992. Developmental memory impairment:
In The Psychology of Facial Expression, J. A. Russell and J. M. Faces and patterns. In Mental Lives, R. Campbell, ed. Ox-
Fernandez-Dols, eds. Cambridge, Mass.: Cambridge Univer- ford: Blackwell.
sity Press, pp. 176-204. TEUNISSE, J.-P., and B. DE GELDER, in press. Impaired categor-
OZONOFF, S., B. F. PENNINGTON, and S. J. ROGERS, 1990. Are ical perception of facial expressions in high-functioning ad-
there emotional perception deficits in young autistic chil- olescents with autism. Child Neuropsychol
dren? J. Child Psychol. Psychiatry 31:343-361. TRONICK, E., H. ALS, L. ADAMSON, S. WISE, and T. B. BRAZEL-
PASCALIS, O., M. DE HAAN, C. A. NELSON, and S. DE SCHONEN, TON, 1978. The infant's response to entrapment between
1998. Long-term recognition memory for faces assessed by contradictory messages in face-to-face interaction. J Am.
visual paired comparison in 3- and 6-month-old infants. J. Acad. Child Psychiatry 17:1-13.
Exp. Psychol.: Learn. Mem. Cogn. 24:249-260. VALENTINE, T, 1991. A unified account of the effects of dis-
PASCALIS, O., S. DE SCHONEN, J. MORTON, C. DERUELLE, and tinctiveness, inversion and race in face recognition. Quart.
M. FABRE-GRENET, 1995. Mother's face recognition in ne- / Exp. Psych. 43(A):161-204.
onates: A replication and an extension. Infant Behav. Dev. VALENTINE, T, and V. BRUCE, 1986. The effects of distinctive-
18:79-86. ness in recognising and classifying faces. Perception 15:525-
QUINN, P. C., 1998. Object and spatial categorisation in young 535.
infants. In Perceptual Development, A. Slater, ed. Hove, UK: VALENZA, E., F. SIMION, V. M. CASSIA, and C. UMILTA, 1996.
Psychology Press, pp. 131-165. Face preference at birth.J Exp. Psychol.: Hum. Percept. Per-
REPP, B. H., 1984. Categorical perception: Issues, methods, form. 22:892-903.
findings. Speech Lang. 10:243-335. WEBSTER, M. J., L. G. UNGERLEIDER, and J. BACHEVALIER,
REYNOLDS, D. M., and M. A.JEEVES, 1978. A developmental 1995. Development and plasticity of the neural circuitry un-
study of hemisphere specialization for recognition of faces derlying visual recognition memory. Can. J. Physiol Phar-
in normal subjects. Cortex 14:511-520. macol 73:1364-1371.
RHODES, G., 1993. Configural coding, expertise, and the right WEEKS, S. H., and R. P. HOBSON, 1987. The salience of facial
hemisphere advantage for face recognition. Brain Cogn. expression for autistic children. J Child Psychol Psychiatry
22:19-41. 28:137-152.
398 COGNITION
27 Spatial Cognitive Development
JOAN STILES
ABSTRACT This chapter focuses on the development of the proposal introduced an alternative to earlier accounts
behavioral and cortical systems that mediate basic spatial cog- in which attention to location and movement was as-
nitive functions. The review of spatial development is orga- sociated principally with the tectal visual system while
nized around the convenient neuroanatomical dissociation
between functions associated with the dorsal and ventral visuo-
form processing was linked to the genicostriate system
spatial streams. The first part of the chapter considers the de- (Schneider, 1967; Trevarthen, 1968). Ungerleider and
velopment of spatial processes associated with the dorsal, or Mishkin's account did not deny the role of the tectal
"where," system, focusing principally on spatial localization system in, particularly, the oculomotor aspects of spatial
and spatial attention. The second part considers the ventral, processing. Rather, it provided a more complete ac-
or "what," system and focuses on the analysis of spatial pat- count of the organization and functioning of higher
terns. The discussion of each spatial cognitive function begins
with an overview of the adult profile of cognitive-neural me- cortical systems, and of spatial cognitive processing
diation. Patterns of development and change in both the be- more generally.
havioral profiles and in the neural substrate are then The dorsal visual pathway begins at the retina and
considered in light of the adult model. projects via the lateral geniculate nucleus (LGN) of the
thalamus to primary visual cortex, area VI. From there
The term spatial cognition refers to a wide variety of the pathway proceeds to areas V2 and V3, then projects
abilities and skills, which include tracking a moving ob- dorsally to the medial (MT/V5) and medial superior
ject, localizing and attending to an object or event in (MST) regions of the temporal lobe, and then to the
the spatial array, and understanding how the parts or ventral inferior-parietal lobe (IP). Input to the dorsal
features of an object combine to form an organized pathway is derived principally, though not exclusively,
whole. Although spatial cognitive functions are not lim- from the large M-type retinal ganglion cells that project
ited to the visual domain, much of the work on spatial to the magnocellular layers of LGN and then to layer
processing has focused on the visuospatial processing 4C(a) of V1. Cells in this pathway are maximally sensi-
system. Neuropsychological studies of visual system ar- tive to movement and direction, and less responsive to
chitecture have identified dozens of interrelated visual color or form. The dorsal stream processes information
areas in the posterior cortices, each of which contrib- about optic flow and other aspects of motion, processes
utes to some aspect of visuospatial processing (e.g., Van information about spatial location, and plays an impor-
Essen, Anderson, and Felleman, 1992). In the early tant role in allocation of spatial attention. It has thus
1980s, Ungerleider and Mishkin (1982) outlined a use- been described as the "where" pathway (for discussions
ful scheme for understanding the organization of this of the role of the inferior parietal lobe in visual control
complex set of cortical areas and functions. Their de- of motor output, see Anderson et al., 1997; Goodale and
scription of a dual visuospatial processing system de- Milner, 1992).
rived from the growing body of evidence from both The ventral visual pathway also begins at the retina
human and animal studies suggesting distinct func- and projects via the LGN of the thalamus to primary
tional dissociations within striate and extrastriate visual visual cortex, area V1. From there, the pathway pro-
areas. According to their proposal, the cortical visual ceeds to areas V2 and V4, then projects ventrally to the
system can be functionally and anatomically subdivided posterior (PIT) and anterior (AIT) regions of the in-
into two principal processing streams—a dorsal path- ferior temporal lobe. Input to the ventral pathway is
way and a ventral pathway (figure 27.1; see color plate derived principally, though not exclusively, from P-type
18). The dorsal stream mediates spatial processing as- retinal ganglion cells that project to the parvocellular
sociated with attention to movement and location, layers of the LGN and then to layer 4C(b) of V1. Par-
while the ventral stream is primarily involved in pro- vocellular input to V1 organizes into distinct areas
cessing information about patterns and objects. Their called the blob and interblob regions (Livingstone and
Hubel, 1984; Wong-Riley, 1979). Cells in the blob
regions are maximally sensitive to form, while cells in
JOAN STILES Department of Cognitive Science, University of the interblob regions respond principally to color. The
California at San Diego, La Jolla, California. ventral stream, which has been described as the "what"
400 COGNITION
pathway, processes information about visual properties SPATIAL LOCATION Neuroimaging studies of adults sug-
of objects and patterns. gest that spatial location tasks activate widely distributed
The dorsal and ventral pathways project rostrally, brain areas. Haxby has used functional MRI to examine
each going to both common and adjacent areas of the brain activation profiles in normal adult subjects on
prefrontal cortex. Finally, there is substantial evidence tasks requiring them to compare the location of objects
that the two pathways are richly interconnected and at in two visually presented arrays (Haxby et al., 1991,
least partially overlapping both in the mature (e.g., 1994). Activation was observed in several areas along
Dobkins and Albright, 1994, 1995, 1998; Marangolo et the dorsal pathway. In the most posterior brain regions,
al., 1998; Merigan and Maunsell, 1993) and the devel- areas of extrastriate cortex were activated bilaterally;
oping visual system (Dobkins and Teller, 1996a,b). these included the calcarine, medial, and lateral areas
The anatomical division of function into the dorsal of the occipital lobe. These areas are presumed to be
and ventral streams provides a convenient organiza- involved in early visual processing and are typically ac-
tional structure for discussing the functional organiza- tivated on a wide array of visual processing tasks. In
tion and development of spatial cognitive processes. addition, the dorsolateral occipital (area 19) and the
Accordingly, the next two sections of this chapter reflect posterior superior parietal areas, extending rostrally to
this anatomical division. The first section focuses on the intraparietal sulcus (area 7), were also activated bi-
processes associated with the dorsal stream and concen- laterally. These areas are considered to be critical to
trates on the complementary issues of the development processing of location information. Finally, the dorso-
of spatial localization and of spatial attention. The sec- lateral prefrontal cortex (area 6) was activated unilat-
ond section takes up issues related to processes associ- erally on the right. This activation may be related to
ated with the ventral stream and is centered on the spatial memory demands or to eye movement. Very
development of pattern processing. similar patterns of activation were observed in studies
It should be noted that although a great deal is of spatial working memory (Jonides et al., 1993; Smith,
known about the neuropsychology of each of these pro- Jonides, and Koeppe, 1996; Smith et al., 1995; for dis-
cesses in adult human and animal populations, the cussion of imaging studies of spatial memory with chil-
study of the development of the neuropsychological un- dren, see also Casey et al., 1998; Nelson et al., 2000).
derpinnings of spatial cognitive processing is, itself, in Overall, subjects showed greater right lateralization
its infancy ... or at least its childhood. To date, there and, importantly, much more extensive frontal activa-
have been very few developmental studies that provide tion, including right dorsolateral prefrontal cortex (ar-
direct evidence mapping specific behavioral changes in eas 46 and 47), and the anterior cingulate.
spatial processing to specific change in the neural sub- There is also strong evidence that the dorsal fronto-
strate. Nonetheless, important basic information about parietal system plays a major role in control of visually
the postnatal development of the visuospatial process- guided motor behavior, including both eye movement
ing system is available. This kind of information, cou- and reach (Johnson et al., 1996; Wise et al., 1997). The
pled with very specific measures of behavioral change, task of looking or reaching to a location in space in-
makes it possible to begin to make inferences about as- volves a complex network of neural areas within the dor-
sociations between change in the neural substrate and sal system pathway. Prefrontal motor areas mediate
change in behavior. planning and preparation for motor action; activation
of these areas typically precedes the actual motor event.
The development of spatial processes associated There is considerable evidence for superior parietal in-
with the dorsal stream: Location and attention put to dorsal premotor and motor cortices. Patterns of
to space activity in linked frontal and superior parietal areas are
concordant, suggesting a network of spatial-motor con-
A variety of spatial processes have been associated with trol. In addition, recent studies have shown that inferior
activation of the dorsal visual pathway. Two of the most parietal areas connect to frontal premotor areas and
basic are considered in this section: (1) processing of play an important modulatory role in spatial-motor ac-
information about spatial location and (2) spatial atten- tivity (Anderson et al., 1997). Cells in this region re-
tion. The division into two distinct processes is some- spond to modification or correction of motor actions.
what artificial in that, for example, localization of an Further, different cells are selective for movement type
object may also require a shift in spatial attention. such that different groups of cells respond to eye move-
Nonetheless, there is substantial evidence for functional ment and reach. Consistent with these findings, Good-
and anatomical independence of key features of each ale and Milner (1992) have reported the case of a
process. woman with bilateral parietal lesions who showed
402 COGNITION
Squire, 1989) have little effect on AB search tal metabolic activity begin at about 8 months of age, a
performance. time that coincides with improved performance on the
Infant rhesus monkeys show a developmental pattern AB search task (Chugani, Phelps, and Mazziotta, 1987;
on the AB search task that is very similar to that of hu- but see also Jacobs et al., 1995, for data from infant
man children. At 1.5-2.5 months, the frequency of AB rhesus monkeys). Second, Fox and Bell (1990) reported
error is high, mirroring the performance of 7.5-8.5- systematic increases in EEG spectral power over frontal
month-old human infants. Between 2.5 and 4 months, brain regions over a time period that corresponded to
performance improves and is affected by increasingly developmental improvement in the AB search task;
longer delays, such that by 4 months, 12-second delays such increases were not observed in parietal or occipital
do not significantly increase the frequency of the AB regions. It should be noted, however, that Fox and Bell
error. At 4.5 months, bilateral lesions of dorsolateral also reported that no such association was observed for
prefrontal cortex produce effects very similar to com- the object retrieval task. Thus, the two marker tasks in-
parable lesions in adult monkeys. dexing frontal lobe maturation yielded different pro-
Diamond suggests that the dorsolateral prefrontal le- files of association on this important brain index.
sion data implicate two important functions necessary Diamond's studies point to the importance of devel-
for the successful performance of the AB search task: opmental change in prefrontal brain systems for suc-
explicit spatial memory and inhibitory control. It has cessful performance on the AB error task, and it is clear
been well documented that prefrontal lesions impair that important changes do take place in frontal systems
spatial memory (Goldman-Rakic, 1987). Studies of rats during this period. However, other data suggest that the
have shown that lesions to dorsomedial prefrontal cor- developmental story may be somewhat more complex.
tex specifically impair memory for temporal order of For example, in addition to their work on frontal EEG
spatial location (Chiba, Kesner, and Gibson, 1997). The spectral power, Bell and Fox (1992) have also examined
AB search task explicitly requires memory for the most patterns of long-range EEG coherence. They reported
recent hiding location in a series of trials. Furthermore, that a systematic increase in anterior-posterior coherence
the effects of lesions to dorsolateral prefrontal cortex was associated with improved performance on the AB
are consistent with the human adult activation studies error task. These data suggest that another important
reported earlier. Although dorsolateral prefrontal acti- factor involved in neural mediation of AB task perfor-
vation was observed in both the perception and mem-
mance is the stabilization of long-range axonal connec-
ory for location tasks, it was enhanced, particularly in
tions within the dorsal stream. As discussed earlier,
the right hemisphere, under conditions of memory
these pathways may be critical for control of spatial
load. Frontal lesions have also been associated with fail-
reaching. Baillargeon and DeVos (1991) have reported
ure of inhibitory control (Welsh and Pennington,
that the frequency of AB error among 4-month-old in-
1988). The AB search task requires children to inhibit
a prepotent response (reach to A) in order to make the fants is greatly reduced when the requirement to reach
new, correct response (reach to B). Diamond (1991; for the hidden object is eliminated. Thus, while related
Diamond, Werker, and Lalonde, 1994) compared per- to frontal lobe development, the coherence data sug-
formance on the AB search task and an object retrieval gest more widespread change in the dorsal system that
task that required inhibition of a prepotent response may well affect different aspects of performance on this
but had no memory component, and found that chil- task. The complexity of the developmental story is also
dren showed similar patterns of developmental change evident in the behavioral studies reported earlier. Data
on both. from a wide range of studies suggest that performance
The results of the lesion studies led Diamond to con- on the AB error task can be affected by a wide variety
clude that "maturational change in the frontal cortices of factors. Under some conditions, children much
underlies improved performance on AB [search] . . . younger than 8 months can solve the task easily, while
and object retrieval between 7.5 and 12 months in hu- under other conditions even toddlers fail the task. Some
man infants and 1.50-2.5 months in infant monkeys" of the factors that affect performance are task-related
(Diamond, 1991, p. 167). She cites two sources of evi- and include such things as landmark information or
dence specifically linking the timing of markers of mat- increasing the spacing between hiding locations. Oth-
urational change in frontal cortex to developmental ers, such as reports of enhanced performance among
change on the behavioral tasks (Diamond, Werker, and healthy preterms or among infants who begin to self-
Lalonde, 1994). First, Chugani's studies of regional locomote early, are related specifically to the child's ex-
change in brain metabolic activation over the first year perience. The simple case of frontal lobe maturation
of life in human children indicate that increases in fron- alone cannot account for all of these data.2
404 COGNITION
slowing of RT following presentation of invalid cues ing for infants with unilateral left anterior perinatal fo-
(Posner et al., 1984). By contrast, patients with supra- cal brain injury (Johnson et al., 1998). These data
nuclear palsy, a disorder associated with damage to the suggest that the systems crucial to the control of atten-
superior colliculus, fail to show IOR responses. Finally, tional processes may change with development, such
patients with thalamic damage are slow to detect targets that early on, frontal regions, and in particular left fron-
presented anywhere in the visual field. tal regions, may be crucial to efficient allocation of at-
There is a small, but growing, literature on infant tention across the spatial field.
ability to shift attention in the visual field. A number of
infant studies have used procedures that allow for ex- NEGLECT Hemispatial neglect is an attentional disor-
amination of both cue facilitation and IOR. These stud- der frequently observed in adults following unilateral
ies show that by at least 6 months, infants show both parietal injury. In severe cases, hemispatial neglect pa-
facilitation and IOR (Clohessy et al., 1991; Harman et tients appear to be unaware of objects and events in the
al., 1994; Hood, 1993; Johnson, Posner, and Rothbart, visual hemispace contralateral to their lesion (Heilman
1994; Johnson and Tucker, 1996). Attempts to evoke and Valenstein, 1993). They may eat only half the food
these responses from younger children have been on their plates, fail to comb half their hair, and on sim-
mixed. However, control of factors, such as SOA dura- ple paper-and-pencil copying tasks, fail to reproduce
tion and cue/target eccentricity, appear to be critical to half of the model form. Impairments associated with
elicitation of the responses. Using 200 and 700 SOAs, neglect cannot be accounted for by sensory deficits. In-
Johnson and Tucker (1996) demonstrated reliable fa- deed, the disruption experienced by neglect patients
cilitation and IOR among 4-month-old children but did may be more devastating than those experienced by pa-
not show facilitation among 6-month-olds. However, tients with visual sensory deficits (McFie, Piercy, and
when a 133-ms SOA was introduced, 6-months-olds Zangwill, 1950). There is evidence that the attentional
showed strong facilitation. This finding suggests that disorder may extend beyond the immediate, percep-
while the basic attentional responses may be robust as tually available environment and may include the
early as 4 months, the timing parameters that elicit the patient's representation of space (e.g., Bisiach and
response may change with development. Similarly, Har- Luzzatti, 1978). In many patients, a milder form of
man and colleagues found no IOR response among 3- hemispatial inattention known as extinction is evident.
month-old children when stimuli were presented at 30° These patients appear to be able to attend to the con-
eccentricity but found a strong response when pre- tralesional field if there are no distractors in the ipsi-
sented at 10°. Thus, distribution of attention across the lesional field. For example, a patient may be able
visual field may also change with development. Few identify an object presented singly to either the right or
studies have examined facilitation and IOR in children left visual field; but when two objects are presented si-
2 months and younger. Johnson reported only weak fa- multaneously, one to each field, the patient fails to iden-
cilitation effects and no IOR effects among 2-month-old tify the object in the contralesional field.
infants. However, Valenza's study of newborns suggests The literature on neglect in child populations is very
that IOR may be present in the first days of life (Valenza, limited (see Temple, 1997, for a summary). There are
Simion, and Umilta, 1994). several reports of documented neglect in cases of pa-
Studies of children with early brain injury provide thology acquired after the age of 5 (Fero, Martins, and
some insight into the neural mechanisms that mediate Tavora, 1984; Heller and Levine, 1989). Fero and col-
these basic attentional processes in development. Craft leagues (1984) provided three case studies of docu-
and White (1994) examined attention shift in a group mented neglect in children with acute RH pathology;
of 33 children with perinatal brain injury, using the shift in all three cases, the neglect resolved after a period of
attention standard task with 100- and 800-ms SOAs. treatment and recovery. Heller and Levine (1989) ex-
Children were grouped by site of lesion: anterior (n = amined children with brain injury occurring either be-
6), posterior (n = 10), diffuse (n = 8), and no appar- fore or after age 6. Although children with LH or RH
ent lesion (n = 9). In contrast to adult patients, chil- injury acquired after age 6 showed evidence of contra-
dren with posterior lesions did not show any of the lesional neglect, no evidence of neglect was found
specific deficits noted for patients with injury to the pa- among children with earlier acquired RH lesions, nor
rietal complex. However, children with anterior lesions among children with LH injury.
failed to show the facilitation effect to valid cues pre- More recently, Thompson reported a case of a child
sented in the right, but not the left, visual field, sug- who suffered right posterior traumatic brain injury at 3
gesting an effect of left anterior injury not observed years, 8 months (Thompson et al., 1991). The child
among adult patients. Johnson reported a similar find- showed a transient period of contralesional neglect and
406 COGNITION
ferior temporal lobes (including inferior-temporal and children use to organize the parts. Furthermore, pat-
fusiform gyri). These areas have been identified in ani- tern complexity affects how children approach the
mal (Desimone, Albright, and Gross, 1984; Tanaka et problem of analysis. In studies using both the Rey-
al., 1991) and in human neuroimaging studies (Haxby Osterrieth complex form and simplified variants, it has
et al., 1991; Malach et al., 1995) as crucial for identify- been shown that simplification of the pattern induced
ing object shape and features. Martinez used both RT more advanced reproduction strategies (Akshoomoff
and measures of functional brain activation to investi- and Stiles 1995a,b).
gate profiles of neural mediation when subjects were Imaging studies of normal children provide evidence
asked to attend to either the global or the local level of for protracted development of the neural systems as-
a pattern. In both tasks, subjects were shown a series of sociated with spatial analytic processing. In a develop-
hierarchical forms (i.e., large patterns made of smaller mental study (Stiles et al., 1999), 15 healthy children
patterns, such as a large square made of small circles) between 11 and 15 years were tested using RT and fMRI
and, on separate blocks of trials, were asked to identify protocols identical to those used by Martinez and col-
targets at either the global or the local level. In the RT leagues (1997) with adult subjects. There were two ma-
study, stimuli were presented centrally, or to the right jor findings of this study. First, the pattern of RT data
or left visual half-fields (RVF, LVF), and the subject's obtained from children across the age span tested in
task was to detect when a target was present at the at- this study did not match that of adults. Across all age
tended level. In the fMRI study, the subject's task was groups, children were faster with global than with local
to count the number of predesignated targets in a rapid targets; however, within task, the same general patterns
series of centrally presented hierarchical stimuli. For of laterality as those observed with adults were obtained.
both RT and fMRI measures, a clear advantage for Second, in the fMRI, children's profiles of activation
global targets was associated with RH processing. Global differed from the patterns observed among adults. For
targets presented to the LVF-RH were processed faster both the attend global and attend local tasks, children
than local targets, and attention to global targets pro- showed statistically greater activation in the RH than in
duced markedly greater levels of brain activation than the LH (figure 27.2; see color plate 19). However, over-
attention to local targets. The opposite pattern of dis- all activation among children was greater than for
sociation between global and local targets was evident, adults, and children showed considerably more bilat-
but much less pronounced, in the LH. RTs to global eral activation, particularly on the local processing
and local targets were comparable. Both areas of acti- tasks, than did adults. These studies suggest that, at least
vation and intensity were marginally greater to the local for these perceptually demanding tasks, children show
than global targets in the LH. both a global processing advantage and overall RH
Studies of normally developing children have shown dominance.
that even young infants are capable of spatial analysis. An extensive series of studies of children with pre-
Cohen and Younger (1984) have shown evidence of sys- and perinatal focal brain injury has provided detailed
tematic change in the complexity of infants' ability to profiles of deficit and recovery for spatial functioning
process pattern information that reflects both global associated with early lateralized brain injury (Stiles et
and local level processing. Deruelle and de Schonen al., 1997, 1998; Stiles-Davis et al., 1988; Vicari et al.,
(1991, 1995) have shown that infants as young as 4 1998). These studies have shown that on construction
months show a profile of lateralized processing differ- and perception tasks, children with RH injury have dif-
ences on global and local processing tasks that is similar ficulty with spatial integration. While they are able to
to those described in the adult imaging study reported segment a spatial form into its elements, they have dif-
by Martinez and colleagues (1997). Change in the com- ficulty organizing those elements to form a coherent
plexity and sophistication of spatial analytic processing whole. A strikingly different profile of deficit is associ-
has been documented across the preschool and school ated with early LH injury. Children with LH injury over-
age period (Akshoomoff and Stiles, 1995a,b; Feeney simplify complex spatial forms and fail to encode the
and Stiles, 1996; Stiles and Stern, in press; Tada and details or elements of these forms. These two profiles
Stiles, 1996). Data from a large series of studies using are consistent with patterns of deficit reported for
different measures and testing children ranging in age adults with injury to comparable brain regions, and
from 3 to 12 years show that, initially, children segment suggest that the basic lateralized dissociation for global
out well-formed, independent units and use simple and local processing is robust from very early in
combinatorial rules to integrate the parts into an overall development.
configuration. With development, change is observed But while the basic adultlike patterns of deficit are
in both the nature of the parts and the relations evident even among very young children, the severity
408 COGNITION
of deficit is greatly attenuated among children, and chil- a complex geometric pattern that has been used for
dren appear to compensate in ways that adults cannot. many years to evaluate spatial planning in adult neu-
In addition to data documenting patterns of specific rological patients. This figure is organized around a
deficit, longitudinal studies of children with early focal central rectangle that is symmetrically divided by verti-
brain injury also provide evidence of functional recov- cal, horizontal, and diagonal bisecting lines; additional
ery. Longitudinal studies indicate that, with develop- pattern details are positioned both within and sur-
ment, children with both LH and RH injury show rounding the core rectangle. The most efficient and
considerable behavioral improvement, eventually advanced strategy for copying the ROCF is to begin with
achieving ceiling level performance on most spatial the core rectangle and bisectors, then add pattern de-
construction tasks. However, the time course over which tails. However, this strategy also places great demands
this improvement occurs is protracted, and several stud- on spatial analytic processing. Akshoomoff and Stiles
ies suggest that although the output of various spatial (1995a,b) have shown that typically developing children
processing tasks appears comparable to that of normal do not regularly use this advanced copying strategy until
peers, the processing strategies of children with early quite late in development. Younger children parse the
lesions continue to reflect persistent deficit. Two studies figure into smaller units and the size of the unit in-
illustrate this point. creases with age: 6-7-year-olds typically use a very piece-
A block construction study with 3-6-year-olds (Stiles meal strategy, drawing each small subdivision separately,
et al., 1996) demonstrated impairment among children while older children use progressively larger subunits
with both RH and LH injury. In this task children were (quadrants, halves) until, finally, by about 10-12 years,
presented with simple block models (e.g., a line of their organization strategy centers around the core
blocks or a double arch) and asked to reproduce them rectangle.
with the model present. Children with LH injury ini- Akshoomoff and Stiles (Akshoomoff and Stiles, in
tially showed delay, producing simplified constructions. press; Akshoomoff, Feroleto, Doyle, and Stiles, submit-
By the time they were 4 years old, they showed an in- ted) examined data collected longitudinally in 18 chil-
teresting dissociation in performance. Most of the chil- dren with pre- or perinatal focal brain injury from the
dren were able to produce accurate copies of the target time when they were 6 to 13 years of age. Two versions
constructions; however, the procedures they used in of the task were administered—a copy task, in which
copying the forms were greatly simplified. This disso- the children were asked to reproduce the figure with
ciation between product and process, which is not ob-
the model present, and an immediate memory task. On
served among normally developing children, persisted
the copying task, the children with both RH and LH
at least through age 6. Children with RH injury were
injury performed worse than normal age-matched con-
also initially delayed on this task, producing only sim-
trols. Deficits were particularly evident among the
plified constructions. At about age 4, they produced
youngest children (age 6-7) in the lesion groups, in
more complex, but disordered and poorly configured
that their drawings were sparser and less accurate than
constructions. At this time the procedures used to gen-
erate these ill-formed constructions were comparable to normal controls. However, there were no striking dif-
those of age-matched controls. By age 6, the perfor- ferences between LH and RH injury groups. With de-
mance profile for this group of children changed. They velopment, performance improved considerably, such
were able to accurately copy the target constructions, that by 9-10 years of age, these children were able to
but, like their LH injured peers, they used simple pro- produce reasonably accurate copies of the ROCF. How-
cedures. This study suggests that there is impairment in ever, analysis of how they generated the figure indicated
spatial processing following early injury, and there is continuing subtle deficit. Specifically, at 10-12 years of
compensation with development. However, close ex- age, the children in the lesion groups continued to use
amination of how spatial constructions are generated the most immature and piecemeal approach to gener-
suggest persistent deficits. These findings have been ating the figure. As was the case with the block construc-
replicated in a study of Italian children with localized tion task, the children's products improved, but
brain injury (Vicari et al., 1998). In addition, that study continuing deficit was evident in their construction pro-
showed that the profiles of deficit were evident among cedures. The failure to find differences between the RH
children with isolated subcortical injury, as well as in and LH injury groups on the ROCF task could reflect
children with injury involving both cortical and subcor- underlying task demands that place equal emphasis on
tical areas. the segmentation and integrative processes, and thus
A parallel set of findings was observed among older equally disrupt performance on the task.
children using a more difficult construction task, the Interestingly, while the copy task data failed to differ-
Rey-Osterrieth Complex Figure (ROCF). The ROCF is entiate the lesion groups, the memory task data did.
410 COGNITION
with a view of early brain development in which both development of spatial analysis. In The Handbook of Rey-
early specification and plastic adaptation play promi- Osterrieth Complex Figure Usage: Clinical and Research Applica-
tions,J. Knight and E. Kaplan, eds.
nent roles. Early injury constitutes a perturbation of AKSHOOMOFF, N. A., C. C. FEROLETO, R. E. DOYLE, AND J.
normal development. Specific neural resources are lost, STILES, submitted. The impact of early unilateral brain in-
and there should be consequent impairment of the sys- jury on perceptual organization and visual memory.
tem, and that is precisely what is observed. However, it ANDERSON, R. A., L. H. SNYDER, D. C. BRADLEY, and J. XING,
is also a developing system, and therefore a system with 1997. Multimodal representation of space in the posterior
an exuberance of resources, the fate of which is deter- parietal cortex and its use in planning movements. Annu.
Rev. Neurosd. 20:303-330.
mined in large measure by input. Thus, the magnitude ARENA, R., and G. GAINOTTI, 1978. Constructional apraxia
and duration of the initial impairment may well depend and visuoperceptive disabilities in relation to laterality of
on a range of factors such as the timing of insult, extent cerebral lesions. Cortex 14:463-473.
and location of injury, and specificity of the neural BAILLARGEON, R., and J. DEVos, 1991. Object permanence in
substrate. young infants: Further evidence. Child Dev. 62:1227-1246.
BAILLARGEON, R., and M. GRABER, 1988. Evidence of location
memory in 8-month-old infants in a nonsearch AB task. Dev.
ACKNOWLEDGEMENTS This work was supported by the Na-
Psychol. 24(4):502-511.
tional Institute of Child Health and Human Development
BELL, M. A., and N. A. Fox, 1992. The relations between fron-
Grant R01-HD25077, National Institute of Neurological Dis-
tal brain electrical activity and cognitive development dur-
orders and Stroke Grant #P50-NS22343, and National Insti-
ing infancy. Child Dev. 63:1142-1163.
tute of Deafness and Communicative Disorders Grant
BELL, M. A., and N. A. Fox, 1996. Crawling experience is re-
#P50-DC01289. The author thanks the parents and children
lated to changes in cortical organization during infancy: Evi-
for their participation in the studies presented in this article.
dence from EEG coherence. Dev. Psychol. 29(7):551-561.
BERTENTHAL, B. I., and J. J. CAMPOS, 1990. A systems approach
to the organizing effects of self-produced locomotion dur-
NOTES ing infancy. In Advances in Infancy Research, C. Rovee-Collier
1. It is possible that the effects of the postural manipulation and L. P. Lipsitt, eds. Norwood, N.J.: Ablex, pp. 1-60.
are not effective across a wide age range. All of the infants BISIACH, E., and C. LUZZATTI, 1978. Unilateral neglect of rep-
in the reported study were 10 months of age at time of resentational space. Cortex 14(1):129-133.
testing. Perhaps such manipulation would not be effective BOUSSAOUD, D., R. DESIMONE, and L. G. UNGERLEIDER, 1991.
with younger children. That is an empirical question. In any Visual topography of area TEO in the macaque. J. Comp.
case, maturation of frontal cortices alone cannot explain Neurol. 306(4):554-575.
these findings. BRODMANN, K,, 1909. Vergleichende Lokalisationslehre der Gros-
2. One could argue that early or even delayed experience af- shimrinde. Leipzig: J. A. Barth.
fects the onset of frontal lobe development, and that could BUTTER WORTH, G. E., N. JARRET, and L. HICKS, 1982. Spatio-
be true. Bell and Fox (1996) have shown an association temporal identity in infancy: Perceptual competence or
between self-locomotion and anterior-posterior EEG coher- conceptual deficit? Dev. Psychol. 18:435-449.
ence. These findings are consistent with a large body of CASEY, B. J., J. D. COHEN, K. O. CRAVEN, R. DAVIDSON, C. A.
evidence documenting experience-expectant or learning- NELSON, D. C. NOLL, X. Hu, M. LOWE, B. ROSEN, C. TRUWIT,
dependent change in other aspects of neural development and P. TURSKI, 1998. Reproducibility of fMRI results across
(e.g., Greenough, Black, and Wallace, 1987). However, this four institutions using a spatial working memory task. Neu-
kind of dynamic interaction is different from a simple mat- roimage 8:249-261.
uration account. If experience induces development in CHIBA, A. A., R. P. KESNER, and C. J. GIBSON, 1997. Memory
frontal regions, what are the consequences of this kind of for temporal order of new and familiar spatial location se-
early or late induction for the rest of the developing cog- quences: Role of the medial prefrontal cortex. Learn. Mem.
nitive or neural system? 4:311-317.
CHUGANI, H. T, M. E. PHELPS, and J. C. MAZZIOTTA, 1987.
Positron emission tomography study of human brain func-
REFERENCES tional development. Ann. Neurol. 22(4):487-497.
AHMED, A., and T. RUFFMAN, 1998. Why do infants make A not CLOHESSY, A. B., M. I. POSNER, M. K. ROTHBART, and S. P.
B errors in a search task, yet show memory for the location VECERA, 1991. The development of inhibition of return in
of hidden objects in a nonsearch task? Dev. Psychol. early infancy.J. Cogn. Neurosd. 3(4):345-350.
34(3) :441-453. COHEN, L. B., and B. A. YOUNGER, 1984. Infant perception of
AKSHOOMOFF, N. A., and J. STILES, 1995a. Developmental angular relations. Infant Behav. Dev. 7(l):37-47.
trends in visuospatial analysis and planning: I. Copying a CORBETTA, M., 1998. Frontoparietal cortical networks for di-
complex figure. Neuropsychology 9(3):364-377. recting attention and the eye to visual locations: Identical,
AKSHOOMOFF, N. A., and J. STILES, 1995b. Developmental independent, or overlapping neural systems? Proc. Natl.
trends in visuospatial analysis and planning: II. Memory for Acad. Sd. USA 95(3):831 -838.
a complex figure. Neuropsychology 9(3):378-389. CORBETTA, M., F. M. MIEZIN, G. L. SHULMAN, and S. E. PE-
AKSHOOMOFF, N. A., and J. STILES, in press. Children's per- TERSEN, 1993. A PET study of visuospatial attention.J. Neu-
formance on the Rey-Osterrieth Complex Figure and the rosd. 13(3):1202-1226.
412 COGNITION
JOHNSON, M. H., M. I. POSNER, and M. K. ROTHBART, 1994. McFiE, J., M. F. PIERCY, and O. L. ZANGWILL, 1950. Visual-
Facilitation of saccades toward a covertly attended location spatial agnosia associated with lesions of the right cerebral
in early infancy. Psychol. Sci. 5(2):90-93. hemisphere. Brain 73:167-190.
JOHNSON, M. H., and L. A. TUCKER, 1996. The development McFiE, J., and O. L. ZANGWILL, 1960. Visual-constructive dis-
and temporal dynamics of spatial orienting in infants.J. Exp. abilities associated with lesions of the left cerebral hemi-
Child Psychol. 63:171-188. sphere. Brain 83:243-259.
JOHNSON, M. H., L. A. TUCKER, J. STILES, and D. TRAUNER, MERIGAN, W. H., and J. MAUNSELL, 1993. How parallel are the
1998. Visual attention in infants with perinatal brain dam- primate visual pathways? Annu. Rev. Neurosci. 16:369-402.
age: Evidence of the importance of anterior lesions. Dev. Sci. NELSON, C. A., 1999. Neural plasticity and human develop-
l(l):53-58. ment. Curr. Direct. Psychol. Sci. 8(2):42-45.
JOHNSON, P. B., S. FERRAINA, L. BIANCHI, and R. CAMINITI, NELSON, C. A., J. LIN, L. J. CARVER, C. S. MONK, K. M. THOMAS,
1996. Cortical networks for visual reaching: Physiological and C. L. TRUWIT, 2000. Functional neuroanatomy of spa-
and anatomical organization of frontal and parietal lobe tial working memory in children. Dev. Psychol. 36:109-116.
arm regions. Cereb. Cortex6(2):102-119. PALMER, S. E., 1977. Hierarchical structure in perceptual rep-
JONIDES, J., E. E. SMITH, R. A. KOEPPE, E. AWH, S. MINOSHIMA, resentation. Cogn. Psychol. 9(4):441-474.
and M. A. MINTUN, 1993. Spatial working memory in hu- PALMER, S. E., 1980. What makes triangles point: Local and
mans as revealed by PET. Nature 363:623-625. global effects in configurations of ambiguous triangles.
KERMOIAN, R., andJ.J. CAMPOS, 1988. Locomotor experience: Cogn. Psychol. 12(3):285-305.
A facilitator of spatial cognitive development. Child Dev. PALMER, S. E., and N. M. BUCHER, 1981. Configural effects in
59(4):908-917. perceived pointing of ambiguous triangles.J Exp. Psychol.:
KITTERLE, F. L., S. CHRISTMAN, andj. HELLIGE, 1990. Hemi- Hum. Percept. Perform. 7(1):88-114.
spheric differences are found in the identification, but not PIAGET,J., 1952. The origins of intelligence in children. New
detection of low versus high spatial frequencies. Percept. Psy- York: International Universities Press.
chophysics 48:297-306. POSNER, M. I., 1980. Orienting of attention. Quart. J. Exp. Psy-
KITTERLE, F. L.,J. B. HELLIGE, and S. CHRISTMAN, 1992. Visual
chol. 32:3-25.
hemispheric asymmetries depend on which spatial frequen-
POSNER, M. I., and Y. COHEN, 1980. Attention and the control
cies are task relevant. Brain Cogn. 20:308-314.
of movements. In Tutorials in Motor Behavior, G. E. Stelmach
KITTERLE, F. L., and L. M. SELIG, 1991. Visual field effects in
and J. Requin, eds. Amsterdam: North Holland Publishing,
the discrimination of sine wave gratings. Percept. Psychophysics
pp. 243-258.
50:15-18.
POSNER, M. I., and S. E. PETERSEN, 1990. The attention system
LAMB, M. R., L. C. ROBERTSON, and R. T. KNIGHT, 1989. At-
of the human brain. Annu. Rev. Neurosci. 13:25-42.
tention and interference in the processing of global and
POSNER, M. L, R. D. RAFAL, L. S. CHOATE, and J. VAUGHN, 1985.
local information: Effects of unilateral temporal-parietal
junction lesions. Neuropsychologia 27(4):47l-483. Inhibition of return: Neural basis and function. Cogn. Neu-
LAMB, M. .R., L. C. ROBERTSON, and R. T. KNIGHT, 1990. Com- ropsychol. 2(3) :211-228.
ponent mechanisms underlying the processing of hierar- POSNER, M. I., J. A. WALKER, F. J. FRIEDRICH, and R. D. RAFAL,
chically organized patterns: Inferences from patients with 1984. Effects of parietal injury on covert orienting of atten-
unilateral cortical lesions.J Exp. Psychol.: Learn. Mem. Cogn. tion.J Neurosci. 4 (7): 1863-1874.
16:471-483. RAFAL, R. D., 1994. Neglect. Curr. Opin. Neurobiol. 4:231-236.
LIVINGSTONE, M. S., and D. H. HUBEL, 1984. Anatomy and RAFAL, R. D., and L. ROBERTSON, 1995. In The Neurology of
physiology of a color system in primate visual cortex.J Neu- Visual Attention, M. S. Gazzaniga, ed. Cambridge, Mass.: MIT
rosci. 4:309-356. Press.
MALACH, R.,J. B. REPPAS, K. K. KWONG, H.JIANG, W. A. KEN- ROBERTSON, L. C., 1992. Perceptual organization and atten-
NEDY, P. J. LEDDEN, T. J. BRADY, B. R. ROSEN, and R. B. Too- tional search in cognitive deficits. In Cognitive Neuropsychol-
TELL, 1995. Object-related activity revealed by functional ogy in Clinical Practice, D. I. Margolin, ed. New York: Oxford
magnetic resonance imaging in human occipital cortex. University Press,
Proc. Natl. Acad. Sci. USA 92(18):8135-8139. ROBERTSON, L. C., and D. C. DELIS, 1986. "Part-whole" pro-
MARANGOLO, P., E. Di PACE, R. RAFAL, and D. SCABINI, 1998. cessing in unilateral brain damaged patients: Dysfunction
Effects of parietal lesions on humans on color and location of hierarchical organization. Neuropsychologia 24(3):363-
priming.J Cogn. Neurosci. 10(6):704-716. 370.
MARTIN, M., 1979. Hemispheric specialization for local and ROBERTSON, L. C., M. R. LAMB, and R. T. KNIGHT, 1988. Effects
global processing. Neuropsychologia 17:33-40. of lesions of temporal-parietal junction on perceptual and
MARTINEZ, A., L. ANLLO-VENTO, and S. A. HILLYARD, 1997. attentional processing in humans.J Neurosci. 8:3757-3769.
Brain electrical activity during selective attention to spatial SCHNEIDER, G. E., 1967. Contrasting visuomotor functions of
frequency. Soc. Neurosci. Abstr. tectum and cortex in the golden hamster. Psychol. Forsch.
MARTINEZ, A., P. MOSES, L. FRANK, R. BUXTON, E. WONG, and 31(l):52-62.
J. STILES, 1997. Hemispheric asymmetries in global and lo- SELTZER, B., and D. N. PANDYA, 1978. Afferent cortical con-
cal processing: Evidence from fMRI. NeuroReport 8:1685- nections and architectonics of the superior temporal sulcus
1689. and surrounding cortex in the rhesus monkey. Brain Res.
MATTHEWS, A., A. E. ELLIS, and C. A. NELSON, 1996. Devel- 149(l):l-24.
opment of preterm and full-term infant ability on AB, recall SERGENT, J., 1982. The cerebral balance of power: Confron-
memory, transparent barrier detour, and means-end tasks. tation or cooperation? J Exp. Psychol.: Hum. Percept. Perform.
Child Dev. 67:2658-2676. 8:253-272.
414 COGNITION
28 Bridging the Gap between Cognition
and Developmental Neuroscience:
The Example of Number
Representation
SUSAN CAREY
ABSTRACT Developmental cognitive neuroscience necessar- between neuroanatomy, neurochemistry, and neuro-
ily begins with a characterization of the developing mind. One development, on the one hand, and the behavioral re-
cannot discover the neural underpinnings of cognition with- search in cognition, on the other, are rather tenuous."
out detailed understanding of the representational capacities
(See Nelson and Bloom, 1997, for a similar lament.)
that underlie thought. Characterizing the developing mind
involves specifying the evolutionarily given building blocks What is needed to bridge the gap, to forge less tenuous
from which human conceptual abilities are constructed, de- connections?
scribing what develops, and discovering the computational
mechanisms that underlie the process of change. Here, I pres- SEEKING STRUCTURAL PARALLELS; STRUCTURAL
ent the current state of the art with respect to one example of ANALOGY Developmental neurobiology aims to under-
conceptual understanding: the representation of number. stand the mechanisms underlying cell differentiation,
Lessons for developmental cognitive neuroscience are drawn cell formation, cell migration, axon migration, synapse
at two levels of analysis: first, structural analogy between de-
velopmental processes described at the neural level and at the
formation, dendrite growth, and so forth. Developmen-
cognitive level, and second, challenges posed at the level of tal psychology (of humans) aims to understand the for-
systems neuroscience. mation of the functional architecture of the mind (how
memory, emotion, attention, executive function, etc.,
come into being) as well as the origins of concepts, lan-
Bridging biology and psychology guage, and the computational capacities that constitute
human reason.
The past 20 years have witnessed stunning advances in Biology and psychology provide different levels of
our understanding of how the brain develops (for re- analysis, but practitioners of the two disciplines share
views aimed at drawing implications for cognitive neu- fundamental intellectual goals. Both seek to under-
roscience, see Gazzaniga, 1995, 1999; Johnson, 1997; stand the role of the genetic code in structuring the
Quartz and Sejnowski, 1997). Similarly, new techniques developing brain and mind, and both seek to specify
for experimental work with human infants have yielded the huge variety of processes that determine the path
a wealth of evidence concerning the unfolding of early to the adult state. Both assume that the functional ar-
perceptual, conceptual, and linguistic capacities (for il- chitecture of the brain, its connectivity and patterns of
lustrative studies, see Baillargeon, 1993; Carey and neural activity, provide the underpinnings of the mind,
Spelke, 1994; Kuhl, 1999; Leslie, 1994; Mandler, 2000; and both assume that a functional account of what the
Marcus et al., 1999; Saffran, Aslin, and Newport, 1996; brain does (i.e., an understanding of the mind) is a
Spelke et al., 1992; Teller, 1999). In spite of the ad- necessary guide to our understanding of how the brain
vances on each of these fronts, Rakic's comment in his works. Therefore, detailed understanding of mecha-
introduction to the Neural and Psychological Develop- nism at one level of analysis is, at the very least, a source
ment section of the 1995 version of The Cognitive Neu- of structural analogy for mechanism at the other.
We appeal to a structural analogy whenever we see a
rosciences (Rakic, 1995) still stands: "Yet the connections
process at one level as similar to a process at the other,
in the absence of any claim that we have reduced one
SUSAN CAREY Department of Psychology, New York Univer- to the other or that we have actually discovered the
sity, New York, New York. neural underpinnings of some cognitive ability or some
416 COGNITION
structural analogy in our thinking about how the de- bolstering the inference that normal development in-
veloping brain underlies and in turn is shaped by the volves dorsolateral prefrontal maturation.
developing mind. This particular case is of great interest to develop-
mental psychologists because of the observation that
CLOSING THE GAP—THE LEVEL OF SYSTEMS NEUROSCI- the delayed response task is almost identical to Piaget's
ENCE, DIAGNOSTIC MARKERS Besides seeking struc- object permanence task (Piaget, 1955). Diamond and
tural parallels between mechanisms of developmental Goldman-Rakic (1989; see also Diamond, 1991) estab-
change specified at different levels, there are many lished that infant monkey development on the A/not
more direct methods of building bridges between the B object permanence task between ages 2 and 4 months
two disciplines, as the chapters in the present volume matches, in parametric detail, human infant develop-
attest. The most obvious derives from a premise that ment between 7 and 11 months, implicating frontal de-
articulates much of the discipline of cognitive neuro- velopment in the human case as well. Confirmatory
science: Large regions of the brain participate in spe- evidence is provided by development of other marker
cialized neural circuits, the functions of which may be tasks for frontal development, such as detour reaching,
studied through a combination of research methods— that are conceptually unrelated to object permanence.
behavioral, imaging, computational, and lesion (exper- Thus, this work provides an alternative explanation to
imentally produced in animal studies, neuropsycholog- Piaget's for young infants' failures on object perma-
ical studies with humans). One descriptive goal of nence tasks. Rather than reflecting different represen-
developmental cognitive neuroscience is to discover tations of objects, as Piaget believed, the failures reflect
when functions known to be subserved by specific brain immature executive function (means/ends planning,
regions come on-line. In some cases, young children fail inhibition of competing responses, deployment of
tasks on which failure is a diagnostic marker of specific working memory).
brain damage in adults. Such a finding is a source of We do not yet know the nature of the maturational
hypothesis—perhaps the child's failure is due to im- change involving prefrontal dorsolateral cortex be-
maturity of the neural structure required for adult per- tween 2 and 4 months in rhesus and between 7 and 11
formance. Of course, there are always alternative months in humans, specified at the level of the neuro-
explanations for the young child's failure. It may be, for biological mechanisms. The method of diagnostic
instance, that the child simply has not yet mastered the markers does not completely close the gap; it establishes
knowledge required to perform the task. Such alter- phenomena related to developing brain function for
natives may be empirically tested. Notice that such re- which neurobiological underpinnings are yet to be
search only partially fills the gap—and that, only insofar worked out.
as the neural substrate of the computations carried out
by the circuit is fully understood, as well as the neuro- OTHER PARTIAL STEPS The method of diagnostic
biological mechanisms underlying the development of markers depends upon studies of brain-damaged hu-
the circuit. These conditions, I submit, have not yet man beings and lesion studies in animals. Studies of
been met in any domain of perceptual, conceptual, or brain-damaged children (whether due to lesions or ge-
linguistic representation. netic abnormalities) add relevant data concerning
Goldman-Rakic (Goldman, 1971) provided a para- dissociations of function and plasticity during develop-
digmatic example of diagnostic marker research. She ment. Noninvasive imaging techniques add new meth-
noted that young rhesus monkeys (less than 2 months ods to the arsenal of tools for bridging the disciplines
of age) performed as did adult monkeys with lesions in at the level of systems neuroscience (for a discussion of
dorsolateral prefrontal cortex on the marker task of de- how imaging techniques might serve this purpose, see
layed response. Shown food hidden in one of two or Nelson and Bloom, 1997). These tools are by nature
more wells, distracted, and then allowed to search, adult interdisciplinary; they require precise psychological
monkeys with prefrontal lesions (and young infant task analysis to provide interpretable data concerning
monkeys) often fail to retrieve the food, especially with the functions of the brain regions implicated in any
repeated trials and with delays of more than a few sec- given lesion or imaging study.
onds. Rather, a variety of perseverative patterns of re- A common thread runs through my remarks so far—
sponse emerge. The infant performance develops to the importance of understanding the mind to under-
adult levels by 4 months of age. Infant monkeys lesioned standing the brain. Before turning to an extended
in dorsolateral prefrontal cortex do not undergo nor- developmental example, I explicitly defend this
mal development; their performance never improves, assumption.
418 COGNITION
studies adequately controlled for nonnumerical bases of ory in prelinguistic infants. In sum, the violation of ex-
discrimination, but others (e.g., Wynn, 1996; Xu and pectancy studies and the choice studies suggest that
Spelke, 2000) did. Insofar as the controls are adequate, preverbal infants discriminate among sets on the basis
these studies reveal representation of number in the of numerosity and represent numerical relations
sense of discrimination on the basis of numerical dif- among sets, such as 1 + 1 — 2 and 2 > 1. Numerical
ferences. But confidence in these representations as representations predate language learning, and thus
number representations requires more; it is important the mechanisms by which they are formed do not in-
that computations defined over them have at least some volve mastering the culturally constructed, explicit,
of the conceptual role of number representations. This integer-list representations of natural language. Indeed,
condition is met: Infants use their representations of these demonstrations open the question of continuity.
number to compute more/less and carry out simple Is the integer-list representation ("one, two, three, four,
addition and subtraction. five, . . .") and the counting algorithm merely an ex-
Some of the evidence for number-relevant computa- plicit form of the representational system underlying
tions derives from experiments using the violation of these infant capacities? Or is the integer-list represen-
expectancy looking time method. Wynn's (1992a) clas- tation discontinuous with those that underlie the in-
sic addition/subtraction studies, now replicated in fant's behavior? To begin to answer this question, we
many laboratories (Koechlin, Dehaene, and Mehler, must consider the format of infants' representation of
1997; Simon, Hespos, and Rochat, 1995; Uller et al., number.
1999) suggest that babies represent the numerical re-
lations among small sets of objects. In a typical experi- GELMAN AND GALLISTEL'S CONTINUITY HYPOTHESIS
ment, an object is placed on an empty stage while the Gelman and Gallistel (1978) suggested that infants es-
baby watches; then a screen is raised that covers the tablish numerical representations through a nonverbal
object, a second object is introduced behind the screen, counting procedure: Babies represent a list of symbols,
and the screen is lowered to reveal either one object or or "numerons," such as &, ^, #, $, @. Entities to be
two. In this 1 + 1 = 2 vs. 1 event, infants of ages ranging counted are put in 1:1 correspondence with items on
from 4 to 10 months look longer at the impossible out- this list, always proceeding through it in the same order.
come of 1 than at the possible outcome of 2. That in- The number of items in the set being counted is rep-
fants expect precisely 2 objects is shown by the fact that resented by the last item on the list reached, and its
they also look longer at the impossible outcome of 3 in numerical value is determined by the ordinal position
a 1 + 1 = 2 vs. 3 event (Wynn, 1992a). Infants also of that item in the list. For example, in the list &, ^, #,
succeed at 2 - 1 = 2 vs. 1. Success with larger numbers $, @, the symbol ^ represents 2, because ^ is the second
such as 3 - 1 and 2 + 1 is less robust (Baillargeon, item in the list.
Miller, and Constantino, 1993; Uller and Leslie, 2000). Gelman and Gallistel's proposal for the nonlinguistic
Preliminary data from a third paradigm, developed representation of number is a paradigm example of a
by Hauser, Carey, and Hauser (2000) to study sponta- continuity hypothesis, for this is exactly how languages
neous representation of number by free-ranging rhesus with explicit integer lists represent the positive integers.
macaques, suggests that prelinguistic infants represent On their hypothesis, the child learning "one, two, three,
the ordinal relations among small sets of different num- four, five, . . . " need only solve a mapping problem:
bers of objects. Each infant gets only one trial; these are identify the list in their language that expresses the an-
totally spontaneous number representations. Ten- and tecedently available numeron list. Originally learning to
12-month-old infants watched as an experimenter put count in Russian should be no more difficult than learn-
a number of small cookies successively into one opaque ing to count in Russian once one knows how to count
container and a different number into a different in English.
opaque container, whereupon they were allowed to
crawl to the container of their choice. Infants of both WYNN'S EASE OF LEARNING ARGUMENT AGAINST THE
ages approached a container with 2 cookies over a con- GELMAN/GALLISTEL CONTINUITY PROPOSAL Between
tainer with just 1 cookie, and a container of 3 cookies the ages of 2 and 4 years, most children learn to count;
over one with just 2 cookies. Infants of both ages failed and contrary to the predictions of a numeron-list the-
at choices involving 3 vs. 4, as well as at choices involving ory, learning to count is not a trivial matter (Fuson,
3 vs. 6; see figure 28.1 (Feigenson et al., 2000). 1988; Wynn, 1990, 1992b). Of course, this fact in itself
As in Wynn's paradigm, infants must compute 1 + 1 does not defeat the continuity hypothesis, for we do not
or 1 + 1 + 1. In addition, this is the first demonstration know in advance how difficult it is to learn an arbitrary
of ordinal comparison of two sets represented in mem- list of words or to discover that one such list (e.g., "one,
two, three, . . . " rather than "a, b, c, . . . " or "Monday, ber of objects (always more than one) and handed them
Tuesday, Wednesday,. . .") is the list in English that rep- to the experimenter. Also, shown two cards depicting,
resents number. However, Wynn's (1990,1992b) studies for example, 2 vs. 3 balloons and asked to indicate
show that children have difficulty discovering the mean- which card had 2 balloons on it, young children re-
ings of specific number words even after they have sponded at chance. Analyses of within-child consistency
solved these two problems. The sequence in which chil- between the "give n" and "which card has n" tasks bol-
dren learn the meanings of the number words there- stered the conclusion that young children count for
fore is at odds with that predicted by the continuity more than a year before they learn what the words in
thesis. the count sequence mean: i.e., before they learn how
Wynn began by confirming that young children do "one, two, three, four, five, . . . " represents number.
not know the numerical meanings of the words in the There is one more observation of Wynn's that is im-
count sequence. First, she identified children who portant to the evaluation of the Gelman/Gallistel con-
could count at least to six when asked "how many" ob- tinuity hypothesis. Wynn (1990, 1992b) showed that
jects there were in an array of toys. These 2-3-year-old from the beginning of learning to count, children know
children honored 1:1 correspondence in their counts, what "one" means. They can pick one object from a pile
and used a consistently ordered list, although some- when asked, and they correctly distinguish a card with
times a nonstandard one such as "one, two, four, six, "one fish" from a card with "three fish." Furthermore,
seven, . . . " She then showed that if such a child were they know that the other words in the count sequence
given a pile of objects and asked to give the adult "two" contrast with one. They always grab a random number
or "three" or any other number the child could use in of objects greater than one when asked to hand over
the game of counting, most 2- and 3-year-old children "two, three, four, . . . " objects, and they also successfully
failed. Instead, young children grabbed a random num- point to a card with three fish when it is contrasted with
420 COGNITION
a card with one, even though their choices are random of number (for reviews, see Gallistel, 1990; Dehaene,
when "three" is contrasted with "two." Thus, Wynn's 1997). Rather than being represented by a list of dis-
studies provide evidence that toddlers learn the English crete symbols, in such systems number is represented
count list and identify the list as relevant to number very by a physical magnitude that is proportional to the num-
early on (younger than age 2H): They know what "one" ber of individuals in the set being enumerated. An ex-
means, they use "two, three, four, etc.," in contrast with ternal analog-magnitude representational system could
"one," and they only use the number words above one represent one as , two as , three as
when presented with sets greater than one. They are in , and so on. In such systems, numerical com-
this state of knowledge for a full year before they work parisons are made by processes that operate over these
out the principle that allows them to determine which analog magnitudes, in the same way that length or time
number each numeral refers to. This state of affairs is comparisons are made by processes that operate over
impossible on the numeron-list continuity hypothesis, representations of these physical magnitudes. For this
whereby the English count list need only be identified reason, number comparisons are subject to Weber's law
and mapped onto the preexisting nonlinguistic nume- (1 and 2 are more discriminable than are 7 and 8).
ron list that the infant already uses to represent number. Sensitivity to Weber's law is one of the main sources of
Wynn's work is important for another reason, be- evidence that both animals and human adults pre-
cause her data constrain our hypotheses about the pro- vented from verbal counting deploy analog-magnitude
cess through which children construct the integer-list representations of number (for reviews of other sources
representation. Her studies show that children learn of evidence, see Dehaene, 1997; Gallistel, 1990).
the meanings of the number words as follows: First, they There are many different ways that analog-magnitude
learn what "one" means, as indicated above, with other representations of number might be constructed. One
number words contrasting with "one"; then they learn proposal is the accumulator model of Meek and Church
what "two" means, with other words contrasting with (1983). The idea is simple: Suppose the nervous system
"two" and referring to numbers bigger than 2. Wynn's has the equivalent of a pulse generator that generates
longitudinal data established that some children are in activity at a constant rate, and a gate that can open to
this state for several months. Then they learn what allow energy through to an accumulator that registers
"three" means; some children induce the meaning of how much has been let through. When the animal is in
all the number words in the list at this point; others a counting mode, the gate is opened for a fixed amount
know what "three" means, and take the other words to of time (say 200 ms) for each individual to be counted.
contrast with "three" and refer indiscriminately to The total energy accumulated then serves as an analog
higher numbers. Wynn found no children who knew representation of number. Meek and Church's model
what "four" meant who had not worked out how the seems best suited for sequentially presented individuals,
whole list represents number. such as bar presses, tones, light flashes, or jumps of a
In spite of the evidence that prelinguistic infants rep- puppet. Gallistel (1990) proposed, however, that this
resent number, it seems that the child is still in the pro- mechanism functions as well in the sequential enumer-
cess of constructing an integer-list representation of ation of simultaneously present individuals. Others
number in the years 2-4. How then do infants repre- (e.g., Dehaene and Changeux, 1989; Church and
sent number, and how do their representations differ Broadbent, 1990) have proposed computational mod-
from the integer-list representation? els that create analog-magnitude representations with-
out implementing any iterative process. Furthermore,
Two POSSIBLE PRELINGUISTIC REPRESENTATIONAL there is unequivocal evidence that in at least some cir-
SYSTEMS FOR NUMBER Two quite different nonlinguis- cumstances, number representations are created by
tic systems of number representation have been pro- noniterative processes. For example, the time that sub-
posed to underlie infants' abilities to discriminate, jects require to discriminate two numerosities depends
compare, and compute over representations of num- on the ratio difference between the numerosities, not
ber: analog-magnitude representations and object-file on their absolute value (Barth, Kanwisher, and Spelke,
representations. I characterize each in turn, consider- under review). In contrast, time should increase mon-
ing the evidence for and against the hypothesis that otonically with N for any iterative counting process.
each one may underlie the behaviors in which infants Moreover, subjects are able to discriminate visually pre-
display knowledge of number. sented numerosities under conditions of stimulus size
and eccentricity in which they are not able to attend to
Analog-magnitude representations Both human adults and individual elements in sequence (Intrilligator, 1997).
animals deploy analog-magnitude representations Their numerosity discrimination therefore could not
422 COGNITION
unlikely to underlie performance on many of the tasks explores an array determines a set of attended objects.
involving small set sizes. If the perceiver can establish a 1:1 correspondence be-
tween objects and active files, therefore, the computa-
Object-file representations A wealth of studies provide evi- tions over object-file representations provide a process
dence that human adults can attend to small numbers for establishing numerical equivalence. Fourth, object
of objects simultaneously, keeping track of each object files represent numerosity exactly for set sizes up to
by following it as it moves continuously and accumulat- about 4 and are not subject to Weber's law.
ing information about its properties (Kahneman, Treis- Object-file representations differ from integer-list
man, and Gibbs, 1992; Pylyshyn and Storm, 1988). representation in three ways. Most importantly, they
Adults are able to perform this task even when distinct contain no symbols for cardinal values. The only sym-
objects have no distinguishing properties other than bols in such models represent the individual objects
their location, and when a single object maintains no themselves. Second, object-file models have an upper
common properties over time as it moves. Conversely, bound (of about 4; see Pylyshyn and Storm, 1988; Trick
when two objects with distinct properties share a com- and Pylyshyn, 1994). Third, object-file representations
mon location (for example, a small circle is embedded serve only to represent individuals that obey the spatio-
in a large one), or when a single object moves discon- temporal constraints on objects; they do not serve to
tinuously, adults sometimes fail to represent the two ob- enumerate events or other entities. Of course, the ex-
jects' distinctness or the single object's identity (Scholl istence of object-file representations does not preclude
and Pylyshyn, 1999; Trick and Pylyshyn, 1994). All these similar representational systems for other types of
phenomena testify to the existence of a spatiotemporal individuals (event-file representations, syllable-file
system for representing objects—up to about four of representations).
them at once—in visual scenes. Such object represen- Many have suggested that infants construct mental
tations have been proposed to underlie adults' rapid models of the objects in number discrimination and ad-
apprehension of numerosity in arrays of simultaneously dition experiments, deploying the object tracking ca-
visible elements, when each element occupies a distinct pacities studied in the adult object-based attention
location and the total number of elements does not ex- literature to update these representations as objects are
ceed about 4 (Trick and Pylyshyn, 1994). Such repre- added and subtracted from the arrays (Carey and Xu,
sentations also have been proposed to underlie adults' in press; Leslie and Scholl, 1999; Simon, 1997; Uller et
rapid apprehension of the identity of individual visible al., 1999). That is, infants may be constructing a rep-
objects through time and their accumulation of infor-
resentation consisting of one object file for each object
mation about the properties of each object. These rep-
in a given array. Object-file representations could un-
resentations have been called by different names, and
derlie infants' discrimination of two individuals from
different theorists do not exactly agree as to their na-
three individuals; in these representations, the small
ture or properties. Here, I abstract over these differ-
sets of objects differ as 0 0 differs from 0 0 0. Object-
ences and, following Kahneman and colleagues (1992),
refer to such representations as "object files." file representations also provide a natural account for
Object-file representations are numerical in four performance in the infant addition experiments. Dur-
senses. First, the opening of new object files requires ing the setup o f a l + l = 2 or l experiment, for
principles of individuation and numerical identity; example, the infant would open an object file for the
models must keep track of whether this object, seen first object introduced on the stage and a second object
now, is the same one as that object seen before. Spatio- file for the second object introduced behind the screen,
temporal information must be recruited for this pur- yielding the representation 0 0. Then when the out-
pose, because the objects in many experiments are come array is revealed, the infant again creates object
physically indistinguishable from each other and be- file representations, 0 0 for the expected outcome of
cause, in any case, property/kind changes within an ob- two objects and 0 for the impossible outcome of 1 ob-
ject are not sufficient to cause the opening of a new ject. The model of the setup array and the model of the
object file (for reviews, see Carey and Xu, in press; Py- outcome array are compared on the basis of 1:1 corre-
lyshyn, in press). Second, the opening of a new object spondence, and a mismatch draws the infant's atten-
file in the presence of other active files provides a nat- tion. Evidence in favor of this account comes from the
ural representation for the process of adding one to an finding that infants do not use property/kind infor-
array of objects. Third, object-file representations pro- mation (e.g., the information that distinguishes a cat
vide implicit representations of sets of objects; the ob- from a dog) to open new object files in addition exper-
ject files that are active at any given time as a perceiver iments (Simon, Hespos, and Rochat, 1995; Xu and
424 COGNITION
large sets of objects in discrimination experiments, they trast, capture all relations among whole numbers, no
appear to represent those sets with analog-magnitude matter how close (14,15,114, 115, 116, etc.). Moreover,
representations. When they are presented with small analog-magnitude representations, especially those pro-
sets of objects in discrimination experiments, they may duced by a noniterative process, obscure the arithmetic
not represent numerosity at all. Finally, it is not clear successor relation (+ 1) between successive states in two
whether infants keep track of small numbers of events ways. First, there is no operation within the process that
by means of analog-magnitude representations or by an constructs these representations that corresponds to
entirely different system of parallel individuation. The the operation of adding one. Second, Weber fraction
hypothesis that human infants are endowed both with considerations make the difference between the states
analog magnitude and object-file representations of that represent 1 and 2 appear larger than the difference
number is not implausible, because a wide variety of between 2 and 3, or 29 and 30.
species, including rats, parrots, chickens, pigeons, and Even if the infant is endowed with both analog mag-
primates, command both systems of representations nitude and object file systems of representation, there-
(for reviews, see Gallistel, 1990; Dehaene, 1997). fore, the infant's capacity to represent number will be
markedly weaker than that of the child who commands
WHY THE INTEGER-LIST REPRESENTATION Is A GENU- the integer-list representation. Neither object files nor
INELY NEW REPRESENTATIONAL RESOURCE The integer- analog magnitudes can serve to represent large exact
list representation differs deeply from both object-file numerosities: Object files fail to capture number con-
and analog-magnitude representations of number, and cepts such as "seven" because they exceed the capacity
it transcends each of them singly and together. Object- limit of four, and analog-magnitude representations fail
file representations are sharply limited to small sets by to capture such concepts because they exceed the limits
constraints on parallel individuation and short-term on their precision. Children endowed with both these
memory for distinct items, whereas integer-list repre- representations therefore lack the representational re-
sentations contain no such limit. There is no intrinsic sources provided by the integer-list system.
upper bound on the lists that one can commit to long- In sum, a consideration of the format of infant num-
term memory (think of the alphabet). Once an integer ber representation reinforces the conclusions from
list embodies a recursive component for generating Wynn's ease of learning argument. The integer-list rep-
symbols in the list, there is no upper limit at all (think resentation is discontinuous with its developmental pre-
of the base-10 system). Moreover, object-file represen- cursors; acquiring it requires the construction of a new
tations contain no symbol for any integer value and no representational resource. How do children build an
counting algorithm, even implicitly, whereas integer-list explicit, integer-list, representation of the natural
representations are constituted by lists of explicit sym-
numbers?
bols for numerosities. Nonetheless, object-file represen-
tations have two important properties in common with
integer-list representations. First, principles of indi- Constructing the integer-list representation
viduation and numerical identity, plus the capacity to of number
compare models on the basis of 1:1 correspondence,
enable sets of 1, 2, and 3 individuals (1 + 1 = 2, 2 > The problem of how the child builds an integer-list rep-
1, etc.) to be precisely, if implicitly, represented. Second, resentation decomposes into the related subproblems
object files provide a natural representation of the suc- of learning the ordered list itself ("one, two, three, four,
cessor function, for adding one corresponds to the open- five, six, . . ."), learning the meaning of each symbol on
ing of a new object file. In integer-list representations, the list (e.g., "three" means three), and learning how the
the successor relation (+ 1) between adjacent numbers list itself represents number such that the child can in-
similarly enables the numerical relations among sets of fer the meaning of a newly mastered integer symbol
individuals to be precisely encoded. (e.g., "eleven") from its position in the integer list. On
Analog-magnitude representations, like integer-list Gelman and Gallistel's (1978) continuity proposal, the
representations, have no intrinsic upper limit and con- second and third questions collapse into the first—the
tain a distinct symbol for each integer value represented child identifies a natural language list such as "one, two
(e.g., represents 2). Discriminability of these three four, five six, ..." as an integer list that corre-
symbols, however, is limited by Weber's law, such that sponds to the mental numeron list, and in doing so
one cannot represent the precise relation between 14 knows the meanings of all its symbols at once. Empirical
and 15, although one may be able to represent that 15 considerations against this proposal were presented
is greater than 12. Integer-list representations, in con- earlier.
426 COGNITION
cognition. But before such behavioral data can inform I have provided evidence for two systems of numeri-
debates at the level of structural analogy between cog- cally relevant representations (object-file representa-
nitive and neurobiological mechanisms, it is important tions, analog-magnitude representations) that become
to discover the nature of the representational systems integrated, during development, in the creation of a
that subserve behavior on these tasks. third system—the explicit integer-list representation
Infants most probably deploy two distinct systems in that supports counting. This work motivates several
tasks that require representation of number—object- questions at the level of systems neuroscience. Are there
file and analog-magnitude representations. Further, I distinct neural circuits for these computationally dis-
sketched some respects in which each is very different tinct representational systems? Do they continue to ar-
from, and weaker than, the explicit integer-list repre- ticulate mathematical reasoning into adulthood?
sentation of number that is mastered in the child's Existing evidence suggests an affirmative answer on
fourth year of life. The acquisition of the integer-list both counts (for reviews, see Dehaene, 1997; Butter-
representation of number thus constitutes an example worth, 1997).
of a developmental discontinuity. I concluded with a The object-file system was originally studied in the
speculative sketch of bootstrapping mechanisms, illus- context of object-based attention, and it clearly contin-
trating with an example of a bootstrapping process that ues to articulate midlevel vision in adulthood. Data
might achieve the new representational resource. These from studies of patients with neuropsychological dam-
latter arguments fall on the side of constructivist ac- age and from imaging studies converge on the conclu-
counts of the creation of new representational re- sion that parietal cortex, especially inferior parietal
sources during development. cortex, is part of the neural circuit underlying midlevel
At the level of structural analogy, then, I have argued object-based attention. For example, Culham, Brandt,
that constructive processes, rather than selective ones, Cavanagh, Kanwisher, Dale, and Tootell (1998) found
underlie the acquisition of the integer-list representa- that activity levels in this area were correlated with the
tion of number, and that this process involves the crea- number of objects being simultaneously being tracked
tion of a representational resource with more power in the Pylyshyn multiple object tracking task (other ar-
than those from which it is built. As far as I am aware, eas, including regions of frontal cortex, also showed this
relation). Parietal damage is also implicated in object-
there are no known neural developmental processes
based neglect (e.g., Behrman and Tipper, 1994) and
analogous to the bootstrapping mechanisms sketched
the even more striking Balint syndrome (for a review,
above. For example, the activity-driven processes that
see Rafel, 1997). Patients with Balint syndrome exhibit
are involved in dendrite growth that Quartz and Se-
simultanagnosia, the incapacity to perceive two objects
jnowski (1997) offer as an example of constructive pro-
at once, in the face of normal sensory processing. They
cesses provide only a very distant analogy to those that
cannot do simple tasks such as telling which of two lines
are involved in the case of cognitive development de-
is longer, and do not report seeing two simultaneously
tailed here. How might we go further in closing the gap
presented disks, although if the disks are joined by a
in this case?
bar, making a dumbbell, they report seeing the whole
dumbbell (Luria, 1959). It seems likely that these syn-
Closing the gap: Integrating cognitive development dromes reflect damage to the midlevel object-file/ob-
with systems neurosdence ject indexing attentional systems (for development of
this point, see Scholl, in press).
Detailed models of representational format play two The object-file system, with its capacity to simulta-
roles in the argument I am developing here. Hereto- neously index and track up to 4 objects, is not a dedi-
fore, my main point has been that it is only by exam- cated number representational system. Number is
ining the initial representational systems in detail that merely implicitly represented here; the objects indexed
we can say in what ways they are continuous and in what constitute a set number that is represented by one
ways discontinuous with later representational systems, object-file for each object in the set. The analog-
and only then can we motivate possible learning mech- magnitude representational systems, in contrast, rep-
anisms that take the initial systems as input and output resent specifically number. There are analog-magnitude
representational systems that transcend them. Here, I representations for other physical quantities, of course,
wish to emphasize a second point: It is the same level such as distance, size, and time; but the computations
of detail (at least) that is needed to begin to close the that establish representations of number in a model
gap between characterizations of the developing mind such as Meek and Church's create an analog symbol for
and characterizations of the developing brain. number alone. Dehaene (1997) has summarized the
428 COGNITION
portant questions become when and how does the CLEARFIELD, M. W., and K. S. Mix, 1999. Number vs. contour
integer-list representation become integrated with length in infants' discrimination of small visual sets. Psycho-
logical Science, 10:408-411.
analog-magnitude representations? Although purely CHIANG, W.-C., and K. WYNN, 2000. Infants' representation
behavioral data will bear on these questions, imaging and tracking of objects: Implications from collections. Cog-
studies with preschool-aged children, should the tech- nition. 77:169-195.
niques become available, also will provide invaluable CHURCH, R. M., and H. A. BROADBENT, 1990. Alternative rep-
information. resentations of time, number and rate. CagTOta'ow33:63-109.
CORNSWEET, T. M. 1970. Visual Perception. New York: Academic
Press.
Final conclusions CULHAM, J., S. BRANDT, P. CAVANAGH, N. KANWISHER, A. L.
DALE, AND R. TOOTELL, 1998. Cortical fMRI activation pro-
We are far from being able to say anything about the duced by attentive tracking of moving targets. Journal of Neu-
growth of the patterns of neural connectivity that un- rophysiology, 80:2657.
derlie the development of number representations DEHAENE, S., 1997. The Number Sense. Oxford: Oxford Univer-
sity Press.
described in this chapter. I have argued that character-
DEHAENE, S., and J. P. CHANGEUX, 1989. Neuronal models of
izing the details of representational format and the cognitive function. Cognition 33:63-109.
computations defined over those representations, as DEHAENE, S., E. SPELKE, P. PINEL, R. STANESCU, and S. Tsiv-
well as characterizing development at these levels, is a KIN, 1999. Sources of mathematical thinking: Behavioral
necessary prerequisite for such understanding. As al- and brain-imaging evidence. Science, 284: 970-974.
ways, an understanding of the mind must guide the DIAMOND, A., 1991. Neuropsychological insights into the
meaning of object concept development. In The Epigenesis
search for its neural underpinnings.
of Mind: Studies in Biology and Cognition, S. Carey and R. Gel-
man, eds. Hillsdale, N.J.: Erlbaum, pp. 67-110.
DIAMOND, A., and P. S. GOLDMAN-RAKIC, 1989. Comparison
REFERENCES of human infants and rhesus monkeys on Piaget's AB task:
Evidence for dependence on dorsolateral prefrontal cortex.
ANTELL, S., and D. KEATING, 1983. Perception of numerical Exp. Brain Res. 74:24-40.
invariance in neonates. Child Dev. 54:695-701. ELMAN, J. L., E. A. BATES, M. H. JOHNSON, A. KARMILOFF-
BAILLARGEON, R., 1993. The object concept revisited: New di- SMITH, D. PARISI, and K. PLUNKETT, 1996. Rethinking In-
rections in the investigation of infants' physical knowledge. nateness: A Connectionist Perspective on Development.
In Visual Perception and Cognition in Infancy, C. E. Granud,
Cambridge, Mass.: MIT Press.
ed. Hillsdale, N.J.: Erlbaum, pp. 265-315.
FEIGENSON, L., S. CAREY, and E. SPELKE, in press. Infants' Dis-
BAILLARGEON, R., K. MILLER, and J. CONSTANTINO, 1993. Ten-
crimination of Number vs. Continuous Extent. Cognitive
month-old infants' intuitions about addition. Unpublished
Psychology.
manuscript, University of Illinois at Urbana-Champaign.
FEIGENSON, L., S. CAREY, and M. HAUSER, 2000. Spontaneous
EARTH, H., KANWISHER, N., and SPELKE, E., under review.
ordinal judgments by preverbal infants. Poster presented at
Constructing representations of large numbers of entities.
the 12th Biennial International Conference on Infant Stud-
BEHRMAN, M., and S. P. TIPPER, 1994. Object-based visual at-
tention: Evidence from unilateral neglect. In C. Umilta & ies, Brighton, UK.
M. Moscovitch, eds. Attention and performance XIV: Conscious FODOR, J., 1975. The Language of Thought. New York: Crowell.
and nonconscious processing and cognitive functioning. Cam- FUSON, K. C., 1988. Children's Counting and Concepts of Number.
bridge, Mass.: MIT Press. New York: Springer-Verlag.
BIJELJAC-BABIC, R.,J. BERTONCINI, and J. MEHLER, 1991. How GALLISTEL, C. R., 1990. The Organization of Learning. Cam-
do four-day-old infants categorize multi-syllabic utterances? bridge, Mass.: MIT Press.
Dev. Psychol. 29:711-721. GALLISTEL, C. R., and R. GELMAN, 1992. Preverbal and verbal
BRANNON, E. M., and H. S. TERRACE, 1998. Ordering of the counting and computation. Cognition 44:43-74.
numerosities 1-9 by monkeys. Science 282:746-749. GAZZANIGA, M. S., 1995. The Cognitive Neurosciences. Cam-
BUTTER WORTH, B., 1999. What Counts: How Every Brain is Hard- bridge, Mass.: MIT Press.
wired for Math. New York: Simon & Shuster, The Free Press. GELMAN, R., and C. R. GALLISTEL, 1978. The Child's Understand-
CAREY, S., in preparation. The Origin of Concepts: Evolution vs. ing of Number. Cambridge, Mass.: Harvard University Press.
Culture. Cambridge, Mass.: MIT Press. GOLD, E. M., 1967. Language identification in the limit. Infor-
CAREY, S., and E. SPELKE, 1994. Domain specific knowledge mat. Control 10:447-474.
and conceptual change. In L. Hirschfeld and S. Gelman, GOLDMAN, P. S., 1971. Functional development of the pre-
eds. Mapping the Mind: Domain Specificity in Cognition and Cul- frontal cortex in early life and the problem of neuronal
ture. Cambridge: Cambridge University Press, 169—200. plasticity. Exp. Neural. 32:366-387.
CAREY, S., and E. SPELKE, in press. On the very possibility of HAUSER, M. D., S. CAREY, and L. B. HAUSER, 2000. Sponta-
conceptual change: The integer-list representation of num- neous number representation in semi-free-ranging rhesus
ber. In Conceptual Development—A Reappraisal, L. Bonati, J. monkeys. Proc. K Soc. Lond. B. Biol. Set. 267(1445):829-833.
Mehler, and S. Carey, eds. Cambridge, Mass.: MIT Press. HELMHOLZ, H., 1910. Treatise on physiological optics. Translated
CAREY, S., and F. Xu, in press. Beyond object-files and object and ed. From the 3rd German ed. Southall, J. P. Rochester,
tracking: Infant representations of objects. Cognition. NY. Optical Society of America.
430 COGNITION
WERKER, J. R, and C. E. LALONDE, 1998. Cross-language WYNN, K., 1992b. Children's acquisition of the number words
speech perception: In capabilities and developmental and the counting system. Cogn. Psychol. 24:220-251.
change. Dev. Psychol. 24(5):672-683. WYNN, K., 1996. Infants' individuation and enumeration of
WEXLER, K., and P. CULICOVER, 1980. Formal Principles of Lan- physical actions. Psychol. Sci. 7:164-169.
guage Acquisition. Cambridge, Mass.: MIT Press. Xu, R, and S. CAREY, 1996. Infants' metaphysics: The case of
WYNN, K., 1990. Children's understanding of counting. Cog- numerical identity. Cogn. Psychol. 30(2) :111-153.
nition 36:155-193. Xu, R, and S. SPELKE, 2000. Large number discrimination in
WYNN, K., 1992a. Addition and subtraction by human infants. 6-month-old infants. Cognition 74:B1-B11.
Nature 358:749-750.
ABSTRACT After a brief overview of the anatomy of dorso- PFC. It is my hope that this review will offer some insight into
lateral prefrontal cortex (DL-PFC) and of its anatomical con- cognitive development, into the role of prefrontal cortex in
nections with other brain regions, findings are summarized cognitive development, and into the role of dopamine in pre-
that DL-PFC subserves working memory and inhibitory con- frontal cortex function.
trol abilities even during infancy. Evidence suggests that one
change in the prefrontal neural circuit helping to make pos- Dorsolateral prefrontal cortex (DL-PFC) undergoes
sible some of the cognitive advances seen in infants between an extremely protracted period of maturation and is
6 and 12 months of age might be changes in the dopaminergic
not fully mature until adulthood (Huttenlocher, 1979,
innervation of prefrontal cortex. The period of 3-6 years is
then examined as a period when there is particularly marked 1984, 1990; Huttenlocher et al., 1982; Orzhekhovskaya,
improvement in the working memory and inhibitory control 1981; Rosenberg and Lewis, 1994; Sowell et al., 1999;
abilities thought to depend upon DL-PFC. Perhaps that im- Thatcher, Walker, and Giudice, 1987; Yakovlev and Le-
provement is made possible, in part, by maturational changes cours, 1967). Growing evidence indicates, however, that
in DL-PFC, perhaps even in its dopamine projection, although some of the cognitive advances seen as early as the first
that remains to be demonstrated. (To propose that changes
in the dopamine innervation of prefrontal cortex play a role year of life (6-12 months) are made possible, in part,
in making possible some cognitive advances is not to negate by early changes in DL-PFC (e.g., Bell and Fox, 1992,
the role of experience nor the role of other maturational 1997; Diamond, 1991a,b; Fox and Bell, 1990). One mat-
changes in the prefrontal neural system.) urational change in DL-PFC that might help make pos-
As an initial way to begin looking at the role of dopamine sible these early cognitive advances is increasing levels
in prefrontal cortex function in human beings during infancy
of the neurotransmitter dopamine in DL-PFC.
and early childhood, we studied a group of children who,
there was reason to believe, have reduced levels of dopamine Prefrontal cortex is richer in dopamine than any
in prefrontal cortex but otherwise normal brains. These are other region of the cerebral cortex (Bjorklund, Divac,
children treated early and continuously for phenylketonuria and Lindvall, 1978; Brown, Crane, and Goldman, 1979;
(PKU), whose phenylalanine levels are 3—5 times normal and Caspar et al., 1989; Levitt, Rakic, and Goldman-Rakic,
whose tyrosine (Tyr) levels are below normal. The rationale
1984; Lewis et al., 1988, 1998; Williams and Goldman-
for studying these children and the results obtained in those
studies are summarized, as are the results from our work with Rakic, 1993, 1995). Not surprisingly, given its high con-
an animal model of early, and continuously, treated PKU. Evi- centration in prefrontal cortex, dopamine plays an
dence on dissociations among tasks that require DL-PFC but important role in DL-PFC function in adult human and
are differentially sensitive to the dopamine content of DL-PFC nonhuman primates (Akil et al., 1999; Brozoski et al.,
is discussed. The evidence shows that not all cognitive tasks 1979; Luciana et al., 1992; Sawaguchi, Matsumura, and
dependent on DL-PFC are dependent on dopamine in DL-
Kubota, 1988; Sawaguchi and Goldman-Rakic, 1991;
Watanabe, Kodama, and Hikosaka, 1997).
ADELE DIAMOND Director, Center for Developmental Cog-
nitive Neuroscience, Eunice Kennedy Shriver Center, Profes- We know that during the period that infant rhesus
sor, Department of Psychiatry, University of Massachusetts macaques are improving on tasks dependent on DL-
School of Medicine, Waltham, Massachusetts. PFC (the A-not-B, delayed response, and object retrieval
434 COGNITION
also Petrides and Pandya, 1999, concerning all of these
interconnections), and the neocerebellum (Diamond,
2000; Leiner, Leiner, and Dow, 1989; Middleton and
Strick, 1994, 1997; Sasaki et al., 1979; Schmahmann and
Pandya, 1995; Yamamoto et al., 1992), mid-DL-PFC can
modulate the activity of those regions, as well as receive
information from, and be modulated by, these regions.
In addition, mid-DL-PFC sends a strong projection to
the caudate nucleus (Arikuni and Kubota, 1986; Gold-
man and Nauta, 1977; Kemp and Powell, 1970; Selemon
and Goldman-Rakic, 1985). The projections from DL-
PFC, posterior parietal cortex, and the superior tem-
poral cortex are intricately interdigitated throughout
the brain, including in the caudate nucleus, providing
multiple opportunities for these neural regions to com-
municate with, and influence, one another (Goldman-
Rakic and Schwartz, 1982; Johnson et al., 1989; Schwartz
and Goldman-Rakic, 1984; Selemon and Goldman-
Rakic, 1985, 1988).
Evidence that DL-PFC subserves cognitive
abilities even during infancy
The A-not-B task has been used in scores of laboratories
throughout the world to study cognitive development
in infants since it was first introduced by Piaget in 1936
(see Piaget, 1954). Under the name "delayed response,"
an almost-identical task has been the classic paradigm
FIGURE 29.1 Diagram of the human brain, indicating the lo-
for studying the functions of DL-PFC in macaques since
cation of dorsolateral prefrontal cortex. (Reprinted with per- it was first introduced for that purpose by Jacobsen
mission from Kandel, E. R., J. H. Schwartz, and T. M. Jessell, (1935, 1936). In the A-not-B/delayed response task, the
eds., 1991. Principles of Neural Science, 3rd ed. New York: participant watches as a desired object is hidden in one
McGraw Hill.) of two hiding places that differ only in their left-right
location, and then a few seconds later is allowed to
lular portion of the mediodorsal nucleus has increased reach to find that object. The participant must hold in
phylogenetically in proportion to the increase in size of mind over those few seconds where the object was hid-
DL-PFC and disproportionately compared even to den. Over trials, the participant must update his or her
other regions of the mediodorsal nucleus (Clark, 1930; mental record to reflect where the reward was hidden
Khokhryakova, 1979; Pines, 1927). last. When the participant reaches correctly, he or she
No area of the brain acts in isolation. A neural region is rewarded by being allowed to retrieve the desired ob-
ject. In this manner, the behavior of reaching to that
functions as part of system of functionally and anatom-
hiding location is reinforced, and hence the tendency
ically interrelated structures. Through its reciprocal
to emit that response is strengthened. When the reward
connections with the superior temporal cortex (Petrides
is then hidden at the other location, the participant
and Pandya, 1988; Seltzer and Pandya, 1989), posterior must inhibit the natural tendency to repeat the re-
parietal cortex (area 7a; Cavada and Goldman-Rakic, warded response and instead respond according to the
1989; Goldman-Rakic and Schwartz, 1982; Johnson et representation held in mind of where the reward was
al., 1989; Petrides and Pandya, 1984; Schwartz and just hidden. Thus, the A-not-B task requires holding in-
Goldman-Rakic, 1984; Selemon and Goldman-Rakic, formation in mind (where the reward was last hidden)
1988), anterior and posterior cingulate (Vogt and Pandya, and inhibition of a prepotent response tendency. By
1987; Vogt, Rosene, and Pandya, 1979), premotor cortex roughly 7^-8 months of age, infants reach correctly to
(Barbas and Mesulam, 1985, 1987; Kiinzle, 1978), SMA the first hiding location with delays as long as 3 s. When
(McGuire, Bates, and Goldman-Rakic, 1991; Wiesen- the reward is then hidden at the other hiding place,
danger, 1981), retrosplenial cortex (Morris, Pandya, and however, infants err by going back to the first location
Petrides, 1999; Morris, Petrides, and Pandya, 1999; see (called the "A-not-B error"). As infants get older, they
436 COGNITION
tex, also produce deficits on the task (e.g., Saint-Cyr et istic pattern to the errors made by infants and by pre-
al, 1988; Schneider and Roeltgen, 1993; Taylor et al., frontally-lesioned macaques on the A-not-B task: Their
1990a,b). (MPTP also affects the level of dopamine in errors tend to be confined to the reversal trials and to
the striatum, but lesions of the striatum do not impair the trials immediately following a reversal error when
performance on the object retrieval task [Crofts et al., the reward continues to be hidden at the new location
1999].) Cumulative doses of 15-75 mg of MPTP do not (Diamond, 1985, 1990a, 1991a,b). If the only source of
produce Parkinsonian-type motor deficits in rhesus ma- error on the task were failure to keep the critical infor-
caques, although larger doses do. At the lower doses of mation in mind, one would expect errors to be random;
MPTP (15-75 mg), monkeys are impaired on the object but they are not (Diamond, Cruttenden, and Neider-
retrieval and A-not-B/delayed response tasks (Schnei- man, 1994). The nonrandom pattern of errors, and the
der and Kovelowski, 1990; Taylor et al., 1990a,b), al- fact that participants occasionally look at the correct
though they perform normally on other tasks such as location (as if they remember that the reward is there)
visual discrimination. while at the same time reaching back to the previously
Importantly, human infants, infant rhesus macaques, correct location (Diamond, 1990a, 1991a; see also Hof-
and infant and adult rhesus macaques with lesions of stadter and Reznick, 1996), suggests that success on the
DL-PFC fail the A-not-B/delayed response task under task also requires resisting, or inhibiting, the tendency
the same conditions and in the same ways. The same is to repeat the previous response. (I have suggested that
true for the object retrieval task. Thus, for example, on there is a predisposition to repeat the previous response
A-not-B, macaques with lesions of DL-PFC and human because it had been rewarded. Smith and colleagues
infants of 7 1/2-9 months succeed when there is no delay [1999] suggest that there is a predisposition to repeat
(macaques: Battig, Rosvold, and Mishkin, 1960; Gold- the previous response simply because the response was
man, Rosvold, and Mishkin, 1970; Harlow et al., 1952; made before [not because of reinforcement], just as it
infants: Gratch et al., 1974; Harris, 1973), succeed when is easier for neurons in visual cortex to process a visual
allowed to circumvent the memory requirements by stimulus if they have previously processed that visual
continuing to stare at or strain toward the correct well stimulus. Either account of the source of the predispo-
during the delay (macaques: Battig, Rosvold, and Mish- sition works equally well for my theoretical position.
kin, 1960; Miles and Blomquist, 1960; Pinsker and The important point is that there is a tendency to repeat
French, 1967; infants: Cornell, 1979; Fox, Kagan, and the previous response; the source of that predisposition
Weiskopf, 1979), succeed if a landmark reliably indi- is unimportant for my argument.)
cates where the reward is located (macaques: Pohl, 1973; A few errors can be elicited simply by taxing how long
infants: Butterworth, Jarrett, and Hicks, 1982), fail even information must be held in mind even when no inhi-
at brief delays of only 2-5 s (macaques: Diamond and bition is required, such as by using a long delay at the
Goldman-Rakic, 1989; Goldman and Rosvold, 1970; in- first hiding location (SophianandWellman, 1983). Sim-
fants: Diamond, 1985; Diamond and Doar, 1989; Fox, ilarly, a few errors can be elicited simply by taxing in-
Kagan, and Weiskopf, 1979; Gratch and Landers, 1971), hibitory control even when the participant does not
and fail if the hiding places differ either in left-right have to remember where the reward was hidden; for
or up-down location (macaques: Fuster, 1980; Goldman, example, a few infants err on the reversal trial even
Rosvold, and Mishkin, 1970; infants: Butterworth, 1976; when the covers are transparent (Butterworth, 1977;
Gratch and Landers, 1971). See figure 29.2 for details. Willatts, 1985). However, the overwhelming majority of
Note that most of these manipulations indicate that errors occur when participants must both hold infor-
the presence of the delay is critical, since even very mation in mind and also exercise inhibitory control
young infants and prefrontally-lesioned macaques per- (i.e., on reversal trials when the covers are opaque and
form well when there is no delay or when the require- a delay is imposed).1
ments of the delay can be circumvented. This suggests Similar close parallels in the parameters determining
that the ability to hold in mind the information on success or failure, and in the characteristics of perfor-
where the reward was last hidden (this might be termed mance, hold for the object retrieval task (infants: Dia-
"sustained attention" or the information-maintenance mond, 1981; 1990b, 1991a; macaques with lesions of
component of "working memory" [Baddeley, 1992]) is DL-PFC: Diamond, 1990b; 199la; Diamond and Gold-
critical to success on this task. The information that man-Rakic, 1985; MPTP-treated macaques: Saint-Cyr et 3al.,
must be held in mind is relational (Was the reward hid- 1988; Schneider and Roeltgen, 1993; Taylor et al.,
den on the right or the left most recently? "Left" is only 1990a,b). Human infants of 7 1/2-9 months, rhesus ma-
left in relation to right and, similarly, "recent" implies caques with lesions of DL-PFC, and macaques treated
a before-and-after relation.) There is also a character- with MPTP all succeed on the object retrieval task when
they are looking through the open side of the box. They the contralateral hand to reach in the opening (see fig-
fail when they are looking through a closed side, and ure 29.3) and show that "awkward reach" on both the
they fail by trying to reach straight through the trans- left and right sides of the box.
parent barrier instead of detouring around it. Human This pattern of performance highlights the impor-
infants of 71/2-9months, rhesus macaques with lesions tance, for success on the object retrieval task, of being
of DL-PFC, and macaques treated with MPTP perform able to inhibit the strong tendency to reach straight in
better when the box is opaque than when the box is the side of the box through which one is looking. Be-
transparent. They lean over to look in the box opening haviors such as the "awkward reach" also highlight the
when the left or right side of the box is open and recruit importance of holding the location of the box opening
438 COGNITION
FIGURE 29.2 (Continued). After 2 trials, there is a switch and error is called the A-not-B error because they are correct on
the reward is hidden at the second hiding place (B). Although the A trials, but not on the B trials; they reach to A, not B.
they all watch the hiding at B, and although the delay at B is (Reprinted with permission from Diamond, 1990a,b.)
no longer than at A, they all err by reaching back to A. This
FIGURE 29.3 Illustration of a 2-month-old infant rhesus ma- the contralateral hand to reach in and retrieve the reward.
caque, 9-month-old human infant, and an adult rhesus ma- This is seen on both sides of the box and does not reflect a
caque in whom dorsolateral prefrontal cortex had been re- hand preference. Because of its appearance, the recruitment
moved bilaterally performing the object retrieval task. They of the contralateral arm is dubbed the "awkward reach. (Re-
lean and look in the side opening of the transparent box, and printed with permission from Diamond, 1990a,b.)"
then while continuing to look through the opening, recruit
in mind when looking at the reward and holding the the prefrontal dopamine system fail the A-not-B/de-
location of the reward in mind when looking at the box layed response and object retrieval tasks under the same
opening, and of integrating the two pieces of informa- conditions and in the same ways (table 29.1). This does
tion. Focusing exclusively on the reward or the box will not prove that maturational changes in DL-PFC during
not work for this task; both must be taken into account. infancy contribute to the emergence of success on these
Reaching through the opening when looking through tasks during infancy, but this body of work makes that
a closed side requires integrating in one's mind looking hypothesis plausible.
at the reward along one route with reaching for the
reward along a completely different route. Infants of Evidence of improvement in the cognitive abilities
8l/2-9 months and prefrontally-lesioned macaques are that depend on DL-PFC during early childhood
only able to succeed when the left or right side of the
transparent box is open by simplifying the task. They DL-PFC continues to mature until early adulthood.
lean over to look in the opening, hence lining up the Marked improvements on tasks that require working
opening and the reward so that they can see both at memory plus inhibition (tasks thought to require the
once and so that their line of sight is the same as the functions of DL-PFC) are seen in children between 3
line along which they will reach. and 6 years of age. At 3 years of age, one can see errors
In sum, human infants of 71/2-9 months, infant ma- reminiscent of the A-not-B error seen in infants and in
caques of 11/2-21/2months, adult macaques with bilateral prefrontally-lesioned macaques, but with a slightly more
removals of DL-PFC, infant macaques of 5 months in difficult task. On this task, children who are 3 years old
whom DL-PFC was removed at 4 months, and adult ma- can sort cards correctly by the first criterion they are
caques who have received MPTP injections to disrupt given (either color or shape; Kirkham, Cruess, and
440 COGNITION
TABLE 29.1
Performance of human infants, infant rhesus monkeys, and adult rhesus monkeys with selective ablations on the same three tasks
Diamond, submitted; Zelazo, Frye, and Rapus, 1996; shape) and only two values for each criterion (e.g., red/
Zelazo and Reznick, 1991; Zelazo, Reznick, and Pirion, blue, truck/star) children are able to succeed at the
1995), just as infants of 71/2-9 months and prefrontally- card sorting task by 4-5 years of age. If the task is made
lesioned macaques are correct at the first hiding place, more complicated, by, for example, adding a third sort-
and just as adults with prefrontal cortex damage are ing dimension, then children cannot succeed until they
correct at sorting cards according to the first criterion are 5-6 years old. The problem for the children appears
(Wisconsin Card Sort test: Drewe, 1974; Milner, 1963, to be in relating two or more dimensions to a single
1964). Three-year-old children err when correct per- stimulus (thinking of a stimulus as either red or blue
formance demands switching to a new criterion, i.e., and also thinking about the same stimulus as either a
when cards previously sorted by color (or shape) must truck or a star) and in inhibiting the tendency to repeat
now be sorted according to the other criterion (shape their previously correct way of categorizing the
or color), just as infants of 7 1/2-9 months and prefron- stimulus.
tally lesioned macaques err when required to switch and Similarly, children 3 years old have great difficulty
search for the reward at the other location, and just as with "appearance-reality" tasks (Flavell, 1986, 1993)
adults with prefrontal cortex damage err when required where, for example, they are presented with a sponge
to switch to a new sorting criterion. that looks like a rock. Three-year-olds typically report,
Although 3-year-old children fail to sort by the new for example, that it looks like a rock and really is a rock,
sorting criterion (sticking steadfastly to the previously whereas a child of 4-5 years correctly answers that it
correct criterion), they can correctly state the new sort- looks like a rock but really is a sponge. The problem for
ing criterion (Kirkham et al., submitted; Zelazo, Frye, the younger children is in relating two conflicting iden-
and Rapus, 1996). Similarly, infants of 71/2-9 months tities to the same object (Rice et al., 1997) and in in-
can sometimes tell you with their eyes that they know hibiting the response that matches their perception
the reward is in the new hiding place even as they persist (thus manipulations that reduce the perceptual sa-
in reaching back to the previously correct location (Di- lience, by removing the object during questioning, find
amond, 1990a, 1991a,b; Hofstadter and Reznick, 1996), significantly better performance by children of 3-4
and patients with prefrontal cortex damage can some- years [e.g., Heberle, Clune, and Kelly, 1999]). "Theory
times tell you correctly the new sorting criterion even of mind" and "false belief" tasks are other tasks that
as they persist in sorting by the previously correct cri- require holding two things in mind about the same sit-
terion (Luria and Homskaya, 1964; Milner, 1963,1964). uation (the true state of affairs and the false belief of
When there are only two sorting criteria (color and another person) and inhibiting the impulse to give the
442 COGNITION
resent; instead they must say the opposite. Children of itself sufficient to account for the poor performance of
31/2-41/2 years find the task terribly difficult; by 6-7 years the younger children on the day-night task.
of age, the task is trivially easy. Children younger than Moreover, children's difficulty with the task depends
6 years of age err often, whereas children of 6-7 years critically on the correct responses being semantically
are correct on roughly 90% of the trials (see figure related to the responses that must be inhibited. When
29.4). Children of 31/2and 4 years show long response we used the same white/sun and black/moon cards, but
latencies on the task (roughly 2 s); older children take instructed the children to say "dog" to one and "pig" to
roughly half as long (1 s). The age-related increase in the other, even the youngest children again performed
the percentage of correct responses is relatively contin- well (Diamond, Kirkham, and Amso, submitted). The
uous from 31/2to 7 years of age, but the decrease in task of holding two rules in mind and inhibiting one's
speed of responding occurs primarily between 3H and natural inclination is sufficiently hard for the younger
4 1/2 years. Passler, Isaac, and Hynd (1985) tested chil- children that they need a long time to formulate their
dren on a similar, though slightly easier, variant of this answers in order to respond correctly. Although we gave
task, which required children to recognize the correct children unlimited time, they tended to speed up their
answer, whereas our task requires that they recall the responses over the 16 test trials, and the accuracy of the
correct answer. They found that children of 6 years were youngest children correspondingly fell. When we made
performing at ceiling on their task, which is consistent the children wait to respond, by singing a brief ditty to
with the excellent performance that we found at 6-7 them on each trial after the stimulus was presented,
years of age. the younger children were able to perform well, even
To test whether the requirement to remember two though the period before their response was filled with
rules alone is sufficient to cause the younger children potentially interfering verbal stimulation (Diamond et
difficulty, we tested a version of our day-night test al., submitted). It is not simply that slowing down the
where each card contained one of two abstract designs testing helped because when the children were made
(Gerstadt, Hong, and Diamond, 1994). Children were to wait before the start of each trial, they performed
instructed to say "day" to one design and "night" to the poorly. The day-night task is sufficiently difficult for
other. Here the children were still required to hold two young children that it takes them several seconds to
rules in mind, but they did not also have to inhibit the compute the answer. Often they do not take the needed
tendency to say what the stimuli really represented time; when forced to take extra time they can perform
well.
because the stimuli were abstract designs. Even the
Luria's tapping test (Luria, 1966) also requires (1)
youngest children performed superbly here. Thus, the
remembering two rules and (2) inhibiting the response
requirement to learn and remember two rules is not in
you were inclined to make, making the opposite re-
sponse instead. Here, one needs to remember the rules,
"Tap once when the experimenter taps twice, and tap
twice when the experimenter taps once," and one needs
to inhibit the tendency to mimic what the experimenter
does. Children improve on this task over the same age
period as they do on the day-night task (see figure
29.4). Over the period of 3 1/2-7 years, children improve
in both speed and accuracy on the tapping task, with
most of the improvement occurring by the age of 6
(Becker, Isaac, and Hynd, 1987; Diamond and Taylor,
1996; Passler, Isaac, and Hynd, 1985).
Adults with large frontal lobe lesions fail this same
tapping task (Luria, 1966). They have similar problems
when instructed to raise a finger in response to the ex-
perimenter's making a fist and to make a fist in re-
sponse to the experimenter's raising a finger (Luria,
1966). The most common error by young children is
always tapping once, or always tapping twice, regardless
FIGURE 29.4 Performance of children, 3 1/2 through 7 years
of age, on the day-night, tapping, and three pegs task. Note of what the experimenter does. It may be that the young
the close parallels in performance on all three tasks through- children are able to keep in mind only one of the two
out this age range. rules. Or, it may be that they lack the ability to switch
444 COGNITION
anine hydroxylase. Phenylalanine hydroxylase is essen- age is thought to be the toxic effects of grossly elevated
tial for hydroxylating (i.e., converting) the amino acid levels of Phe in the brain.
phenylalanine (Phe) into the amino acid tyrosine (Tyr)
(DiLella et al., 1986; Lidsky et al., 1985; Woo et al., 1983;
see figure 29.5). In the roughly 1 in every 10,000 people THE TREATMENT FOR PKU: A DIET Low IN PHENYLAL-
born with PKU, phenylalanine hydroxylase activity is ANINE The treatment for PKU consists of a diet low
either absent or markedly reduced. Hence, PKU is a in Phe. Since Phe is a constituent of protein, the low-
member of the class of disorders called "inborn [i.e., Phe diet severely restricts the intake of milk and milk
genetic] errors of metabolism." In the case of PKU, the products (such as ice cream, butter, and cheese), and
error is in the metabolism of Phe. all meat and fish. When PKU is treated early and con-
Since little, if any, Phe is metabolized, Phe levels in tinuously by a diet low in Phe, gross brain damage and
the bloodstream reach dangerously high levels. Indeed, severe mental retardation are averted (Bickel, Hudson,
when PKU is untreated, levels of Phe in the bloodstream and Woolf, 1971; Holtzman et al., 1986). Note that here
rise to well over 10 times normal (>20 mg/dL [>1200 is an example of how a behavioral change (changing
mmol/L]). Since little or no Tyr is produced from Phe, what you eat) can profoundly affect your biochemistry
the level of Tyr in the bloodstream is low (Nord, Mc- and your brain.
Cabe, and McCabe, 1988). (Tyr levels would be still
lower were it not for the availability of Tyr directly Limitations of the diet: Why problems might still exist when
through the foods we eat.) This imbalance in blood lev- PKU is treated The low-Phe diet rarely results in fully
els of Phe and Tyr, if not corrected early, causes wide- normal levels of Phe or Tyr. This is because the need to
spread brain damage and severe mental retardation minimize Phe intake must be balanced with the need
(Cowie, 1971; Hsia, 1967; Koch et al., 1982; Krause et for protein. Eliminating all Phe from the diet would
al., 1985; Tourian and Sidbury, 1978). Indeed, PKU is require eliminating all protein. Outside of protein, Phe
the most common biochemical cause of mental retar- is not present in any naturally occurring food. The hu-
dation. The primary cause of the widespread brain dam- man body needs to ingest protein; moreover, the body
FIGURE 29.5 Diagram illustrating the reasoning leading to have a selective decrease of dopamine in prefrontal cortex and
the hypothesis that children treated early and continuously a selective deficit in the cognitive abilities dependent on pre-
for PKU, whose blood Phe levels are 6-10 mg/dL, would frontal cortex.
446 COGNITION
in the level of Tyr than other brain regions. The brain PFC (Simon, Scatton, and LeMoal, 1980). Similarly, in-
needs Tyr to make dopamine (see figure 29.6). Indeed, jections of MPTP that disrupt the dopamine projection
the hydroxylation of Tyr is the rate-limiting step in the to prefrontal cortex, but are of sufficiently low dose that
synthesis of dopamine. Most dopamine systems in the motor deficits are avoided, impair performance on the
brain are unaffected by small decreases in the amount A-not-B/delayed response and object retrieval tasks
of available Tyr. Not so prefrontal cortex. The dopa- (Schneider and Kovelowski, 1990; Taylor et al.,
mine neurons that project to prefrontal cortex are un- 1990a,b).
usual in that they have a higher firing rate and higher
rate of dopamine turnover than other dopamine neu- SUMMARY or THE REASONING LEADING TO THE PRE-
rons (Bannon, Bunney, and Roth, 1981; Roth, 1984; FRONTAL DOPAMINE HYPOTHESIS IN TREATED PKU For
Thierry et al., 1977). These unusual properties of the these reasons it seemed plausible that the moderate im-
prefrontally projecting dopamine neurons in the ven- balance in the Phe:Tyr ratio in the bloodstreams of chil-
tral tegmental area (VTA) make prefrontal cortex dren treated early and continuously for PKU might well
acutely sensitive to even a modest change in the supply result in deficits in the cognitive abilities dependent on
of Tyr (Tarn et al., 1990; Wurtman et al., 1974). Reduc- prefrontal cortex (because of the unusual vulnerability
tions in the availability of Tyr too small to have much of the dopamine projection to prefrontal cortex to a
effect on other dopamine systems in other neural moderate reduction in the amount of available tyro-
regions (such as the striatum) have been shown to pro- sine) without significantly affecting other brain regions
foundly reduce dopamine levels in prefrontal cortex or other cognitive abilities. Hence, we hypothesized that
(Bradberry et al., 1989). here was a mechanism by which the modest elevation
in the Phe:Tyr ratio in the bloodstream of some chil-
Reducing the level of dopamine in prefrontal cortex produces dren treated for PKU, which results in moderate reduc-
deficits in the cognitive abilities dependent on prefrontal cortex tions in the level of Tyr reaching the brain, might
As mentioned earlier, selectively depleting DL-PFC of selectively affect prefrontal cortex (by modestly decreas-
dopamine can produce cognitive deficits as severe as ing the level of Tyr reaching the brain).
those found when DL-PFC is removed altogether (Bro-
zoski et al., 1979). Local injection of dopamine A 4-year longitudinal study of children treated
antagonists into DL-PFC impairs performance in a early and continuously for PKU
precise, dose-dependent manner (Sawaguchi and
Goldman-Rakic, 1991). Destruction of the dopamine To investigate our prediction that children treated early
neurons in the VTA that project to prefrontal cortex and continuously for PKU have selective deficits in the
also impairs performance on tasks dependent on DL- cognitive functions dependent on prefrontal cortex, we
FIGURE 29.6 Diagram illustrating the mechanism by which tyrosine via two routes, the hydroxylation of Phe and directly
the neurotransmitter dopamine is produced in the body. Per- through diet. The hydroxylation of tyrosine is the rate-limiting
sons with PKU either lack the enzyme phenylalanine hydrox- step in the production of dopamine.
ylase or have it in an inactive form. Note that the body acquires
448 COGNITION
TABLE 29.2
List of tasks
tasks requiring working memory plus inhibitory control A linear relationship between Phe level and performance The
(6 tasks X 4 comparisons per task), PKU children with higher a PKU child's current Phe level (the higher a
higher Phe levels performed significantly worse than child's Phe:Tyr ratio), the worse that child's perfor-
the comparison groups on 79% of these comparisons mance on the tasks that required the working memory
using the stringent criterion of p < .005 for each test to and inhibitory control functions dependent on DL-
correct for multiple comparisons (see table 29.3). This PFC. PKU children whose blood Phe levels had been
pattern of 19 out of 24 comparisons in the predicted maintained at 2-6 mg/dL performed comparably to all
direction would be very unlikely to occur by chance (p control groups on our tasks. Thus, at least in this sub-
< .004 [binomial distribution]). In short, the impair- group of PKU children, deficits in the ability to exercise
ment of the PKU children, whose blood Phe levels were working memory and inhibitory control simultaneously
3-5 times above normal, on the tasks that require the did not appear to be a necessary, unavoidable conse-
working memory and inhibitory control functions de- quence of being born with PKU. The effect of elevated
pendent on DL-PFC was clear and consistent. Phe levels appeared to be acute rather than chronic:
This finding of deficits in the working memory and Performance on these tasks was most strongly and con-
inhibitory control abilities dependent on DL-PFC in sistently related to current blood Phe levels, not mean
PKU children whose blood Phe levels are mildly ele- Phe levels over a wide age range during the first year of
vated (3-5 X normal) is consistent with the results of a life or during the first month of life. As current Phe
number of other studies. The most relevant are those levels varied, so too, inversely, did behavioral perfor-
by Welsh and colleagues (1990) and Smith and col- mance on five of the six tasks that required acting
leagues (1996), as these investigators used cognitive counter to one's initial tendency on the basis of infor-
tasks tailored to the functions of DL-PFC. mation held in mind (the exception being A-not-B with
The cognitive deficits documented in many studies of invisible displacement). Indeed, over time, changes in
children treated for PKU could be explained away by blood Phe levels within the same child were accompanied
saying that (1) the blood Phe levels of many of the chil- by concomitant, inverse changes in performance on
dren were outside the "safe" range (i.e., > 5 X normal); these cognitive tasks.
(2) even if current Phe levels were not excessively ele- The findings that performance is most closely tied to
vated, earlier Phe levels had been (during the years the current blood Phe levels (rather than to Phe levels ear-
children had been off diet); and/or (3) the low-Phe diet lier in life) and that performance covaries with a child's
had been started too late to avert early brain damage. current blood Phe levels is consistent with the afore-
Those disclaimers are not applicable to the study done mentioned biological mechanism concerning the cause
by Diamond and colleagues (1997). of the cognitive deficits. That is, these findings are con-
450 COGNITION
FIGURE 29.7 Performance of PKU children whose blood Phe from the general population. Note also that they are signifi-
levels are 6-10 mg/dL (3-5 times normal) on tasks requiring cantly impaired in the youngest age range investigated (as in-
both working memory and inhibitory control. Note that they fants they are impaired on the A-not-B/delayed response and
are significantly impaired compared to each comparison object retrieval tasks) and in the oldest age range investigated
group: other PKU children with Phe levels closer to normal, (as young children on the day-night, tapping, and three-pegs
siblings of the PKU children, control children matched to the tasks). (Reprinted with permission from Diamond et al., 1997.)
PKU children on a large number of variables, and children
sistent with an effect of reduced dopamine on prefron- Like us, Welsh and colleagues (1990) and Smith and
tal cortex function, which would vary directly with colleagues (1996) found that performance on measures
changes in the Phe:Tyr ratio in the bloodstream, as op- of DL-PFC function was significantly and negatively cor-
posed to structural, neuroanatomical changes, which related with concurrent Phe levels and less so with life-
might be more fixed. time Phe levels. Brunner, Jordan, and Berry (1983)
found that cognitive neuropsychological performance A developmental delay or absolute, lasting deficits ? Are the
was significantly correlated with concurrent Phe levels cognitive deficits in treated PKU children indicative of
but not with Phe levels during infancy. Using IQ and a developmental delay or of lasting deficits? On the one
school achievement as the outcome measures, Dobson hand, all children, even PKU children with Phe levels
and colleagues (1976) also found a significant, negative 3-5 times above normal, improved over time on our
correlation with concurrent blood Phe levels, and a tasks. On the other hand, the impression that PKU chil-
much weaker association with Phe levels earlier in life. dren may "catch up" to other children is probably mis-
Like us, Stemerdink and colleagues (1995) found that leading. In almost all cases this "catch up" was due to
when blood Phe levels were kept below 3X normal ceiling effects: The same tasks were administered over
from birth to the present, PKU children showed no cog- a wide age range, and these tasks were often too easy
nitive deficits. The only contrary finding is the report for children at the upper end of an age range. We have
of Sonneville and colleagues (1990), who found that repeatedly found that the between-group differences
Phe levels during the 2 years preceding cognitive testing reappeared on the next battery of tasks for the next age
were a better predictor of speed of responding on a group. The impairment of the PKU children with
continuous performance test than were concurrent Phe higher Phe levels in simultaneously holding informa-
levels. tion in mind and inhibiting a prepotent response was
The relationship found between blood Phe level and as evident in our oldest age range (31/2-7years) as it
performance in three studies (Diamond et al., 1997; was in our youngest age range (6-12 months). The def-
Smith and Beasley, 1989; Welsh et al., 1990) is particu- icit showed no evidence of subsiding within the age
larly impressive considering the truncated range of Phe range we studied (6 months to 7 years).
levels; all PKU children in those studies were on a die- The oldest children tested by Diamond and col-
tary regimen and their Phe levels were generally within leagues (1997) were 7 years old. One cannot tell from
the "acceptable" range. Because participants in the our study whether some time after 7 years PKU children
study by Diamond and colleagues (1997) were followed whose Phe levels remain only moderately elevated
longitudinally, we are able to present evidence for the might no longer show the kinds of cognitive deficits we
first time that performance on tasks requiring the work- have documented. Many studies of elementary school-
ing memory and inhibitory control functions of DL- age PKU children on the low-Phe diet have found cog-
PFC covaried inversely with Phe levels in the same child nitive deficits (Smith and Beasley, 1989; Weglage et al.,
over time. Because of the evidence of cognitive deficits 1995; Welsh et al., 1990). Recent studies by Ris and col-
in PKU children whose blood Phe levels are 6-10 mg/ leagues (1994) and Smith and colleagues (1996) report
dL, the national guidelines for the treatment of PKU deficits in the cognitive abilities dependent on prefron-
have been changed in the United Kingdom, the Neth- tal cortex in young adults with PKU. However, dietary
erlands, and Denmark: Phe levels higher than 6 mg/dL compliance tends to become progressively more lax af-
are no longer considered acceptable. In addition, sev- ter children enter school, so that these studies have in-
eral clinics in the United States have similarly revised cluded participants whose blood Phe levels were higher
their guidelines. than 10 mg/dL. What would happen if blood Phe levels
452 COGNITION
were maintained at 3-5 times normal: Would the cog- Welsh and colleagues (1990) and Smith and colleagues
nitive deficits eventually disappear? The data to answer (1996) who found (1) greater impairments on tasks de-
that question do not currently exist. Amino acid uptake pendent on prefrontal cortex than on tasks dependent
across the blood-brain barrier changes during devel- on parietal cortex or the medial temporal lobe in those
opment, offering more protection against blood Phe treated early and continuously for PKU, and (2) an in-
elevations as children get older (Greengard and Brass, verse relationship between Phe levels and performance
1984; Lajtha, Sershen, and Dunlop, 1987). Thus, it is on tasks dependent on prefrontal cortex function but
quite possible that the blood Phe levels we found to be no such relationship for tasks dependent on parietal
detrimental during infancy and early childhood might cortex or the medial temporal lobe. This is an example
be more benign in later childhood or adolescence. of a very specific, selective effect resulting from a global
Early cognitive deficits or developmental delays—es- insult (a moderate elevation in Phe and a moderate re-
pecially when they extend over a long period (such as duction in Tyr in the bloodstream that feeds the entire
the 6-year period we have documented) —are likely to body, and moderately too little Tyr in the entire brain).
have profound and enduring effects, even if the cog- The reason for the specificity is the differential, unique
nitive deficits themselves are subsequently resolved. sensitivity of prefrontally projecting dopamine neurons
They affect children's perceptions of, and expectations to a mild reduction in the dopamine precursor,
for, themselves and the perceptions and expectations of tyrosine.
others for the children. Such perceptions and expec- This finding of a deficit in the working memory and
tations can be inordinately difficult to change and can
inhibitory control functions of DL-PFC, but not in the
have major effects in shaping development and
cognitive functions dependent on other neural systems,
behavior.
is consistent with the mechanism I have hypothesized
as the cause of the cognitive deficits: A moderate im-
Selective, rather than global, cognitive deficits The same
balance in the Phe:Tyr ratio in blood (as when Phe lev-
children who were impaired on all six working memory
els are 3-5 times normal in PKU children) adversely
plus inhibitory control tasks performed well on the ten
affects the dopamine concentration in prefrontal cor-
control tasks, which required other cognitive abilities
tex but not other dopamine systems in the brain be-
dependent on other neural systems such as parietal cor-
cause of the special properties of the prefrontally
tex or the medial temporal lobe. Performance on the
projecting dopamine neurons, which makes them un-
control tasks, moreover, was not related to current
blood Phe levels. For each of the ten control tasks, we usually vulnerable to modest reductions in the level of
compared the performance of PKU children with Tyr reaching the brain. The specificity of the deficits
higher blood Phe levels (6-10 mg/dL; 3-5 X normal) suggests that the cause of those deficits is probably too
to that of the four comparison groups: other PKU chil- little Tyr reaching the brain, rather than too much Phe
dren with lower blood Phe levels, siblings of the PKU reaching the brain, because all neural regions would be
children, matched controls, and children from the gen- equally vulnerable to the negative effects of too much
eral population. This yielded a total of 40 pairwise com- Phe; the functions of DL-PFC would not be dispropor-
parisons (10 tasks X 4 comparisons per task). PKU tionately affected. That is, if the cause of the cognitive
children with higher Phe levels performed worse on deficits were too much Phe in the brain, the cognitive
only 10% of these comparisons (see table 29.3). This deficits should be global, rather than limited to the pre-
pattern of 36 out of 40 comparisons in the predicted frontal neural system.
direction would be extremely unlikely to occur by
chance (p < .001 [z distribution]). The consistency of Findings we had not predicted: Preserved performance on self-
the deficits of the PKU children with Phe levels 3-5 ordered pointing and temporal order memory tasks The
times normal on the working memory plus inhibitory mechanism I have proposed to explain the cause of the
control tasks and the paucity of deficits on the control cognitive deficits in children treated early and contin-
tasks is quite striking: 55 out of 64 comparisons in the uously for PKU, whose Phe levels are 3-5 times normal,
predicted direction (86%), p < .0001, z distribution. rests on the special properties of the dopamine neurons
Thus, for children treated early and continuously for that project to prefrontal cortex. I had not hypothesized
PKU whose blood Phe levels are 3-5 times normal, the that only certain cognitive functions dependent on DL-
cognitive deficits appear to be selective. The functions PFC would be affected. I was surprised, therefore, when
of parietal cortex and of the medial temporal lobe ap- we found that PKU children with Phe levels 3-5 times
pear to be spared, even if the children's Phe levels go normal performed normally on three tasks dependent
up to 6-10 mg/dL. This is consistent with reports by on DL-PFC: the three- and six-boxes tasks (boxes scram-
454 COGNITION
TABLE 29.4
Summary of the results of the 1998 study by Collins, Roberts, Dias, Everitt, and Robbins
Behavioral Task
Delayed response Self-ordered pointing
Type of lesion to frontal cortex (requires working memory + inhibition) (requires working memory
Excitotoxic (cell bodies destroyed) Performance impaired Performance impaired
6-OHDA (dopamine depleted) Performance impaired Performance spared
impairments on the day-night and tapping tasks, which testing was randomized across experimental condition.
are also nonspatial. Blood samples were collected at multiple time points to
determine the animals' Phe levels. High-performance
An animal model of mild, chronic plasma liquid chromatographic (HPLC) analyses of the brain
Phe elevations tissue assessed the distributions and concentrations of
dopamine, serotonin, norepinephrine, and their me-
With children it was possible only to measure blood lev- tabolites in various brain regions (prefrontal cortex,
els of Phe and Tyr and cognitive performance. To in- caudate-putamen, and nucleus accumbens).2
vestigate the biological mechanism underlying the The most dramatic neurochemical effects of the
cognitive deficits of children treated for PKU more di- moderate elevation in blood Phe levels were the reduc-
rectly, we developed and characterized the first animal tion in dopamine and in the dopamine metabolite HVA
model of treated PKU (Diamond et al., 1994) and sub- in prefrontal cortex in each of the PKU-model animals.
sequently worked with the genetic mouse model of PKU There was almost no overlap between HVA levels in the
(Zagreda et al., 1999). The animal model enabled us to prefrontal cortex of controls and either PKU-model
study the effect of moderate, chronic plasma Phe ele- group: All control animals but one had higher HVA lev-
vations on neurotransmitter and metabolite levels in els in prefrontal cortex than any animal in either ex-
specific brain regions. Thus, we could directly investi- perimental group. In contrast, as predicted, the levels
gate our hypothesis that the cognitive deficits associated of dopamine and dopamine metabolites were not re-
with moderately elevated plasma Phe levels are pro- duced elsewhere in the brain, and norepinephrine lev-
duced by a selective reduction in dopamine synthesis in els were not reduced elsewhere in the brain or in
prefrontal cortex. prefrontal cortex. We had predicted that norepineph-
Building on work modeling the untreated PKU con- rine levels would be unaffected (even though norepi-
dition (Brass and Greengard, 1982; Greengard, Yoss, nephrine is made from dopamine) because previous
and Del Valle, 1976), Diamond and colleagues (1994) work had shown that norepinephrine levels are rela-
administered a phenylalanine hydroxylase inhibitor (a- tively insensitive to alterations in precursor availability
methylphenylalanine) plus a small supplement of Phe (Irie and Wurtman, 1987).
to mildly and chronically elevate the blood Phe levels The PKU-model animals were impaired on delayed
in rat pups. (The Phe supplement was needed because alternation in the same ways and under the same con-
a-methylphenylalanine does not inhibit phenylalanine ditions as are animals with prefrontal cortex lesions. On
hydroxylase completely.) There were two experimental the delayed alternation task, the animal is rewarded
groups: (1) pups whose blood Phe levels were elevated only for alternating goal arms (i.e., for selecting the
postnatally, and (2) pups whose blood Phe levels were goal arm not selected on the previous trial). Thus, the
elevated pre- and postnatally. Control animals came animal must remember which goal arm was last entered
from the same litters as the first group and received over the delay between trials and must inhibit repeating
daily control injections of saline. that response. The hallmark of performance after pre-
All were tested on delayed alternation, a task sensitive frontal cortex is removed is that subjects fail when a
to prefrontal cortex dysfunction (Battig, Rosvold, and delay is imposed between trials, although they are un-
Mishkin, 1960; Bubser and Schmidt, 1990; Kubota and impaired at learning the alternation rule or in perform-
Niki, 1971; Larsen and Divac, 1978; Wikmark, Divac, ing the task when no delay is imposed (in rats: Bubser
and Weiss, 1973). Testers were blind to the group as- and Schmidt, 1990; Larsen and Divac, 1978; Wikmark,
signment of their animals. Each of the testers was as- Divac, and Weiss,1973; in monkeys: Battig, Rosvold, and
signed four animals in each group and the order of Mishkin, 1960; Jacobsen and Nissen, 1937; Kubota and
FIGURE 29.8 Rats with chronic, mild elevations in their blood to dorsolateral prefrontal cortex has been lesioned. That is,
Phe levels, to create an animal model of early and continu- they can learn the delayed alternation rule and perform well
ously treated PKU, show the same pattern of performance on when there is no delay, but are impaired when a delay is
the delayed alternation task, as do monkeys whose dorsolateral introduced.
prefrontal cortex has been lesioned and rats whose homolog
456 COGNITION
ing and dopamine turnover rates (Fernstrom, Volk, and
Fernstrom, 1986; luvone et al., 1978, 1989). Moreover,
the competition between Phe and Tyr at the blood-
retinal barrier is fully comparable to their competitive
uptake at the blood-brain barrier (Fernstrom, Volk,
and Fernstrom, 1986; Hjelle et al., 1978; Rapoport,
1976; Tornquist and Aim, 1986). Indeed, it has been
shown that a small reduction in the level of Tyr reaching
the retina dramatically reduces retinal dopamine syn-
thesis (Fernstrom and Fernstrom, 1988; Fernstrom,
Volk, and Fernstrom, 1986), mirroring the effect on do-
pamine synthesis in prefrontal cortex. Therefore, to be
consistent, we had to predict that retinal function
should also be affected in PKU children who have been
on a low-Phe diet since the first month of life, but who
have moderately elevated blood Phe levels (levels FIGURE 29.9 PKU children whose blood Phe levels are 6-10
roughly 3-5 times normal [6-10 mg/dL; 360-600 mg/dL were found to be significantly impaired in contrast
l^mol/L]), even though no visual deficit had been re- sensitivity compared to children of the same age at every spa-
ported in these children before. tial frequency investigated.
The aspect of retinal function most firmly linked to
the level of dopamine in the retina is contrast sensitivity.
Contrast sensitivity refers to the ability to detect differ- IQs were below 90 were omitted from the analyses. In-
ences in luminance (brightness) of adjacent regions in deed, at the next to the highest spatial frequency (12
a pattern. Your contrast sensitivity threshold is the limit cycles per degree), the "group" variable accounted for
of how faint items printed in gray can become before 70% of the variance, controlling for acuity, gender, age,
you fail to perceive them at all. People with good con- and test site. At no spatial frequency was the contrast
trast sensitivity can perceive fainter lines than those who sensitivity of any PKU child better than that of his or
require a greater luminance difference between fore- her own sibling. Standard eye exams had never detected
ground and background. Patients with Parkinson's dis- a problem in this population because acuity is normally
ease, who have greatly reduced levels of dopamine, tested under conditions of high contrast; an impair-
have impaired contrast sensitivity (Bodis-Wollner, 1990; ment in contrast sensitivity was not revealed before be-
Bodis-Wollner et al., 1987; Kupersmith et al., 1982; Re- cause no one had tested for it.
gan and Neima, 1984; Skrandies and Gottlob, 1986). It At the time, we interpreted these results as providing
is thought that this occurs because dopamine is impor- converging evidence in support of the biological mech-
tant for the center-surround organization of retinal re- anism I had proposed. I had predicted the contrast sen-
ceptive fields (Bodis-Wollner, 1990; Bodis-Wollner and sitivity deficit for the same reason I had predicted
Piccolino, 1988). DL-PFC cognitive deficit. Both predictions had been
To investigate contrast sensitivity, we (Diamond and based on the special sensitivity of dopamine neurons
Herzberg, 1996) tested children between the ages of 5.4 that fire rapidly and turn over dopamine rapidly to
and 9.8 years on the Vistech test (Gilmore and Levy, moderate reductions in the level of available tyrosine.
1991; Ginsberg, 1984; Lederer and Bosse, 1992; Man- We had found two superficially unrelated behavioral ef-
tyjarvi et al., 1989; Rogers, Bremer, and Leguire, 1987; fects, a selective deficit in cognitive functions depen-
Tweten, Wall, and Schwartz, 1990). We found that chil- dent on DL-PFC and a selective visual defect in contrast
dren treated early and continuously for PKU, whose sensitivity, both of which had been predicted based on
blood Phe levels were 6-10 mg/dL (3-5X normal), the same underlying hypothesis.
were impaired in their sensitivity to contrast at each of However, I was troubled by one lack of convergence.
the five spatial frequencies tested (1.5-18.0 cycles per In the Diamond and Herzberg study (1996), we found
degree; see figure 29.9). Even though all children had that contrast sensitivity performance did not correlate
been tested under conditions of 20/20 acuity, the PKU with children's current blood Phe levels, but rather with
children were significantly less sensitive to visual con- their Phe levels during the first month of life. In an-
trast than their same-aged peers across the entire range other study (Diamond et al., 1997), however, we had
of spatial frequencies. These group differences re- found that, on the cognitive tasks that required the
mained robust even when the two PKU children whose working memory and inhibitory control functions
458 COGNITION
candy better than their younger siblings, the reverse of 1999; Smith andjonides, 1999; Smith et al., 1998). Un-
the pattern seen in the PKU sibling pairs. The earlier der such conceptualizations, holding information in
born PKU siblings (who started the diet at 1 1/2-2 weeks mind plus inhibiting a dominant response becomes
of age) also showed worse contrast sensitivity than their part of a subset of "holding information in mind plus
same-age peers. another cognitive operation." I am in complete accord
These results suggest that extremely high Phe levels with such formulations. In general, tasks that require
during the first weeks of life, even if subsequently low- DL-PFC are more difficult than tasks that do not. How-
ered and maintained at lower levels, may cause long- ever, if one increases how much information must be
lasting damage to the visual system. Evidently, a held in mind so that the task is as difficult as one that
short-term exposure to high concentrations of Phe of requires both holding information in mind plus either
only a couple of weeks during the sensitive neonatal inhibition (Diamond, O'Craven, and Savoy, 1998) or
period can have long-lasting effects on contrast sensitiv- alphabetizing the information held in mind (Postle,
ity, evident 9-16 years later. This is significant because Berger, and D'Esposito, 1999; Rypma and D'Esposito,
it suggests that the current practice of allowing up to 2 1999), then that task, too, will activate DL-PFC.
weeks to pass before beginning treatment for an infant There may be differential developmental trajectories
born with PKU may be ill-advised. Since the blood sam- for the working memory and inhibitory control abilities
ple to test for PKU is taken at birth, it would be feasible dependent on DL-PFC. Between 71/2-12 months of age
to start the diet earlier. These results also suggest that there is evidence of clear age-related improvements in
although we obtained the results I had predicted for inhibitory control (from results with the object retrieval
contrast sensitivity in the Diamond and Herzberg task) and in how long infants can hold information in
(1996) study, we may have obtained those results for mind (from results with the A-not-B task) (Diamond,
reasons other than the ones predicted. The deficits in 1988, 1991). For example, infants make the A-not-B er-
the working memory and inhibitory control abilities de- ror at delays under 2 s at 7 1/2-8 months and at delays of
pendent on DL-PFC do indeed appear to occur for the 10 s or more at 12 months, an average improvement of
reason hypothesized (because of reduced levels of do- about 2 s per month in their ability to hold information
pamine in DL-PFC due to elevated blood Phe:Tyr ra- in mind (see figure 29.11). On the other hand, between
tios). Those deficits covary with concurrent levels of 4 and 22 years of age there are marked improvements
Phe in the bloodstream. However, the retinal deficit in in inhibitory control but little improvement in the abil-
contrast sensitivity appears to be caused, at least in part, ity to hold information in mind, which appears to be
by the inordinately high levels of Phe in the first weeks quite robust even by the age of 4 (see figures 29.12 and
of life, does not covary with current levels of Phe, and
appears to be structural.
Conclusions
FIGURE 29.12 Performance on our directional Stroop task patibility"). The addition of an inhibitory requirement low-
when inhibitory control was not required minus when it was ered the accuracy (panel A) and increased the reaction times
required. Trials that required inhibitory control were trials (panel B) of younger children more than older children and
where participants had to resist the tendency to respond with older children more than adults.
the hand on the same side as the stimulus ("spatial incom-
460 COGNITION
FIGURE 29.13 Performance on our directional Stroop task load was greater was harder for all participants, but it was not
when memory load was greater minus when it was less. Mem- differentially harder for any age group. Hence, the effect
ory load was greater when the rules for 6 abstract shapes had on the accuracy (panel A) and on reaction times (panel B) of
to be remembered than when the rules for only 2 abstract younger children, older children, and adults were compa-
shapes had to be remembered. The condition where memory rable.
decreases in the amount of precursor (i.e., Tyr). How- nal function would also be affected in early and contin-
ever, the dopamine neurons that project to prefrontal uously treated PKU children whose Phe levels are 3-5
cortex are different. They fire faster and turn over do- times normal. Indeed, we found the predicted impair-
pamine faster, and are acutely sensitive to even a modest ment in contrast sensitivity. However, this effect appears
change in the level of Tyr. Because of the special prop- to be due to the exceedingly high Phe levels during the
erties of this dopamine projection, we predicted and first two weeks of life, when most children with PKU
found a specific, localized effect (prefrontal cortex af- have not yet begun treatment.
fected but not other regions of the brain) even though Cognitive deficits in children treated early and con-
the insult is global (a mildly increased PherTyr ratio tinuously for PKU whose blood Phe levels are 3-5 times
throughout the bloodstream and mildly reduced Tyr normal went officially unrecognized for years, despite
levels throughout the brain). the protestations of parents and teachers that some-
The dopamine neurons in the retina share the same thing was wrong, in part because the children per-
special properties as the dopamine neurons that project formed within the normal range on IQ tests. Their IQ
to prefrontal cortex. We predicted, therefore, that reti- scores were in the 80s and the 90s, just as are the IQ
462 COGNITION
FIGURE 29.14 (A) Figure illustrating the stimuli in the "dots- opposite instructions. Participants were given a two-button re-
side" variant of the directional Stroop task. Here, half the par- sponse box: one button for their right thumb and one for their
ticipants were instructed to press the response button on the left.)
same side as the dot if the dot is striped and on the opposite (continued on following page)
side if the dot is gray. (Half the participants were given the
ipant sees one of six abstract figures at the midline, and ride the strong tendency to go straight to that goal when
needs to remember the rule associated with each figure an indirect route is required (as in the object retrieval
("press right" or "press left"; see figure 29.14c). The and windows tasks). To sustain the focused concentra-
abstract-center-six variant taxes working memory tion required for hearing someone in a noisy room or
heavily (participants must hold six rules in mind), but for a difficult task, one needs to be able to resist dis-
it requires little or no inhibition (as the stimuli appear traction; to relate multiple ideas to one another, one
at the center of the screen and do not preferentially needs to resist focusing exclusively on only one idea;
activate the right or left hand). It is work with this task when visual perception is misleading, one needs to be
that produced the results on the age-related improve- able to resist acting in accord with what one sees; and
ments between 4 and 22 years in inhibitory control, but to act in new ways, one needs to resist falling back into
not in the amount of information that could be held in one's usual way of acting or thinking. That is, one needs
mind, as illustrated in figures 29.12 and 29.13. inhibitory control, dependent upon prefrontal cortex.
Because inhibitory control undergoes such a pro- The ability to exercise inhibitory control, which pre-
tracted development, young children often fail to in- frontal cortex makes possible, frees us to act according
hibit the prepotent response, despite their best to what we choose to do rather than being simply crea-
intentions and despite knowing what they should do. It tures of habit or immediate perception. The ability to
would be a shame to mistakenly label such a young child hold information in mind, which also depends upon
as "bad," "stupid," or "willful." It is not enough to know prefrontal cortex, enables us to remember what we are
something or remember it; one must get that knowl- supposed to do, to consider alternatives, to remember
edge into one's behavior. Infants and young children, the past and consider the future, and to use what we
in whom prefrontal cortex is not yet mature, sometimes know—not just what we see—to help guide our actions
do the wrong thing even though they know what they and choices. These abilities make it possible for us to
should do and are trying to do it. Their attention is solve new, undreamed-of challenges, permitting us to
sometimes so captured by the desired goal object that exercise free will and self-determination. Not that it's
they either cannot inhibit responding (as in delay of easy, of course; but prefrontal cortex helps make it
gratification or Go-NoGo paradigms) or cannot over- possible.
464 COGNITION
NOTES BATTIG, K., H. E. ROSVOLD, and M. MISHKIN, 1960. Compar-
ison of the effects of frontal and caudate lesions on delayed
1. At the core of my hypothesis about the cause of the A-not-
response and alternation in monkeys. J. Comp. Physiol.
B error has always been the notion of a competition or 53:400-404.
battle between the information held in mind (i.e., where BAUER, R. H., and J. M. FUSTER, 1976. Delayed-matching and
the toy was last hidden) and the prepotent tendency (a type delayed-response deficit from cooling dorsolateral prefron-
of procedural or implicit memory) formed by the experi- tal cortex in monkeys.J. Comp. Physiol. Psychol 90:293-302.
ence of previous trials. The key element is conflict: What is BECKER, M. G., W. ISAAC, and G. W. HYND, 1987. Neuropsy-
required is not simply holding in mind the newest infor- chological development of nonverbal behaviors attributed
mation, but that stored information has to win against a to "frontal lobe" functioning. Dev. Neuropsychol. 3:275-298.
competitor (a conditioned tendency), which is probably BELL, J. A., and P. J. LIVESEY, 1985. Cue significance and re-
subcortical in origin, since even extremely simple organ- sponse regulation in 3- to 6- year-old children's learning of
isms can show conditioned tendencies. multiple choice discrimination tasks. Dev. Psychobiol. 18:229-
2. We had intended to include two regions of frontal cor- 245.
tex (prefrontal cortex and the anterior cingulate). BELL, M. A., and N. A. Fox, 1992. The relations between fron-
However, it turned out the brain sections we sampled to tal brain electrical activity and cognitive development dur-
assess the anterior cingulate were actually from prefrontal ing infancy. Child Dev. 63:1142-1163.
cortex. BELL, M. A., and N. A. Fox, 1997. Individual difference in
object permanence performance at 8 months: Locomotor
ACKNOWLEDGMENTS The work summarized here was sup- experience and brain electrical activity. Dev. Psychobiol.
ported by grants from NIMH (R01-#MH41842, R01- 31:287-297.
#MH09007, R01-#MH00298, & R01-#MH38456), NICHD BENTON, A. L., 1969. Disorders of spatial orientation. In P.
(R01-#HD35453, R01-#HD34346, R01-#HD10094, P30- Vincken and G. Bruyn, eds., Handbook of Clinical Neurology,
Vol. 3. Amsterdam: North-Holland Publishing.
#HD26979, P30-#HD04147), BRSG (S07-#RR07054-22, S07-
BERRY, H. K., D. J. O'GRADY, L. J. PERLMUTTER, and M. K.
#RR07083-23, & S07-#RR07083-26), March of Dimes (#12-253
BOFINGER, 1979. Intellectual development and achievement
& #12-0554), NSF (Doctoral Dissertation Grant
of children treated early for phenylketonuria. Dev. Med.
#BNS8013447), the Sloan Foundation, and the Arc Founda-
Child Neurol. 21:311-320.
tion. I would like to thank Natasha Kirkham and Kristin Shutts
BICKEL, H., F. P. HUDSON, and L. I. WOOLF, 1971. Phenylketon-
for their astute comments on an earlier draft.
uria and Some Other Inborn Errors of Metabolism. Stuttgart:
Georg Thiese Verlag.
BJORKLUND, A., I. DIVAC, and O. LINDVALL, 1978. Regional
REFERENCES distribution of catecholamines in monkey cerebral cortex,
evidence for a dopaminergic innervation of the primate
AKERT, K., 1964. Comparative anatomy of frontal cortex and prefrontal cortex. Neurosci. Lett. 7:115-119.
thalamofrontal connections. In The Frontal Granular Cortex BODIS-WOLLNER, L, 1990. Visual deficits related to dopamine
and Behavior, ]. M. Warren and K. Akert, eds. New York: deficiency in experimental animals and Parkinson's disease
McGraw-Hill, pp. 372-376. patients. Trends Neural Sci. 13:296-302.
AKIL, M., J. N. PIERRI, R. E. WHITEHEAD, C. L. EDGAR, C. BRADBERRY, C. W., D. H. KARASIC, A. Y. DEUTSCH, and R. H.
MOHILA, A. R. SAMPSON, and D. A. LEWIS, 1999. Lamina- ROTH, 1989. Regionally-specific alterations in mesotelen-
specific alterations in the dopamine innervation of the pre- cephalic dopamine synthesis in diabetic rats: Associations
frontal cortex in schizophrenic subjects. Am. J. Psychiatry with precursor tyrosine.J. Neural Transmission 78:221-229.
156:1580-1589. BRASS, C. A., and O. GREENGARD, 1982. Modulation of cere-
ARIKUNI, T., and K. KUBOTA, 1986. The organization of pre- bral catecholamine concentrations during hyperphenyla-
frontocaudate projections and their laminar origin in the laninaemia. Biochem. J. 208:765-771.
macaque monkey: A retrograde study using HRP-gel. /. BRODMANN, K., 1912. Neue ergebnisse iiber die vergleichende
Comp. Neurol 244:492-510. histologische lokalisation der grosshirnrinde mit beson-
BACHEVALIER, J., M. BRICKSON, and C. HAGGER, 1993. Limbic- derer berucksichtigung des stirnhirns. Anat. Am.
dependent recognition memory in monkeys develops early 41(suppl.):157-216.
BROWN, R. M., A. M. CRANE, and P. S. GOLDMAN, 1976. Cate-
in infancy. NeuroReport 4:77-80.
cholamines in neocortex of rhesus monkeys: Regional dis-
BACHEVALIER, J., and M. MISHKIN, 1986. Visual recognition
tribution and ontogenetic development. Brain Res. 124:576-
impairment follows ventromedial but not dorsolateral pre-
580.
frontal lesions in monkeys. Behau Brain Res. 20:249-261. BROWN, R. M., A. M. CRANE, and P. S. GOLDMAN, 1979. Re-
BADDELEY, A., 1992. Working memory. Science 255:556-559. gional distribution of monoamines in the cerebral cortex
BALAMORE, U., and R. H. WOZNIAK, 1984. Speech-action co- and subcortical structures of the rhesus monkey: Concen-
ordination in young children. Dev. Psychol. 20:850-858. trations and in vivo synthesis rates. Brain Res. 168:133-150.
BARBAS, H., and M. M. MESULAM, 1985. Cortical afferentinput BROWN, R. M., and P. S. GOLDMAN, 1977. Catecholamines in
to the principalis region of the rhesus monkey. Neuroscience neocortex of rhesus monkeys: Regional distribution and on-
15:619-637. togenetic development. Brain Res. 127:576-580.
BANNON, M. J., E. B. BUNNEY, and R. H. ROTH, 1981. Meso- BROZOSKI, T. J., R. M. BROWN, H. E. ROSVOLD, and P. S. GOLD-
cortical dopamine neurons: Rapid transmitter turnover MAN, 1979. Cognitive deficit caused by regional depletion
compared to other brain catecholamine systems. Brain Res. of dopamine in prefrontal cortex of rhesus monkey. Science
218:376-382. 205:929-932.
466 COGNITION
DIAMOND, A., and C. HERZBERG, 1996. Impaired sensitivity to FRITZ, A. S., 1991. Is there a reality bias in young children's
visual contrast in children treated early and continuously emergent theories of mind? Paper presented at the biennial
for PKU. Brain 119:101-116. meeting of the Society for Research in Child Development,
DIAMOND, A., K. M. O'CRAVEN, M. DAVIDSON, C. CRUESS, R. Seattle.
BERGIDA, and R. L. SAVOY, 1999b. Further fMRI-based stud- FUSTER, J. M., 1973. Unit activity in prefrontal cortex during
ies of memory and inhibition in prefrontal cortex of adults. delayed-response performance: Neuronal correlates of
Paper presented at Cognitive Neuroscience Society Annual transient memory. J Neurophysiol. 36:61-78.
Meeting, Washington, D.C., April. FUSTER, J. M., 1980. The Prefrontal Cortex: Anatomy, Physiology,
DIAMOND, A., K. M. O'CRAVEN, and R. L. SAVOY, 1998. Dor- and Neuropsychology of the Frontal Lobe. New York: Raven Press.
solateral prefrontal cortex contributions to working mem- FUSTER, J. M., and G.E. ALEXANDER, 1970. Delayed response
ory and inhibition as revealed by fMRL Soc. Neurosci. Abstr. deficit by cryogenic depression of frontal cortex. Brain Res.
24:1251. 61:79-91.
DIAMOND, A., M. PREVOR, G. CALLENDER, and D. P. DRUIN, FUSTER, J. M., and G. E. ALEXANDER, 1971. Neuron activity
1997. Prefrontal cortex cognitive deficits in children treated related to short-term memory. Science 173:652-654.
early and continuously for PKU. Monogr. Soc. Res. Child Dev. CASPAR, P., B. BERGER, A. FEBVRET, A. VIGNY, andj. P. HENRY,
62(4). 1989. Catecholamine innervation of the human cerebral
cortex as revealed by comparative immunohistochemistry
DIAMOND, A., and C. TAYLOR, 1996. Development of an aspect
of tyrosine hydroxylase and dopamine-(3-hydroxylase. J.
of executive control: Development of the abilities to remem-
Comp. Neurol 279:249-271.
ber what I said and to "Do as I say, not as I do." Dev. Psy-
GERSTADT, C., Y. HONG, and A. DIAMOND, 1994. The relation-
chobiol. 29:315-334.
ship between cognition and action: Performance of 3.5-7
DIAMOND, A., S. ZOLA-MORGAN, and L. R. SQUIRE, 1989. Suc-
year old children on a Stroop-like day-night test. Cognition
cessful performance by monkeys with lesions of the hippo- 53:129-153.
campal formation on A-not-B and object retrieval, two tasks GILMORE, G. C., and J. A. LEVY, 1991. Spatial contrast sensitiv-
that mark developmental changes in human infants. Behav. ity in Alzheimer's disease: A comparison of two methods.
Neurosci. 103:526-537. Optom. Vis. Sci. 68:790-794.
DILELLA, A. G., J. MARVIT, A. S. LIDSKY, F. GUTTLER, and S. GINSBERG, A. R., 1984. A new contrast sensitivity vision test
L. C. Woo, 1986. Tight linkage between a splicing mutation chart. Am.J. Optom. Physiol. Optics 61:403-407.
and a specific DNA haplotype in phenylketonuria. Nature GOLDMAN, P. S., and W. J. H. NAUTA, 1977. An intricately pat-
322:799-803. terned prefrontal-caudate projection in the rhesus monkey.
DOBSON, J. C., E. KUSHIDA, M. L. WILLIAMSON, and E. G. J. Comp. Neurol. 171:369-385.
FRIEDMAN, 1976. Intellectual performance of 36 phenylke- GOLDMAN, P. S., and H. E. ROSVOLD, 1970. Localization of
tonuric patients and their non-affected siblings. Pediatrics function within the dorsolateral prefrontal cortex of the
58:53-58. rhesus monkey. Exp. Neurol. 29:291-304.
DREWE, E. A., 1974. The effect of type and area of brain lesion GOLDMAN, P. S., H. E. .ROSVOLD, and M. MISHKIN, 1970. Evi-
on Wisconsin Card Sorting Test performance. Cortex dence for behavioral impairment following prefrontal lo-
10:159-170. bectomy in the infant monkey. J. Comp. Physiol. Psychol.
FAUST, D., D. LIBON, and S. PUESCHEL, 1986. Neuropsycho- 70:454-463.
logical functioning in treated phenylketonuria. Intl. J. Psy- GOLDMAN-RAKIC, P. S., 1987. Circuitry of primate prefrontal
chiatry Med. 16:169-177. cortex and regulation of behavior by representational mem-
FERNSTROM, J. D., and M. H. FERNSTROM, 1988. Tyrosine avail- ory. In Handbook of Physiology, the Nervous System, Higher Func-
ability and dopamine synthesis in the retina. In Dopaminergic tions of the Brain, Vol. 5, F. Plum, ed. Bethesda, Md.:
Mechanisms in Vision, I. Bodis-Wollner and M. Piccolino, eds. American Physiological Society, pp. 373-417.
New York: Alan R. Liss, pp. 59-70. GOLDMAN-RAKIC, P. S., and L. J. PORRINO, 1985. The primate
FERNSTROM, M. H., E. A. VOLK, and J. D. FERNSTROM, 1986. In mediodorsal (MD) nucleus and its projection to the frontal
vivo inhibition of tyrosine uptake into rat retina by large lobe.J Comp. Neurol. 242:535-560.
neutral, but not acidic, amino acids. Am. J. Physiol. GOLDMAN-RAKIC, P. S., and M. L. SCHWARTZ, 1982. Interdigi-
25LE393-E399. tation of contralateral and ipsilateral columnar projections
to frontal association cortex in primates. Science 216:755-
FITTS, P. M., and C. M. SEGER, 1953. S-R compatibility: Spatial
757.
characteristics of stimulus and response codes. J. Exp. Psy-
GOLDSTEIN, K., 1936. The modifications of behavior conse-
chol 81:174-176.
quent to cerebral lesions. Psychiatric Quart. 10:586-610.
FLAVELL, J. H., 1986. The development of children's knowl- GOLDSTEIN, K., 1944. The mental changes due to frontal lobe
edge about the appearance-reality distinction. Amer. Psychol- damage.J Psychol. 17:187-208.
ogist 41:418-425. GRATCH, G., K. J. APPEL, W. F. EVANS, G. K. LECOMPTE, and
FLAVELL, J. H., 1993. The development of children's under- N. A. WRIGHT, 1974. Piaget's stage IV object concept error:
standing of false belief and the appearance-reality distinc- Evidence for forgetting or object conception? Child Dev.
tion. Intl.J. Psychol. 28:595-604. 45:71-77.
Fox, N. A., and M. A. BELL, 1990. Electrophysiological indices GRATCH, G., and W. F. LANDERS, 1971. Stage IV of Piaget's
of frontal lobe development: Relations to cognitive and af- theory of infant's object concepts: A longitudinal study.
fective behavior in human infants over the first year of life. Child Dev. 42:359-372.
Ann. N.Y. Acad. Sci. 608:677-704. GREENGARD, O., and C. A. BRASS, 1984. Developmental
Fox, N., J. KAGAN, and S. WEISKOPF, 1979. The growth of mem- changes of cerebral phenylalanine uptake from severely el-
ory during infancy. Genetic Psychol. Monogr. 99:91-130. evated blood levels. Neurochem. Res. 9:837-848.
468 COGNITION
LEWIS, D. A., S. L. FOOTE, M. GOLDSTEIN, andj. H. MORRISON, MILES, R. C., and A. J. BLOMQUIST, 1960. Frontal lesions and
1988. The dopaminergic innervation of monkey prefrontal behavioral deficits in monkey.J. Neurophysiol. 23:471-484.
cortex: A tyrosine hydroxylase immunohistochemical study. MILLER, L., L. D. BRAUN, W. M. PARDRIDGE, and W. H. OLD-
Brain Res. 565:1-13. ENDORF, 1985. Kinetic constants for blood-brain, barrier
LEWIS, D. A., and H. W. HARRIS, 1991. Differential laminar amino acid transport in conscious rats. J Neurochem.
distribution of tyrosine hydroxylase-immunoreactive axons 45:1427-1432.
in infant and adult monkey prefrontal cortex. Neurosci. Lett. MILNER, B., 1963. Effects of different brain lesions on card
125:151-154. sorting: The role of the frontal lobes. Arch. Neurol. 9:90-100.
LEWIS, D. A., S. R. SESACK, A. I. LEVEY, and D. R. ROSENBERG, MILNER, B., 1964. Some effects of frontal lobectomy in man.
1998. Dopamine axons in primate prefrontal cortex: Spec- In The Frontal Granular Cortex and Behavior,J. M. Warren and
ificity of distribution, synaptic targets, and development. K. Akert, eds. New York: McGraw-Hill, pp. 313-334.
Adv. Pharmacol. 42:703-706. MILNER, B., 1982. Some cognitive effects of frontal-lobe lesions
LIDOW, M. S., and P. RAKIC, 1992. Scheduling of monoamin- in man. Phil. Trans. R Soc. (London), 298(suppl B):211-226.
ergic neurotransmitter receptor expression in the primate MILNER, B., P. CORSI, and G. LEONARD, 1991. Frontal-lobe con-
neocortex during postnatal development. Cereb. Cortex tribution to recency judgments. Neuropsychologia 29:601-
2:401-416. 618.
LIDSKY, A. S., M. L. LAW, H. G. MORSE, F. T. KAO, and S. L. C. MISCHEL, H. N., and W. MISCHEL, 1983. The development of
Woo, 1985. Regional mapping of the human phenylalanine children's knowledge of self-control strategies. Child Dev.
hydroxylase gene and the PKU locus on chromosome 12. 54:603-619.
Proc. Natl. Acad. Sci. USA 82:6221-6225. MORRIS, R., D. N. PANDYA, and M. PETRIDES, 1999. Fiber sys-
LIVESEY, D.J., and G. A. MORGAN, 1991. The development of tem linking the mid-dorsolateral frontal cortex with the re-
response inhibition in 4- and 5-year-old children. Austral. J. trosplenial/presubicular region in the rhesus monkey. /.
Psychol. 43:133-137. Comp. Neurol. 407:183-192.
LUCIANA, M., and P. F. COLLINS, 1997. Dopaminergic modu- MORRIS, R., M. PETRIDES, and D. N. PANDYA, 1999. Architec-
lation of working memory for spatial but not object cues in ture and connections of retrosplenial area 30 in the rhesus
normal humans.J. Cogn. Neurosci. 9:330-347. monkey (Macaca mulatto). Eur.J. Neurosci. 11:2506-2518.
LUCIANA, M., R. A. DEPUE, P. ARBISI, and A. LEON, 1992. Fa-
NIKI, H., 1974. Differential activity of prefrontal units during
cilitation of working memory in humans by a D2 dopamine
right and left delayed response trials. Brain Res. 70:346-349.
receptor agonist.J. Cogn. Neurosci. 4:58-68.
NORD, A. M., L. McCABE, and E. R. McCABE, 1988. Biochem-
LURIA, A. R., and E. D. HOMSKAYA, 1964. Disturbance in the
ical and nutritional status of children with hyperphenylalan-
regulative role of speech with frontal lobe lesions. In The
inaemia,J Inherit. Metab. Disorders 11:431-432.
Frontal Granular Cortex and Behavior, J. M. Warren and K.
OLDENDORF, W. H., 1973. Stereospecificity of blood-brain
Akert, eds. New York: McGraw-Hill, pp. 353-371.
barrier permeability to amino acids. Am. J. Physiol. 224:967-
LURIA, A. R., 1966. The Higher Cortical Functions in Man. New
York: Basic Books. 969.
MANTYJARVI, M. L, M. H. AUTERE, A. M. SILVENNOINEN, and ORZHEKHOVSKAYA, N. S., 1981. Fronto-striatal relationships in
T. MYOHANEN, 1989. Observations of the use of three dif- primate ontogeny. Neurosci. Behau Physiol. 11:379-385.
ferent contrast sensitivity tests in children and young adults. OWEN, A. M., N. J. HERROD, D. K. MENON, J. C. CLARK, S. P.
/. Pediatr. Ophthalmol. Strabismus 26:113-119. DOWNEY, T. A. CARPENTER, P. S. MINHAS, F. E. TURKHEIMER,
MCDONALD, J. D., V. C. BODE, W. F. DOVE, and A. SHEDLOVSKY, E. J. WILLIAMS, T. W. ROBBINS, B. J. SAHAKIAN, M. PETRIDES,
1990. Pahhph-5: A mouse mutant deficient in phenylala- and J. D. PICKARD, 1999. Redefining the functional orga-
nine hydroxylase. Proc. Natl. Acad. Sci. 87:1965-1967. nization of working memory processes within human lateral
McGuiRE, P. K.,J. F. BATES, and P. S. GOLDMAN-RAKIC, 1991. prefrontal cortex. Eur.J. Neurosci. ll(2):567-574.
Interhemispheric integration: I. Symmetry and conver- OWEN, A. M., R. G. MORRIS, B. J. SAHAKIAN, C. E. POLKEY, and
gence of the corticocortical connections of the left and the T. W. ROBBINS, 1996. Double dissociations of memory and
right principal sulcus (PS) and the left and the right sup- executive functions in a self-ordered working memory task
plementary motor area (SMA) in the rhesus monkey. Cereb. following frontal lobe excision, temporal lobe excisions or
Cortex 1:390-407. amygdalahippocampectomy in man. Brain 119:1597-1615.
McKEE, R. D., and L. R. SQUIRE, 1993. On the development PARDRIDGE, W., 1977. Regulation of amino acid availability to
of declarative memory. J. Exp. Psych.: Learn. Mem. and Cogn. the brain. In Nutrition and the Brain, R.J. Wurtman andJ.J.
19:397-404. Wurtman, eds. New York: Raven Press, pp. 141-204.
McLARDY, T., 1950. Thalamic projection to frontal cortex in PARDRIDGE, W. M., and W. H. OLDENDORF, 1977. Transport of
man.J Neurol. Neurosurg. Psychiatry 13:198-202. metabolic substrates through the blood-brain barrier.J. Neu-
MEUNIER, M., J. BACHEVALIER, M. MISHKIN, and E. A. MUR- rochem. 28:5-12.
RAY, 1993. Effects on visual recognition of combined and PASSLER, P. A., W. ISAAC, and G. W. HYND, 1985. Neuropsy-
separate ablations of the entorhinal and perirhinal cortex chological development of behavior attributed to frontal
in rhesus monkeys.J Neurosci. 13:5418-5432. lobe functioning in children. Dev. Neuropsychol. 4:349-370.
MIDDLETON, F. A., and P. L. STRICK, 1994. Anatomical evi- PELLI, D. G., J. G. ROBSON, and A. J. WILKINS, 1988. The design
dence for cerebellar and basal ganglia involvement in of a new letter chart for measuring contrast sensitivity. Clin.
higher cognitive function. Science 266:458-461. Vis. Sci. 2:187-199.
MIDDLETON, F. A., and P. L. STRICK, 1997. Cerebellar output PENNINGTON, B. E, and S. OZONOFF, 1996. Executive function
channels. In The Cerebellum and Cognition, J. D. Schmah- and developmental psychopathology.J. Child Psychol. Psychi-
mann, ed. San Diego: Academic Press, pp. 61-82. atry 37:51-87.
470 COGNITION
SAWAGUCHI, T, M. MATSUMURA, and K. KUBOTA, 1990. Effects DE SONNEVILLE, L. N. J., E. SCHMIDT, U. MICHEL, and U. BAT-
of dopamine antagonists on neuronal activity related to a ZLER, 1990. Preliminary neuropsychological test results. Eur.
delayed response task in monkey prefrontal cortex. /. Neu- J Pediatr. 149(suppl. 1): S39-S44.
rophysiol. 63:1401-1412. SOPHIAN, C., and H. M. WELLMAN, 1983. Selective information
SCHMAHMANN, J. D., and D. N. PANDYA, 1995. Prefrontal cor- use and perseveration in the search behavior of infants and
tex projections to the basilar pons in rhesus monkey: Im- young children.J Exp. Child Psychol. 35:369-390.
plications for the cerebellar contribution to higher SOWELL, E. R., P. M. THOMPSON, C. J. HOLMES, T. L. JERNIGAN,
function. Neurosci. Lett. 199:175-178. and A. W. TOGA, 1999. In vivo evidence for post-adolescent
SCHNEIDER, J. S., and C. J. KOVELOWSKI II, 1990. Chronic ex- brain maturation in frontal and striatal regions. Nature Neu-
posure to low doses of MPTP. I. Cognitive deficits in motor rosci. 2:859-861.
asymptomatic monkeys. Brain Res. 519:122-128. SQUIRE, L. R., S. ZOLA-MORGAN, and K. S. CHEN, 1988. Human
SCHNEIDER, J. S., and D. P. ROELTGEN, 1993. Delayed match- amnesia and animal models of amnesia: Performance of
ing-to-sample, object retrieval, and discrimination reversal amnesic patients on tests designed for the monkey. Behav.
deficits in chronic low dose MPTP-treated monkeys. Brain Neurosci. 102:210-221.
Res. 615:351-354. STEMERDINK, B. A., J. J. VAN DER MEERE, M. W. VAN DER Mo-
SCHWARTZ, M. L., and P. S. GOLDMAN-RAKIC, 1984. Callosal LEN, A. F. KALVERBOER, M. M. T. HENDRIKX, J. HUISMAN, L.
and intrahemispheric connectivity of the prefrontal associ- W. A. VAN DER SCHOT, F. M. E. SLIJPER, F. J. VAN SPRONSEN,
ation cortex in rhesus monkey: Relation between intrapar- and P. H. VERKERK, 1995. Information processing in patients
ietal and principal sulcal cortex. J. Comp. Neurol. 226:403- with early and continuously-treated phenylketonuria. Eur.J.
420. Pediatr. 154:739-746.
SELEMON, L. D., and P. S. GOLDMAN-RAKIC, 1985. Longitudinal STUSS, D. T, and D. F. BENSON, 1986. The Frontal Lobes. New
topography and interdigitation of corticostriatal projections York: Raven Press.
in the rhesus monkey.J Neurosci. 5:776-794. STUSS, D. T, and D. F. BENSON, 1987. The frontal lobes and
SELEMON, L. D., and P. S. GOLDMAN-RAKIC, 1988. Common control of cognition and memory. In The Frontal Lobes Re-
cortical and subcortical targets of the dorsolateral prefron- visited, E. Perecman, ed. New York: IRBN Press, pp. 141-
tal and posterior parietal cortices in the rhesus monkey: Evi- 158.
dence for a distributed neural network subserving spatially TAM, S. Y, J. D. ELSWORTH, C. W. BRADBERRY, and R. H. ROTH,
guided behavior.J. Neurosci. 8:4049-4068. 1990. Mesocortical dopamine neurons: High basal firing
SELTZER, B., and D. N. PANDYA, 1989. Frontal lobe connec- frequency predicts tyrosine dependence of dopamine syn-
tions of the superior temporal sulcus in the rhesus monkey. thesis./ Neural Transmission 81:97-110.
J Comp. Neurol. 281:97-113. TAYLOR, J. R.,J. D. ELSWORTH, R. H. ROTH,J. R. SLADEK,JR.,
SHALLICE, T., 1988. From Neuropsychology to Mental Structure. and D. E. REDMOND, JR., 1990a. Cognitive and motor deficits
Cambridge: Cambridge University Press. in the acquisition of an object retrieval detour task in MPTP-
SHEDLOVSKY, A.,J. D. MCDONALD, D. SYMULA, and W. F. DOVE, treated monkeys. Brain 113:617-637.
1993. Mouse models of human phenylketonuria. Genetics TAYLOR, J. R., R. H. ROTH, J. R. SLADEK, JR., and D. E. RED-
134:1205-1210. MOND, JR., 1990b. Cognitive and motor deficits in the per-
SIMON, H., B. SCATTON, and M. LEMOAL, 1980. Dopaminergic formance of the object retrieval detour task in monkeys
A10 neurons are involved in cognitive functions. Nature (Cercopithecus aethiops sabaeus) treated with MPTP: Long-
286:150-151. term performance and effect of transparency of the barrier.
SIMON, R. J., and K. BERBAUM, 1990. Effect of conflicting cues Behav. Neurosci. 104:564-576.
on information processing: the 'Stroop Effect' vs. the THATCHER, R. W., R. A. WALKER, and S. GIUDICE, 1987. Hu-
man cerebral hemispheres develop at different rates and
'Simon Effect.' Acta Psychol. 73:159-170.
ages. Science 230:1110-1113.
SIWEK, D. F., and D. F. PANDYA, 1991. Prefrontal projections
THIERRY, A. M., J. P. TASSIN, A. BLANC, L. STINUS, B. SCATTON,
to the mediodorsal nucleus of the thalamus in the rhesus
and J. GLOWINSKI, 1977. Discovery of the mesocortical do-
monkey.J Comp. Neurol. 312:509-524.
paminergic system: Some pharmacological and functional
SKRANDIES, W., and I. GOTTLOB, 1986. Alterations of visual
characteristics. Adv. Biomed. Psychopharmacol. 16:5-12.
contrast sensitivity in Parkinson's disease. Hum. Neurobiol.
TOBIAS, T. J., 1975. Afferents to prefrontal cortex from the
5:255-259.
thalamic mediodorsal nucleus in the rhesus monkey. Brain
SMITH, E. E., J. JONIDES, C. MARSHUETZ, and R. A. KOEPPE,
Res. 83:191-212.
1998. Components of verbal working memory: Evidence TORNQUIST, P., and A. ALM, 1986. Carrier-mediated transport
from neuroimaging. Proc. Natl. Acad. Sci. 95(3):876-882. of amino acids through the blood-retinal and blood-brain
SMITH, E. E., and J. JONIDES, 1999. Storage and executive pro- barriers. Graefe's Arch. Clin. Exp. Ophthalmol. 224:21-25.
cesses in the frontal lobes. Science 283:1657-1661. TOURIAN, A. Y, and J. B. SIDBURY, 1978. Phenylketonuria. In
SMITH, L, and M. BEASLEY, 1989. Intelligence and behaviour The Metabolic Basis of Inherited Disease, J. D. Stanbury, J. B.
in children with early treated phenylketonuria. Eur. J. Clin. Wyngaarden, and D. Fredrickson, eds. New York: McGraw-
Nutrit. 43:1-5. Hill, pp. 240-255.
SMITH, L. B., E. THELEN, R. TITZER, and D. McLiN, 1999. TWETEN, S., M. WALL, and B. D. SCHWARTZ, 1990. A compar-
Knowing in the context of acting: The task dynamics of the ison of three clinical methods of spatial contrast-sensitivity
A-not-B error. Psychol. Rev. 106:235-260. testing in normal subjects. Graefe's Arch. Clin. Exp. Ophthal-
SMITH, M. L., P. KLIM, E. MALLOZZI, and W. B. HANLEY, 1996. mol. 228:24-27.
A test of the frontal-specificity hypothesis in the cognitive VOGT, B. A., and D. N. PANDYA, 1987. Cingulate cortex of the
performance of adults with phenylketonuria. Dev. Neuro- rhesus monkey: II. Cortical afferents. J Comp. Neurol.
psychol 12:327-341. 262:271-289.
472 COGNITION
30 Age-Related Changes in Working
Memory and Frontal Lobe Function:
A Review
MONICA FABIANI
AND EMILY WEE
ABSTRACT The frontal lobes are considered to be an essen- Working memory refers to a system for temporarily
tial component of a circuit that supports working memory holding and manipulating information (Baddeley,
function. This chapter examines several aspects of the changes 1986; Cowan, 1995). As such, working memory is likely
that occur in frontal lobe function and working memory
during normal aging. We review data from cognitive, neuro- to contribute to a wide variety of tasks, including learn-
psychological, and neuroimaging research that suggest the ing and comprehension, as well as the planning and
existence of diminished frontal lobe function in normal aging, maintenance of goals and actions and the monitoring
leading to diminished performance in a wide variety of tasks and supervision of one's own behavior. Given this view
involving working memory. These include age-related changes of working memory, it is not surprising that age-related
in the effects of memory load, in the processing of different
types of stimuli, in the susceptibility to distractors, and in the changes in this function may be far-reaching and have
ability to inhibit prepotent but unwanted responses. In addi- effects on a variety of laboratory and real-life tasks.
tion, we discuss dissociations between performance in source Evidence has accumulated that suggests that the fron-
memory tasks and recognition memory tasks, as well as meth- tal lobes, and particularly the dorsolateral prefrontal
odological and theoretical issues pertaining to this research. cortex (Brodmann area 46), are an essential compo-
nent of a circuit that sustains working memory function
Introduction and background (Fuster, 1973; Goldman-Rakic, 1987; 1992). Damage to
the frontal lobes does not necessarily lead to memory
WORKING MEMORY, THE FRONTAL LOBES, AND AGING loss per se, but it may lead to impairments on tasks re-
One of the most common complaints of normal older quiring the organization of external cues and the ini-
adults is a reduction in their memory capacity. Research tiation and execution of strategic behavior (Knight,
on metamemory suggests that the elderly commonly ex- Grabowecky, and Scabini, 1995; Milner, 1995; Milner
pect memory failures, and sometimes exaggerate their and Petrides, 1984; Moscovitch and Winocur, 1992;
likelihood of occurrence and extent (Hess and Pullen, Shimamura, 1995; Stuss, Eskes, and Foster, 1994). In
1996; Parr and Siegert, 1993). However, an examina- fact, patients with frontal lobe lesions show minor per-
tion of the literature concerning age-related memory formance decrements in tasks involving "item memory"
changes suggests that some aspects of memory (such as (memory for the items or events themselves), but are
direct vocabulary knowledge, priming, and recognition often impaired in tasks requiring "source memory"
memory1) appear to remain fairly stable throughout (memory for contextual details referring to the remem-
the adult life span, whereas other aspects of memory
bered items), and may lack flexibility when confronted
are affected by increasing age (for reviews, see Kausler,
with changing task demands. A typical task in which
1991; Light, 1996; Woodruff-Pak, 1997). Among the
frontal-lesioned patients show diminished performance
memory functions that appear to be most affected by
is the Wisconsin Card Sorting Test (WCST, Heaton,
normal aging is "working memory," especially in situa-
1981; Milner, 1963). In this task patients with frontal
tions that require the recall and "on-line" manipulation
lesions have difficulty adjusting their response strategies
of contextual details, coupled with a fair amount of flex-
ibility and strategic adjustments (Moscovitch and Win- on the basis of feedback from the experimenter, and
ocur, 1992; Parkin and Walter, 1991; West, 1996). tend to persevere with responses that were originally
correct but are no longer valid.
MONICA FABIANI AND EMILY WEE Department of Psychology, Recent evidence suggests that altered frontal func-
University of Missouri, Columbia, Missouri. tion may frequently accompany not only pathological,
FABIANI AND WEE: WORKING MEMORY AND FRONTAL LOBE FUNCTION 473
but also normal aging (West, 1996). Several reports sug- processed in working memory; and (5) evidence per-
gest that older adults' performance is diminished with taining to memory errors for contextual information
respect to that of young adults in neuropsychological (source memory). In closing, we briefly summarize
tests such as the WCST (e.g., Fabiani and Friedman, some theoretical issues and future directions in this re-
1995; Fabiani, Friedman, and Cheng, 1998; Parkin and search domain.
Walter, 1991), in source memory tasks (Fabiani and
Friedman, 1997; Schacter et al, 1993; Trott et al., 1999), Methodological issues
and in other tasks requiring the monitoring of strategies
and/or the inhibition of prepotent responses (e.g., con- "OLD" AGE AND INDIVIDUAL DIFFERENCES The defini-
flict paradigms such as the Stroop tasks; Spieler, Balota, tion of "old" age is a source of controversy in aging re-
and Faust, 1996). There is also evidence that age-related search because there is no exact correspondence
changes in brain anatomy (Haug and Eggers, 1981; for between chronological age and the extent of age-re-
a review, see West, 1996) and function (e.g., Gur et al., lated cognitive change. In an effort to increase their
1987) are more extensive in the frontal lobes than in understanding of the aging process, researchers some-
other areas of the brain, such as sensory and motor ar- times divide older subjects into groups characterized by
eas. These data are consistent with the proposal that more restricted age ranges, such as young-old (ages 60-
many of the cognitive changes observed in later life may 74) and old-old (ages 74 and over; e.g., Parkin and Java,
be due to reduced frontal function (Moscovitch and 1999). However, there is no universally accepted de-
Winocur, 1992; Parkin and Walter, 1991; Stuss et al., marcation between middle age and older adulthood.
1996; West, 1996). The studies reviewed in this chapter mostly involve two
subject groups: young adults (most often college stu-
LAYOUT OF THE REVIEW In summary, it is hypothesized dents below the age of 30) and older adults (typically
that age-related problems in working memory may be 60 years of age or older).
associated, at least in part, with diminished frontal func- Individual differences tend to increase with age. Life-
tion. In this chapter, we present converging evidence of style, general health, and other factors greatly contrib-
(1) changes in the working memory function of nor- ute to how much an individual actually ages with respect
mally aging adults and (2) age-related changes in fron- to other members of his/her cohort. For example, a
tal lobe function in the same population. As will number of investigators have looked at the role of physi-
become apparent in the course of this review, the exact cal fitness on cognitive function in aging (e.g., Bashore,
mapping and degree of overlap between working mem- 1990; Bashore and Goddard, 1993; Kramer et al., 1999).
ory and frontal lobe function remain to be completely Older adults who exercise regularly tend to perform
clarified. better in cognitive tasks than those who do not, espe-
Although by no means exhaustive, this review in- cially if their exercise programs involve aerobic activi-
cludes representative examples from the cognitive, ties. These effects may be due to the benefits of
neuropsychological, and brain imaging literatures, in- enhanced overall health, or to more indirect effects
cluding event-related brain potentials (ERPs), positron such as an increased blood circulation to the brain.
emission tomography (PET), and functional magnetic Of course, the role of individual differences in aging
resonance imaging (fMRI). We begin the review with a begs the question of whether engaging in frequent and
discussion of methodological issues in aging research, complex mental activity may delay or counteract some
including the definition of "old" age and the role of of the impairments observed in normal cognitive aging
individual differences, as well as some of the problems (the "use-it-or-lose-it" hypothesis). Snowdon and col-
associated with the interpretation of brain imaging and leagues (1996, 1997) initiated a longitudinal study in-
neuropsychological data. These are issues that should volving a group of nuns who joined the Sisters of Notre
be considered when examining literature on aging, and Dame in the 1920s, and were tested at regular intervals
may guide the interpretation of the remainder of the throughout old age. Data published in 1996 (Snowdon
chapter. We then describe (1) possible age-related et al., 1996) suggested the possibility of predicting the
changes in sensory memory; (2) the effects of working onset of Alzheimer's disease in old age by examining
memory load, both in terms of the number of items to the writing complexity of the same women at age 20.
be held in memory and in terms of the number of However, although very intriguing in principle, this
operations to be performed on the items; (3) evidence study has been criticized for the subjectivity of the writ-
of increased distractibility and decreased ability to in- ing analysis and for the small number of subjects in-
hibit prepotent, but incorrect, responses; (4) the dif- volved. A more general issue is that it is difficult, in
ferential effects of the type of task and material to be studies involving individual differences, to distinguish
474 COGNITION
between causes and effects (i.e., does mental activity or ERP effects. Although this approach is potentially
"protect" the brain from aging, or is a more "capable" useful, it can be criticized because many of these tests
brain more resistant to aging?). In summary, the role are not sufficiently specific to the brain region whose
of a more active cognitive life in preventing some of the functions they are designed to assess (Eslinger and Grat-
adverse effects of aging on cognition is far from clear tan, 1993; Mountain and Snow, 1993). In addition, it
and difficult to address even in the context of longitu- should also be kept in mind that, although the effects
dinal studies because a number of alternative explana- of normal aging may be accompanied by some cell at-
tions are usually possible. rophy and neurotransmitter depletion, they are not
characterized by the extensive cell death typical of brain
THE INTERPRETATION OF BRAIN IMAGING DATA FROM lesions. Thus, the analogy between normal aging pro-
OLDER ADULTS Technological advances in the last two cesses and the effects of frontal lesions should not be
decades have greatly enhanced the tools available to in- overemphasized. For instance, Fabiani and Friedman
vestigators to examine the relationships between behav- (1995) observed an increased frontal P3 in aging,
ior and underlying brain function in older adults whereas Knight (1984) reported decreased frontal P3
(Fabiani, Gratton, and Coles, 2000; Gratton and Fabi- in patients with frontal lesions. This underscores the
ani, 1998; Kutas and Federmeier, 1998; Toga and difference that may exist between a full-fledged lesion
Mazziotta, 1996). However, several caveats in the inter- and the microscopic, though diffuse, alterations that
pretation of brain imaging data should be considered, may occur in normal aging.
especially with respect to their use in aging research.
For example, a number of the ERP studies described in Sensory memory
this chapter were designed on the basis of a priori hy-
potheses of frontal effects and did report marked age- Although this chapter mainly focuses on age-related
related changes of brain activity at frontal electrode changes in working memory and frontal function, in
sites. However, it is generally not possible to make in- this section we briefly review some recent studies sug-
ferences about the brain origin of electrical potentials gesting that early and largely preattentive stimulus pro-
only on the basis of scalp-recorded data (Fabiani, Grat- cessing may also show changes during normal aging.
ton, and Coles, 2000), although some inferences can be This early processing can be roughly identified with sen-
made on the basis of converging evidence from intra- sory memory—a memory buffer that can briefly hold
cranial (Fernandez et al., 1999) and brain injury large amounts of information before some of it enters
(Knight, 1984) studies. attention and working memory (Naatanen, 1992). It
Although fMRI may provide more precise localiza- may be hypothesized that information from the sensory
tion data than ERP recordings, there are a number of systems is continuously "fed" into working memory
limitations in the application of fMRI to the study of from sensory memory (Broadbent, 1957). Regardless of
aging. Some of these limitations are practical (such as the theoretical framework adopted, it follows that if the
the need for the subjects to be motionless for extended capacity and/or duration of sensory memory is in some
periods and the necessity of avoiding glasses or other way degraded in aging, this may, in turn, influence sub-
prostheses containing metal), and can probably be sequent working memory processing.
solved for a number of older participants. There are, One component of the ERP, the mismatch negativity
however, some more fundamental issues concerning (MMN; Naatanen and Alho, 1995; Ritter et al., 1995),
the degree to which the coupling between neuronal has been used as an index of automatic deviance detec-
function and blood flow remains the same in old age, tion in auditory sensory memory. In fact, the MMN in-
and whether it remains consistent for different areas of creases in amplitude in response to deviant tones (e.g.,
the brain (D'Esposito et al., 1999). These problems may high-pitched) occasionally intermixed within a stream
complicate the interpretation of hemodynamic imaging of homogeneous (standard) tones (e.g., low-pitched),
data in the study of age-related cognitive changes. Fur- even when the subject is ignoring the tones and reading
ther understanding of the neuronal-hemodynamic a book (passive oddball paradigm). This amplitude in-
coupling in aging may help address these issues. crease to the rarely occurring deviant stimuli implies
that the memory of the standard stimuli is maintained
THE USE OF NEUROPSYCHOLOGICAL TESTING IN AGING automatically.
RESEARCH Neuropsychological tests are typically de- The average ERP waveform elicited by the standard
signed to characterize patients with specific brain le- tones is subtracted from that of the deviant tones to
sions. In aging research, they can also be used to help yield the MMN, which, as its name indicates, is negative
characterize the underlying brain source of behavioral in polarity. The MMN has an onset latency as short as
FABIANI AND WEE: WORKING MEMORY AND FRONTAL LOBE FUNCTION 475
50 ms, a peak latency of 100-200 ins, and is most likely cessing leading to tone discrimination may be intact in
generated in primary auditory cortex and/or immedi- older subjects, at least at short intervals, but that the
ately adjacent areas, with at least one additional contri- information required to determine whether a stimulus
bution from the frontal lobes (Alho, Huotilainen, and is deviant may decay at a faster rate in the elderly than
Naatanen, 1995; Alho et al., 1994; Javitt et al., 1994; it does in young adults. These discrepant findings may
Rinneetal., 1999). be due to methodological differences among studies.
Because the MMN is generated in response to any For example, a fixed tone volume was used in some
discriminable change between the standard and deviant studies (Czigler, Csibra, and Csontos, 1992), and ad-
tones even when the subject is not attending to the stim- justed in others (Woods, 1992; Pekkonen et al., 1996).
uli, it is likely to reflect the automatic and preattentive Thus, peripheral hearing loss may have contributed to
processing of deviant features. In addition, because the the MMN differences observed between young and old
MMN is not elicited after long (10 s or more) intersti- subjects. Similarly, eye movement artifacts may have led
mulus intervals, it is hypothesized that it may be based to increased trial rejection in the elderly with respect to
on a type of memory that is transient in nature. the young. Finally, some of the studies used paradigms
A few studies have investigated the effects of aging on that involved some degree of attention switching
the MMN. Some of these studies have reported a de- (Woods, 1992), and therefore may have imposed an ad-
creased MMN amplitude in the elderly, irrespective of ditional memory load, whereas others did not (Pekko-
the length of the interval interposed between standard nen et al., 1996). In conclusion, although it is still
and deviant stimuli (Czigler, Csibra, and Csontos, 1992; unclear whether or not there is decreased automatic
Gaeta et al., 1998; Kazmerski, Friedman, and Ritter, deviance detection in aging in general, or whether this
1997; Woods, 1992; see figure 30.1) and therefore sug- decrease is limited to conditions in which relatively long
gest a decreased sensitivity to deviance as a function of (e.g., 4 s or more) ISIs separate deviants from standard
age. In contrast, others have reported a normal-ampli- stimuli, there appears to be some amount of change in
tude MMN in the elderly when short ISIs are used, but early auditory sensory memory processing in normal
a reduced MMN when long intervals are interposed be- aging.
tween stimuli (Pekkonen et al., 1996; see figure 30.2).
The latter finding suggests that the early sensory pro- Effects of working memory load
476 COGNITION
FIGURE 30.2 Results from a study by Pekkonen and col- subjects at long (bottom panel) but not at short ISIs (top
leagues (1996) showing diminished MMN amplitude in older panel). (Reproduced with permission from Taylor & Francis.)
FABIANI AND WEE: WORKING MEMORY AND FRONTAL LOBE FUNCTION 477
Investigators have used ERPs to specify further the memory. For example, the load was defined as low when
causes of the age-related RT delay that occurs with in- the two study stimuli were identical on both dimensions
creased memory load in Sternberg's task. The most (e.g., two squares above fixation), and therefore only
common approach has been to use the latency of the two features needed to be kept in working memory. The
P300 (also called P3 or P3b; Sutton et al., 1965; Fabiani, load was considered medium when the two stimuli
Gratton, and Coles, 2000) elicited by the probe stimuli matched on only one dimension (e.g., two squares pre-
as an index of decision processes independent of re- sented in different positions). Finally, load was consid-
sponse selection processes (McCarthy and Donchin, ered high when the two stimuli differed on both
1981). In other words, because the P300 is elicited by dimensions (e.g., a diamond and a square presented in
target stimuli (such as the probes in a Sternberg's task; different positions), requiring the maintenance of four
Johnson and Donchin, 1978), it can be used as an ad- separate features and their correct pairings. Note that
ditional measure of processing time. subjects did not know which dimension they needed to
Ford and colleagues (1979) reported that, in Stern- use for the test, and therefore had to encode all relevant
berg's task, the slope of the increases in P3 latency as a information. As in Sternberg's task, the RT at test in-
function of set size was less than that for the RT in creased as a function of load, especially for older adults.
young adults. In a subsequent study, Ford and col- Although there were small P300 effects as a function of
leagues (1982) reported that, in older adults, the la- load, the most significant ERP changes consisted of var-
tency of P3 was unaffected by set size or response type, iations in the amplitude of a frontally distributed slow
possibly suggesting strategic differences between young wave elicited by the second study stimulus, which be-
and older adults. Strayer, Wickens, and Braune (1987) came progressively more negative as load increased
suggested that perceptual encoding, response criterion from low to high. These slow-wave effects at encoding
adjustment, and response execution were affected by were more pronounced in older adults than in young
aging more than memory search speed per se (see also adults. This slow-wave component may reflect encoding
Strayer and Kramer, 1994; for a meta-analytic study, see efforts, and, if so, these data may indicate that older
Bashore, Osman, and Heffley, 1989). Thus, these data adults require more effort to encode and/or maintain
suggest that (1) different phenomena may contribute information in working memory as load increases.
to the age-related changes in the RT/item slope in
Sternberg's task; (2) aging produces differential effects, NUMBER or OPERATIONS To BE PERFORMED IN WORK-
depending on the aspects of processing involved (see ING MEMORY The effects of working memory load can
also Chao and Knight, 1997a); and (3) response selec- also be assessed with respect to the number of opera-
tion and execution may play an important role in the tions (or amount of processing) that subjects are re-
slowing of RT in aging, possibly reflecting increased dis- quired to perform "on-line." The WCST (Parkin and
tractibility and consequent response conflict. Walter, 1991) and the continuous TV-back task (Jonides
A different approach to investigating the effects of et al., 1997) are both tasks in which the subject has to
working memory load was taken recently by Wee and hold several pieces of information in memory as well as
Fabiani (1999a,b). They used a modified 2-back task in complex instructions on how to handle each new item.
which encoding and retrieval processes could be sepa- For example, Parkin and Walter (1991) reported that,
rated more easily than in the traditional continuous in older adults, performance on the WCST correlates
TV-back task (Jonides et al., 1997). In one of the exper- with performance in the Brown-Peterson task. In the TV-
imental conditions, subjects were asked to encode two back task, Jonides and colleagues (1997), using PET,
successive stimuli, each varying on two dimensions showed progressively poorer performance and in-
(shape: square or diamond; position: above or below fix- creased activity in frontal areas as the working memory
ation). Following stimulus encoding, the subjects were load was increased by varying the value of N in young
given a cue indicating which of the two dimensions they adults (see also Braver et al., 1997). In the modified 2-
should use in formulating a response. The cue was fol- back task described earlier, older adults' performance
lowed by two test stimuli presented one after the other. at test decreased with load more than the performance
For each of these test stimuli, subjects had to indicate of young adults (see Wee and Fabiani, 1999a).
whether or not it matched the stimulus presented in These data suggest that the age-related effects of load
position 2-back with respect to the dimension indicated are increased further if older subjects have to maintain
by the cue. The sequence then ended and a new trial a number of items in memory while performing a task
started again with the study stimuli. In this task, memory that involves complex instructions and/or frequent
load during encoding was defined as the number of strategic adjustments and flexibility. In fact, data from
separate stimulus features that had to be maintained in Kramer, Hahn, and Gopher (1999) suggest that, with
478 COGNITION
practice in a task-switching paradigm, older adults im- relevant stimuli (Rabbitt, 1965). It has been also
prove to the level of young adults under low load con- hypothesized that the dorsolateral prefrontal cortex
ditions. However, under conditions of high memory plays a crucial role in a system aimed at inhibiting pre-
load, older adults are maximally affected by task switch- potent but unwanted responses to irrelevant stimuli
ing and are not able to benefit from practice. (Chao and Knight, 1997b; Cohen, Braver, and O'Reilly,
1996; Knight, Grabowecky, and Scabini, 1995). In this
Effects of type of material to be processed section we examine two different lines of evidence for
in working memory this hypothesis. The first is an examination of results
obtained by recording ERPs in two variants of the odd-
An important variable in the analysis of working mem- ball paradigm in young and older adults, and in patients
ory function is the type of material to be held in mem- with frontal lobe lesions. The second is an examination
ory. Both fMRI (McCarthy et al., 1994) and PET studies of data obtained in paradigms in which two stimulus
(Jonides et al., 1993; Smith et al., 1995) have reported dimensions call for opposite responses (only one of
an increased activation in the middle frontal gyrus which is correct), thereby creating a conflict in the pro-
(Brodmann area 46) during the performance of spatial cessing of information.
working memory tasks in young adults. This activation
is predominant in the right hemisphere, and it is part AGE-RELATED CHANGES IN THE AMPLITUDE AND SCALP
of a circuit that involves occipital, parietal, and prefron- DISTRIBUTION OF P300 The P300 component of the
tal areas. Verbal working memory tasks appear to acti- ERPs is typically recorded in an active oddball para-
vate regions in the left hemisphere (Smith, Jonides, and digm, in which two types of stimuli, differing in proba-
Koeppe, 1996, 1998), whereas not much prefrontal ac- bility of occurrence, are presented in a Bernoulli
tivity has been reported for object working memory sequence, and the subjects are instructed to respond to
tasks (Smith et al., 1995). or count the rare (target) stimuli. It has been suggested
Only a handful of imaging data are currently avail- that the P300 is involved in some aspects of working
able for older subjects. Jonides and colleagues (2000) memory processes (Donchin, 1981).
reported that, in a verbal working memory task in which Several age-related changes in the P300 recorded in
inhibitory activity was required, older adults (unlike the oddball paradigm have been reported, including an
young adults) failed to show increased activation in pre- increased latency and decreased amplitude as a func-
frontal cortex. In addition, whereas in young adults tion of age (Ford and Pfefferbaum, 1991; Polich, 1991,
there appears to be a lateralization of working memory 1996; Polich, Howard, and Starr, 1985). The age-related
function depending on the type of material used (ver- amplitude decrease is not uniformly distributed across
bal vs. spatial), either bilateral activation or the opposite the scalp. The P300 is diminished in amplitude at pos-
pattern of activation is observed in older adults (see terior electrodes (where it is typically largest in young
figure 30.3 [see color plate 20]; Gabrieli, 1999; Reuter- adults), whereas it does not change much at frontal
Lorenz, 1999; Reuter-Lorenz et al., 2000). electrode sites, thus resulting in an apparent age-related
ERP data obtained in young subjects suggest that change in scalp distribution (see figure 30.4; Fabiani
working memory function may be indexed by negative and Friedman, 1995; Fabiani, Friedman, and Cheng,
slow waves (Mecklinger, 1998; Ruchkin et al., 1995). 1998; Friedman, Kazmerski, and Fabiani, 1997).
Wee and Fabiani (1999a,b) found a reduction in This age-related change in the scalp distribution of the
the frontal negative activity between young and older P300 has little apparent correspondence in behavioral
adults, and more bilateral activity in older than young changes: In most cases the oddball task is extremely easy,
adults when comparing spatial and object memory. and performance is asymptotic in both young and older
Taken as a whole, these blood flow and ERP data sug- adults. Furthermore, RT typically is not slower in older
gest that older adults may either need to recruit more adults in this task. However, hypotheses about the pos-
brain areas than do young adults to assist in processing sible meaning of these changes can be derived by an
(bilateral vs. lateralized effects), or may be invoking examination of findings obtained in a variant of the
processes that are inappropriate for the task at hand. oddball paradigm, the "novelty" oddball task.
Decreased inhibition of unwanted responses THE P300 IN THE NOVELTY ODDBALL TASK As in the
and increased susceptibility to distractors active oddball task, in the novelty oddball task subjects
are presented with a series of stimuli and asked to re-
For many years it has been known that it is often dif- spond to the target (rare) stimulus. However, in addi-
ficult for older adults to inhibit the processing of ir- tion to target and standard stimuli, novel stimuli are
FABIANI AND WEE: WORKING MEMORY AND FRONTAL LOBE FUNCTION 479
FIGURE 30.3 Data from Reuter-Lorenz (1999), showing bi- young subjects. (Reproduced with permission from Taylor &
lateral frontal activation in the older subjects during a verbal Francis.)
working memory (WM) task, and left activation only in the
also intermittently presented in the series. Much of the also Friedman, Simpson, and Fabiani, 1999). In other
research in this paradigm has been done within the au- words, it could be hypothesized that the scalp-recorded
ditory domain (but see Courchesne, Hillyard, and Gal- P300 represents the summation of the activity of neural
ambos, 1975; Yamaguchi and Knight, 1991a,b), and the generators in a circuit encompassing frontal, parietal
novel stimuli are unexpected nontonal sounds (e.g., a and temporal areas (Knight, 1997; Knight et al., 1989;
dog bark) rarely presented within a series of tones Yamaguchi and Knight, 1991a,b; see also Goldman-
(Fabiani and Friedman, 1995; Fabiani, Friedman, and Rakic, 1992, for a similar circuit in the monkey). Scalp
Cheng, 1998; Fabiani et al., 1996; Friedman, Simpson distribution changes observed in lesioned patients and
and Hamberger, 1993; Knight, 1987). It is hypothesized older adults may represent a change in the relative
that the rarely occurring novel stimuli may capture at- "weight" of the various generators within this circuit,
tention, and may therefore elicit an orienting response which, in turn, may reflect functional changes.
(which is not elicited by the rare but repeated targets). Using this paradigm and theoretical framework in
However, because no response is assigned to these stim- the context of normal aging research, Fabiani and
uli, it is also hypothesized that they may elicit inhibitory Friedman (1995) found that both young and older
processes to suppress response execution (Knight and adults showed increased frontal activity in response to
Nakada, 1998). nonrepeated novel sounds as compared to the P300 ac-
Using this paradigm in control subjects and patients tivity elicited by target tones (see figures 30.4 and 30.5).
with frontal lesions, Knight (1984) reported that, in Although the behavior of young and older adults did
normal control subjects, the novel sounds elicited a not differ much in this paradigm, there was a trend
P300-like activity (called novelty P3a or novelty P3) for the older adults to produce more false alarms in
characterized by increased amplitude at frontal elec- response to the novel sounds (for which the motor re-
trodes in comparison to the P300 elicited by rare target sponse should be inhibited), at least early on in the
stimuli. However, this increased activity in response to session. In addition, older adults showed frontally ori-
the novel stimuli was absent in patients with lesions ented P300 activity in response to the rare but repeated
to the dorsolateral frontal cortex. This lends support to target tones, whereas young adults had a typical
the hypothesis that the anterior aspect of the novelty- parietal-maximum P3 in response to targets. Finally,
PS activity may be generated in the frontal lobes (see when the P3 response to target stimuli was examined
480 COGNITION
FIGURE 30.4 Data from Fabiani and Friedman (1995) show- in aging. Positive potentials are plotted downward. (Reprinted
ing changes in P3 amplitude, latency and scalp distribution with permission of Cambridge University Press.)
over time, starting with the practice block, it was appar- the target P3 observed in young subjects) were much
ent that young adults showed a frontally oriented P3 in less impaired in tests of frontal function such as the
response to the initial presentations of the target stim- WCST than were those older adults whose P3 was largest
uli. This response faded out with time on task and was at the frontal electrodes (and thus more similar to a
replaced by a parietal-maximum P3 (see figure 30.6). novelty P3). This relationship between brain activity
In older adults, the frontal focus of this component in and neuropsychological profile was specific to tests of
response to targets remained stable over time, regard- frontal lobe function, as the two groups of older sub-
less of repeated exposures to the target tone. This sug- jects did not differ on the basis of IQ or other "non-
gests that novel stimuli engage working memory/ frontal" variables. This suggests that the frontal portion
orienting processes in both young and older adults. of the P3 may represent the engagement of a working
However, with aging, these processes may also be en- memory/orienting system that may normally be sup-
gaged inappropriately in response to repeated stimuli pressed with repeated stimulus presentation. The per-
(such as the oddball targets) that should no longer re- sistence of frontal activity in response to repeated
quire them. This may occur because working memory targets in older subjects may therefore signal the inap-
templates decay faster in old age, and/or because older propriate elicitation of orienting responses. In turn, this
adults are more susceptible to distractors and increased may indicate the presence of working memory dysfunc-
workload. tion and be associated with increased distractibility.
In a follow-up study, Fabiani, Friedman, and Cheng
(1998) found that those older adults whose target P3 THE STROOP AND OTHER CONFLICT TASKS A more di-
was more parietally focused (and thus more similar to rect test of the hypothesis that older adults are more
FABIANI AND WEE: WORKING MEMORY AND FRONTAL LOBE FUNCTION 481
FIGURE 30.5 Data from Fabiani and Friedman (1995) show- maps at the peak of P3 for target and novel stimuli. (Reprinted
ing surface potential (SP) and current source density (CSD) with permission of Cambridge University Press.)
482 COGNITION
whereas only the contralateral motor cortex is activated This position is further corroborated by recent imaging
in trials with no conflict. data obtained with PET (Cabeza et al., 2000).
Spieler, Balota, and Faust (1996) found, using ex- Fabiani and Friedman (1997) compared the perfor-
Gaussian analyses, that healthy older adults exhibited a mance of young and older adults in a continuous task
disproportionate increase in Stroop interference with that involved both recognition and recency judgments.
respect to young adults. This is consistent with the hy- In this task, study trials and trials testing recency and
pothesis that additional processing time may be re- recognition memory were intermixed. The subjects'
quired to resolve conflict and produce the correct task was to indicate which of two items (either pictures
response. Similarly, Duchek and colleagues (1995) or words) was presented most recently. In recognition
found that the elderly showed deficits in the control of trials only one of the pictures had been studied earlier,
interfering information. This suggests that irrelevant in- making that the most recent by default. In recency test
formation may be processed to a greater extent by older trials, both pictures had been previously presented, so
adults, as compared to younger subjects, leading to the subjects had to remember the relative temporal
greater response competition. occurrence of the two in order to produce a correct
ERP data obtained in the Stroop task by Lavoie, response. This continuous paradigm has some charac-
Bherer, and Belleville (1999) showed that, whereas teristics that make it advantageous for studying changes
young adults had an enhanced P300 in response to con- in recency memory in normal aging. First, the subjects
flict trials, older adults did not (see also West and Alain, need not know whether they are performing a recency
2000). Similarly, Jonides and colleagues (2000) re- or recognition task, and therefore task differences per
ported that, in a task requiring inhibition of prepotent se are minimized. Second, by mixing study and test tri-
responses, older adults were slower and less accurate als and by building an appropriate sequence, the effects
than young adults, and, unlike young adults, did not of lag can be studied. Third, like all continuous para-
show increased frontal activation. These data suggest digms, this task imposes an additional load on working
that older adults may lack the resources necessary to memory, thereby allowing one to study encoding and
implement inhibitory processes. retrieval of contextual attributes in a situation of in-
creased workload, in which age-related effects should
Memory for contextual details be maximized.
Using this paradigm, Fabiani and Friedman (1997)
In recent years much attention has been paid to age- found that older subjects did not differ from young
related changes in the memory for contextual details adults on the recognition trials, whereas they per-
(source of the information, relative time of occurrence, formed more poorly than the young subjects on the
etc.). Behavioral studies comparing young and older recency trials (in fact, they performed at chance level
adults suggest that older adults, although able to indi- on these trials). In addition, for the old subjects (age
cate whether or not they have been presented with range 65-88) the performance on the recency task, but
the information (recognition judgments—item mem- not that on the recognition task, was correlated with
ory1) have increased difficulty when asked specific ques- performance on the Wisconsin Card Sorting Test and
tions about details of their episodic memory (source other tests of frontal function.
memory; Kausler, Lichty, and Davis, 1985; Kausler, Salt- In a follow-up study in which ERPs were recorded,
house, and Saults, 1988; Schacter et al., 1993). These Fabiani and colleagues (1999) found that there were
results are also corroborated by findings suggesting that clear individual differences among the elderly in the
recognition performance (which does not show sub- performance of this task. Older adults with higher levels
stantial age-related decrements) may be supported by of education and professional jobs were not impaired
different processes in young and older adults (Tulving, in their recency judgments, whereas older adults with
1985). In fact, when asked to judge whether a correctly lower levels of education and socioeconomic status per-
recognized word is "remembered" (i.e., details about formed at chance levels on the recency task. In addi-
the episodes are retrieved) or just "known" (i.e., accom- tion, highly educated older adults showed enhanced
panied by familiarity without detail), older adults tend negative slow wave activity at frontal electrodes in the
to give more "know" responses, whereas young adults performance of the recency task, which was not dis-
give more "remember" responses (Parkin and Walter, played by the older adults with lower educational levels
1992; for a similar dissociation, see Jennings andjacoby, or by the young adults. This suggests the existence
1997). These results bear similarities with data obtained of compensatory activity in those older adults who
in patients with frontal lobe lesions (Milner, 1995), sug- achieved high performance in the recency task.
gesting a possible frontal contribution to these tasks. Whether similar effects could be found in young adults
FABIANI AND WEE: WORKING MEMORY AND FRONTAL LOBE FUNCTION 483
remains unclear. This leaves open the issue of whether aging adults. This view may account for a number of the
these phenomena may be due only to education, or to memory findings in the aging literature.
the protective role that high levels of education might Several unresolved theoretical issues are also likely to
have on the effects of cognitive aging. inform the research framework presented in this chap-
Trott and colleagues (1997, 1999) compared young ter, together with further developments of imaging
and older adults in a task in which subjects were asked methods. For example, there is evidence that many of
to indicate in which of two lists a given noun had oc- the effects of cognitive aging can be accounted for by
curred. Compared with young adults, older adults had generalized slowing of function (Salthouse, 1996). It
greater source memory (study list) performance dec- has also been known for a number of years that the
rements. In addition, only the young produced a fron- alpha rhythm of the EEG slows down with age, probably
tal-maximal, late-onset old/new effect that differed as a because of diminished arousal to the cortex from the
function of subsequent source attribution (see also reticular formation (for a review, see Woodruff-Pak,
Senkfor and Van Petten, 1996, 1998). 1997). It will be important to determine how a gener-
The studies just described indicate that older adults alized slowing may interact with working memory func-
have more difficulty than young adults in acquiring tions. For instance, it is conceivable that a slower
and/or maintaining at least some of the contextual de- processor may require the maintenance of information
tails of the memory episode. One extreme form of for a longer period of time, or having more information
source memory error is the phenomenon of false mem- available at any one time. Both conditions may result in
ory, which involves recalling or recognizing an event an increased working memory load. If that happens,
that did not actually occur. This phenomenon has re- apparent working memory deficits may actually reflect
ceived much attention in recent years owing to the use slower processing abilities.
of a clever experimental paradigm that makes it possi- From the review presented in this chapter, it should
ble to produce it easily in a laboratory setting (the be evident that a number of tasks are expected to have
Deese-Roediger-McDermott, or DRM paradigm; Deese, a substantial contribution from the frontal lobes. How-
1959; Roediger and McDermott, 1995). In this para- ever, as the concept of working memory may not be
digm, subjects are presented with lists of words that all unitary, the role of frontal lobes and in particular of
relate to a nonstudied "lure" (e.g., the words slumber, dorsolateral prefrontal cortex may be more differenti-
night, bed, and snore are all related to sleep, which is not ated than currently hypothesized (Petrides et al., 1993;
presented at study). In this paradigm, even young sub- West, 1996).
jects produce a large number of false alarms to the non- In addition, some of the effects attributed to the fron-
studied lures when they are included at test as distractor tal lobes may result from slowed down or deficient trans-
items (the false memory effect, Roediger and Mc- fer of information among brain areas, as suggested by
Dermott, 1995). In older adults the false memory effect the review of the auditory sensory memory literature
is increased, owing to the increased number of false presented here. Methods such as diffusion tensor im-
alarms and the lower overall memory for studied items aging, which allows one to examine the intactness of
(true memory—Balota et al., 1999; Schacter, 1997; white matter fiber tracts, may be particularly useful in
Schacter, Israel, and Racine, 1999). Behaviorally, it has assessing age-related changes in the transfer of infor-
proved difficult to distinguish between true and false mation between brain regions (Shimony et al., 1999).
memory. In fact, attempts at finding blood flow or ERP Similarly, fast optical imaging (EROS, Gratton and Fa-
differences between these two types of items have also biani, 1998) may provide a useful bridge for the inte-
proved elusive (Johnson et al., 1997; Schacter et al., gration of imaging methods based on hemodynamics
1996, 1997), although recent ERP data in young adults and methods based on electrophysiology. Finally, the
suggest that some evidence of sensory signatures for ve- integration of existing methods is likely to afford a view
ridical but not false memory may be found (Fabiani, of brain function over the lifespan that is not possible
Stadler, and Wessels, 2000). with any single method in isolation (Fabiani, Gratton,
and Coles, 2000; Gratton and Fabiani, 1998; Kutas and
Federmeier, 1998; Liu, Belliveau, and Dale, 1998).
Summary and future directions
ACKNOWLEDGMENTS We would like to acknowledge the sup-
This chapter presented a brief and selective overview of
port of a grant from the Alzheimer's Disease and Related Dis-
research suggesting that age-related changes in frontal orders Program (ADRD) to Monica Fabiani. We also thank
lobe function may underlie changes in working mem- David Friedman and Gabriele Gratton for reviewing an earlier
ory capacity that are commonly observed in normally version of this manuscript.
484 COGNITION
NOTE COHEN, J. D., and D. SERVAN-SCHREIBER, 1992. Context, cor-
tex and dopamine: A connectionist approach to behavior
1. The lack of performance decrements in recognition tasks
and biology in schizophrenia. Psychol. Rev. 99:45-77.
in older adults can often be attributed to a shift from rec-
COURCHESNE, E., S. A. HILLYARD, and R. GALAMBOS, 1975.
ollection-based to familiarity-based processes (Jennings Stimulus novelty, task relevance and the visual evoked po-
and Jacoby, 1997; Parkin and Walter, 1992). tential in man. Electroencephalogr. Clin. Neurophysiol. 39:131-
143.
COWAN, N., 1995. Attention and Memory: An Integrated Frame-
REFERENCES work. New York: Oxford University Press.
ALHO, K., M. HUOTILAINEN, and R. NAATANEN, 1995. Are CZIGLER, I., G. CSIBRA, and A. CSONTOS, 1992. Age and inter-
memory traces for simple and complex sounds located in stimulus interval effects on event-related potentials to fre-
different regions of auditory cortex? Recent MEG studies. quent and infrequent auditory stimuli. Biol. Psychol. 33(2-
Electroencephalogr. Clin. Neurophysiol. 44(suppl.):197—203. 3):195-206.
ALHO, K., D. L. WOODS, A. ALGAZI, R. T. KNIGHT, and DEESE, J., 1959. On the prediction of occurrence of particular
R. NAATANEN, 1994. Lesions of frontal cortex diminish the verbal intrusions in immediate recall. J. Exp. Psychol. 58:17-
auditory mismatch negativity. Electroencephalogr. Clin. Neuro- 22.
physiol. 91(5):353-362. DESOTO, M. C., M. FABIANI, D. C. GEARY, and G. GRATTON, in
ANDERS, T. R., J. L. FOZARD, and T. D. LILLYQUIST, 1972. Ef- press. When in doubt, do it both ways: Brain evidence of
fects of age upon retrieval from short-term memory. Dev. the simultaneous activation of conflicting responses in a
Psychol. 6(2):214-217. spatial Stroop task. J. Cogn. Neurosci.
BADDELEY, A. D., 1986. Working Memory. Oxford, UK: Oxford D'EsposiTO, M., E. ZARAHN, G. K. AGUIRRE, and B. RYPMA,
University Press. 1999. The effect of normal aging on the coupling of neural
BALOTA, D. A., M. J. CORTESE, J. M. DUCHEK, D. ADAMS, H. L. activity to the bold hemodynamic response. Neuroimage
ROEDIGER III, K. B. MCDERMOTT, and B. E. YERYS, 1999. 10(1):6-14.
Veridical and false memories in healthy older adults and DONCHIN, E., 1981. Surprise! . . . Surprise? Psychophysiology
in dementia of the Alzheimer's type. Cogn. Neuropsychol. 18:493-513.
16:361-384. DUCHEK, J. M., D. A. BALOTA, M. E. FAUST, and F. R. FERRARO,
BASHORE, T. R., 1990. Age, physical fitness, and mental pro- 1995. Inhibitory processes in young and older adults in a
cessing speed. In Annual Review of Gerontology and Geriatrics, picture-word task. Aging Cognit. 2 (2): 156-167.
Vol. 9, M. P. Lawton, ed. New York: Springer, pp. 120-144. DUNCAN-JOHNSON, C. C., and B. S. KOPELL, 1981. The Stroop
effect: Brain potentials localize the source of interference.
BASHORE, T. R., and P. H. GODDARD, 1993. Preservative and
restorative effects of aerobic fitness on the age-related slow- Science 214:938-940.
ESLINGER, P. J., and L. M. GRATTAN, 1993. Frontal lobe and
ing of mental processing speed. In Cerella, J., J. Rhybash,
frontal-striatal substrates for different forms of human cog-
and W. Hoyer, eds. Adult Information Processing: Limits on
nitive flexibility. Neuropsychologia 31(l):17-28.
Loss, New York: Academic Press, pp. 205-228.
FABIANI, M., and D. FRIEDMAN, 1995. Changes in brain activity
BASHORE, T. R., A. OSMAN, and E. F. HEFFLEY, 1989. Mental
patterns in aging: The novelty oddball. Psychophysiology
slowing in elderly persons: A cognitive psychophysiological
32(6):579-594.
analysis. Psychol. Aging 4(2):235-244.
FABIANI, M., and D. FRIEDMAN, 1997. Dissociations between
BRAVER, T. S., J. D. COHEN, L. E. NYSTROM, J. JONIDES, E. E.
memory for temporal order and recognition memory in
SMITH, and D. C. NOLL, 1997. A parametric study of pre-
aging. Neuropsychologia 35(2):129-141.
frontal cortex involvement in human working memory. Neu- FABIANI, M., D. FRIEDMAN, and J. C. CHENG, 1998. Individual
roimage 5 (1): 49 - 62. differences in P3 scalp distribution in old subjects, and their
BROADBENT, D. E., 1957. A mathematical model for human relationship to frontal lobe function. Psychophysiology
attention and immediate memory. Psychol. Rev. 64:205-215. 35:698-708.
CABEZA, R., N. D. ANDERSON, S. HOULE, J. MANGELS, and L. FABIANI, M., D. FRIEDMAN, J. C. CHENG, E. WEE, and C. TROTT,
NYBERG, 2000. Age-related differences in neural activity 1999. Use it or lose it: Effects of aging and education on
during item and temporal-order memory retrieval: A posi- brain activity in the performance of recency and recogni-
tron emission tomography study. J. Cogn. Neurosci. tion memory tasks. J. Cogn. Neurosci. Suppl. 83B:18.
12(1):197-206. FABIANI, M., G. GRATTON, and M. G. H. COLES, 2000. Event
CARTER, C. S., T. S. BRAVER, D. M. BARCH, M. M. BOTVINICK, related brain potentials: Methods, theory and applications.
D. NOLL, and J. D. COHEN, 1998. Anterior cingulate cortex, In Tutorials in Psychophysiology: Physical, Social and Inferential
error detection, and the online monitoring of perfor- Elements, 2nd ed., J. T. Cacioppo, L. G. Tassinary, and G.
mance. Science 280 (5364) :747-749. Bernston, eds. New York: Cambridge University Press, pp.
CHAo, L. L., and R. T. KNIGHT, 1997a. Age-related prefrontal 53-84.
alterations during auditory memory. Neurobiol. Aging FABIANI, M., V. A. KAZMERSKI, Y. M. CYCOWICZ, and D. FRIED-
18(l):87-95. MAN, 1996. Naming norms for brief environmental sounds:
CHAo, L. L., and R. T. KNIGHT, 1997b. Prefrontal deficits in Effects of age and dementia. Psychophysiology 33 (4) :462-475.
attention and inhibitory control with aging. Cereb. Cortex FABIANI, M., M. A. STADLER, and P. WESSELS, 2000. True mem-
7(l):63-69. ories but not false ones produce a sensory signature in hu-
COHEN, J. D., T. S. BRAVER, and R. C. O'REILLY, 1996. A com- man lateralized brain potentials. J. Cogn. Neurosci. 12(6):
putational approach to prefrontal cortex, cognitive control 1-10.
and schizophrenia: Recent developments and current chal- FERNANDEZ, G., A. EFFERN, T. GRUNWALD, N. PEZER, K. LEH-
lenges. Phil. Trans. R. Soc. Lond. [B] 351 (1346) :1515-1527. NERTZ, M. DUMPELMANN, D. VAN ROOST, and C. E. ELGER,
FABIANI AND WEE: WORKING MEMORY AND FRONTAL LOBE FUNCTION 485
1999. Real-time tracking of memory formation in the hu- cortical mechanisms of mismatch negativity (MMN) gen-
man rhinal cortex and hippocampus. Science 285:1582- eration. Brain Res. 667(2):192-200.
1585. JENNINGS, J. M., and L. L. JACOBY, 1997. An opposition pro-
FISK, A. D., and W. A. ROGERS, 1991. Toward an understanding cedure for detecting age-related deficits in recollection:
of age-related memory and visual search effects. J. Exp. Psy- Telling effects of repetition. Psychol. Aging l2(2):352-361.
chol. Gen. 120(2):131-149. JOHNSON, M. K., S. F. NOLDE, M. MATHER, J. KOUNIOS, D. L.
FORD, J. M., and A. PFEFFERBAUM, 1991. Event-related poten- SCHACTER, and T. CURRAN, 1997. The similarity of brain
tials and eyeblink responses in automatic and controlled activity associated with true and false recognition memory
processing: Effects of age. Electroencephalogr. Clin. Neurophy- depends on test format. Psychol. Sci. 8(3):250-257.
siol. 78:361-377. JOHNSON, R., JR., and E. DONCHIN, 1978. On how P300 ampli-
FORD, J. M., A. PFEFFERBAUM, J. R. TINKLENBERG, and B. S. tude varies with the utility of the eliciting stimuli. Electroen-
KOPELL, 1982. Effects of perceptual and cognitive difficulty cephalogr. Clin. Neurophysiol. 44(4):424-437.
on P3 and RT in young and old adults. Electroencephalogr. JONIDES, J., C. MARSHUETZ, E. E. SMITH, P. A. REUTER-LORENZ,
Clin. Neurophysiol. 54(3):311-321. and R. A. KOEPPER, 2000. Age differences in behavior and
FORD, J. M., W. T. ROTH, R. C. MOHS, W. F. HOPKINS, and PET activation reveal differences in interference resolution
B. S. KOPELL, 1979. Event-related potentials recorded from in verbal working memory. J. Cogn. Neurosci. 12(1):188-196.
young and old adults during a memory retrieval task. Elec- JONIDES, J., E. H. SCHUMACHER, E. E. SMITH, E. J. LAUBER,J.
troencephalogr. Clin. Neurophysiol. 47(4):450-459. AWH, S. MINOSHIMA, and R. A. KOEPPE, 1997. Verbal work-
FRIEDMAN, D., V. KAZMERSKI, and M. FABIANI, 1997. An over- ing memory load affects regional brain activation as mea-
view of age-related changes in the scalp distribution of P3b. sured by PET. J Cogn. Neurosci. 9(4):462-475.
Electroencephalogr. Clin. Neurophysiol. 104(6):498—513. JONIDES, J., E. E. SMITH, R. A. KOEPPE, E. AWH, S. MINOSHIMA,
FRIEDMAN, D., M. FABIANI, and G. SIMPSON, 1999. J. Cogn. and M. A. MINTUN, 1993. Spatial working memory in hu-
Neurosci. 31A (suppl):21. mans as revealed by PET [see comments]. Nature
FRIEDMAN, D., G. SIMPSON, and M. HAMBERGER, 1993. Age- 363 (6430) :623-625.
KAUSLER, D. H., 1991. Experimental Psychology, Cognition and Hu-
related changes in scalp topography to novel and target
man Aging, 2nd ed. New York: Springer-Verlag.
stimuli. Psychophysiology 30:383-396.
KAUSLER, D. H., W. LICHTY, and R. T. DAVIS, 1985. Temporal
FUSTER, J. M., 1973. Unit activity in prefrontal cortex during
memory for performed activities: Intentionality and adult
delayed response performance: Neuronal correlates of tran-
age differences. Dev. Psychol. 21(6):1132-1138.
sient memory.J. Neurophysiol. 36:61-78.
KAUSLER, D. H., T. A. SALTHOUSE, and J. S. SAULTS, 1988. Tem-
GABRIELI, J., 1999. FMRI evidence for age-associated altera-
poral memory over the adult lifespan. Am. J. Psychol.
tions in prefrontal contributions to the encoding and re-
101(2):207-215.
trieval of new memories.J. Cogn. Neurosci. Suppl. 10. KAZMERSKI, V. A., D. FRIEDMAN, and W. RITTER, 1997. Mis-
GAETA, H., D. FRIEDMAN, W. RITTER, and J. CHENG, 1998. An match negativity during attend and ignore conditions in
event-related potential study of age-related changes in sen- Alzheimer's disease. Biol. Psychiatry 42(5):382-402.
sitivity to stimulus deviance. Neurobiol. Aging 19(5):447-459. KNIGHT, R. T, 1984. Decreased response to novel stimuli after
GOLDMAN-RAKIC, P. S., 1987. Circuitry of primate prefrontal prefrontal lesions in man. Electroencephalogr. Clin. Neurophy-
cortex and regulation of behavior by representational mem- siol. 59:9-20.
ory. In Handbook of Physiology—The Nervous System, Vol. 5, KNIGHT, R. T, 1987. Aging decreases auditory event-related
V. B. Mountcastle, F. Plum, and S. R. Geiger, eds. Bethesda, potentials to unexpected stimuli in humans. Neurobiol. Aging
Md.: American Physiological Association, pp. 373-417. 8(2):109-113.
GOLDMAN-RAKIC, P. S., 1992. Working memory and the mind. KNIGHT, R. T, 1997. Distributed cortical network for visual
Scientific American 267(3) :110-117. attention. J. Cogn. Neurosci. 9(1):75-91.
GRATTON, G., and M. FABIANI, 1998. Dynamic brain imaging: KNIGHT, R. T, M. F. GRABOWECKY, and D. SCABINI, 1995. Role
Event-related optical signal (EROS) measures of the time of human prefrontal cortex in attention control. Adv. Neu-
course and localization of cognitive-related activity. Psy- rol. 66:21-34; discussion 34-36.
chonom. Bull. Rev. 5:535-563. KNIGHT, R. T, and T. NAKADA, 1998. Cortico-limbic circuits
GUR, R. C., R. E. GUR, W. D. OBRIST, B. E. SKOLNICK, and and novelty: A review of EEG and blood flow data. Rev. Neu-
M. REIVICH, 1987. Age and regional cerebral blood flow at rosci. 9(1):57-70.
rest and during cognitive activity. Arch. Gen. Psychiatry KNIGHT, R. T, D. SCABINI, and D. L. WOODS, 1989. Prefrontal
44(7):617-621. cortex gating of auditory transmission in humans. Brain Res.
HAUG, H., and R. EGGERS, 1991. Morphometry of the human 504:338-342.
cortex cerebri and corpus striatum during aging [com- KRAMER, A. F., S. HAHN, N. J. COHEN, M. T. BANICH, E.
ment]. Neurobiol. Agwgl2(4):336-338; discussion 352-355. McAuLEY, C. R. HARRISON, J. CHASON, E. VAKIL, L. BAR-
HEATON, R. K., 1981. Wisconsin Cards Sorting Test Manual. DELL, R. A. BOILEAU, and A. COLCOMBE, 1999. Ageing, fit-
Odessa, Fla.: Psychological Assessment Resources. ness and neurocognitive function [letter]. Nature
HESS, T. M., and S. M. PULLEN, 1996. Memory in context. In 400(6743) :418-419.
Perspectives on Cognitive Change in Adulthood and Aging, F. KRAMER, A. E, S. HAHN, and D. GOPHER, 1999. Task coordi-
Blanchard-Fields and T. H. Hess, eds. New York: McGraw- nation and aging: Explorations of executive control pro-
Hill. cesses in the task switching paradigm. Acta Psychol., 101(2-
JAVITT, D. C., M. STEINSCHNEIDER, C. E. SCHROEDER, H. G. 3):339-378.
VAUGHAN, JR., and J. C. AREZZO, 1994. Detection of stimulus KUTAS, M., and K. D. FEDERMEIER, 1998. Minding the body.
deviance within primate primary auditory cortex: Intra- Psychophysiology 35(2):135-150.
486 COGNITION
LAVOIE, M. E., L. BHERER, and S. BELLEVILLE, 1999. Effect of Brunia, G. Mulder, and M. N. Verbaten, eds. Amsterdam:
aging on event-related potentials topography during a Elsevier, pp. 304-323.
Stroop task. Psychophysiology 36:S72. POLICH, J., 1996. Meta-analysis of P300 normative aging stud-
LIGHT, L. L., 1996. Memory and aging. In Memory. Handbook ies. Psychophysiology 33(4):334-353.
of Perception and Cognition, Vol. xxii, 2nd ed., E. L. Bjork, ed. POLICH, J., L. HOWARD, and A. STARR, 1985. Effects of age on
San Diego: Academic Press, pp. 443-490. the P300 component of the event-related brain potential
LIU, A. K,J. W. BELLIVEAU, and A. M. DALE, 1998. Spatiotem- from auditory stimuli: Peak definition, variation, and mea-
poral imaging of human brain activity using functional MRI surement. J. Gerontol. 40(6):721-726.
constrained magnetoencephalography data: Monte Carlo RABBITT, P., 1965. Age-decrement in the ability to ignore ir-
simulations. Proc. Natl. Acad. Sci. USA 95(15):8945-8950. relevant information. J. Gerontol. 20:233-238.
MCCARTHY, G., A. M. BLAMIRE, A. PUCE, A. C. NOBRE, G. REUTER-LORENZ, P. A., 1999. Cognitive neuropsychology of
BLOCH, F. HYDER, P. GOLDMAN-RAKIC, and R. G. SHULMAN, the aging brain. In Cognitive Aging: A Primer, D. C. Park and
1994. Functional magnetic resonance imaging of human N. Schwarz, eds. Philadelphia: Psychology Press, pp. 93-
prefrontal cortex activation during a spatial working mem- 114.
ory task. Proc. Natl. Acad. Sci. USA 91(18):8690-8694. REUTER-LORENZ, P. A., J. JONIDES, E. E. SMITH, A. HARTLEY,
MCCARTHY, G., and E. DONCHIN, 1981. A metric for thought: A. MILLER, C. MARSHUETZ, and R. A. KOEPPE, 2000. Age
A comparison of P300 latency and reaction time. Science differences in the frontal lateralization of verbal and spatial
211 (4477) :77-80. working memory revealed by PET. J. Cogn. Neurosci.
MECKLINGER, A., 1998. On the modularity of recognition 12(1):174-187.
memory for object form and spatial location: A topographic RINNE, T, G. GRATTON, M. FABIANI, N. COWAN, E. MACLIN, A.
ERP analysis. Neuropsychologia36(5):44l-46Q. STINARD, J. SINKKONEN, K. ALHO, and R. NAATANEN, 1999.
MILNER, B., 1963. Effects of different brain lesions on card Scalp-recorded optical signals make sound processing from
sorting. Arch. Neurol. 9:90-100. the auditory cortex visible. Neuroimage 10:620-624.
MILNER, B., 1995. Aspects of human frontal lobe function. In RITTER, W., D. DEACON, H. GOMES, D. C. JAVITT, and H. G.
Epilepsy and the Functional Anatomy of the Frontal Lobe, Vol. 66, VAUGHAN, JR., 1995. The mismatch negativity of event-
H. H. Jasper, ed. New York: Raven Press, pp. 67-84. related potentials as a probe of transient auditory memory:
MILNER, B., and M. PETRIDES, 1984. Behavioural effects of A review. Ear Hearing 16(1) :52—67.
frontal-lobe lesions in man. Trends Neurosci. 7:403-407. ROEDIGER, H. L., and K. B. MCDERMOTT, 1995. Creating false
MOSCOVITCH, M., and G. WINOCUR, 1992. The neuropsychol- memories: Remembering words not presented in lists. J.
ogy of memory and aging. In The Handbook of Aging and Exp. Psych.: Learning, Mem. & Cogn. 21(4):803-814.
Cognition, F. I. M. Craik and T. A. Salthouse, eds. Hillsdale, RUCHKIN, D. S., H. L. CANOUNE, R. JOHNSON, JR., and W. RIT-
N.J.: Erlbaum, pp. 315-372. TER, 1995. Working memory and preparation elicit differ-
MOUNTAIN, M. A., and W. G. SNOW, 1993. Wisconsin Card Sort- ent patterns of slow wave event-related brain potentials.
ing Test as a measure of frontal pathology: A review. Clin. Psychophysiology 32(4):399-410.
Neuropsychol. 7 (1): 108-118. SALTHOUSE, T. A., 1996. The processing-speed theory of adult
NAATANEN, R., 1992. Attention and Brain Function. Hillsdale, age differences in cognition. Psychol. Rev., 103(3):403-428.
N.J.: Erlbaum. SCHACTER, D. L., R. L. BUCKNER, W. KOUTSTAAL, A. M. DALE,
NAATANEN, R., and K. ALHO, 1995. Mismatch negativity—A and B. R. ROSEN, 1997. Late onset of anterior prefrontal
unique measure of sensory processing in audition. Intl. J. activity during true and false recognition: An event-related
Neurosci. 80(l-4):3l7-337. fMRI study. Neuroimage 6(4):259-269.
PARKIN, A. J., and R. I. JAVA, 1999. Deterioration of frontal SCHACTER, D. L., L. ISRAEL, and C. RACINE, 1999. Suppressing
lobe function in normal aging: Influences of fluid intelli- false recognition in younger and older adults: The distinc-
gence versus perceptual speed. Neuropsychology 13(4):539- tiveness heuristic. J. Mem. Lang. 40(1):1-24.
545. SCHACTER, D. L., J. F. KIHLSTROM, A.W. KASZNIAK, and M.
PARKIN, A. J., and B. M. WALTER, 1991. Aging, short-term VALDISERRI, 1993. Preserved and impaired memory func-
memory, and frontal dysfunction. Psychobiology 19(2) :175- tions in elderly adults. In Adult Information Processing: Limits
179. on Loss, J. Cerella, ed. San Diego: Academic Press, pp. 327-
PARKIN, A. J., and B. M. WALTER, 1992. Recollective experi- 350.
ence, normal aging, and frontal dysfunction. Psychol. Aging SCHACTER, D. L., W. KOUTSTAAL, M. K. JOHNSON, M. S. GROSS,
7(2):290-298. and K. E. ANGELL, 1997. False recollection induced by pho-
PARR, W. V, and R. SIEGERT, 1993. Adults' conceptions of ev- tographs: A comparison of older and younger adults. Psy-
eryday memory failures in others: Factors that mediate the chol. Aging 12(2) :203-215.
effects of target age. Psychol. Aging 8(4):599-605. SCHACTER, D. L., E. REIMAN, T. CURRAN, L. S. YUN, D. BANDY,
PEKKONEN, E., T. RINNE, K REINIKAINEN, T. KUJALA, K. ALHO, K. B. MCDERMOTT, and H. L. ROEDIGER III, 1996. Neuroan-
and R. NAATANEN, 1996. Aging effects on auditory process- atomical correlates of veridical and illusory recogni-
ing: An event-related potential study. Exp. Aging Res. tion memory: Evidence from positron emission. Neuron
22(2):171-184. 17(2):267-274.
PETRIDES, M., B. ALIVISATOS, A. C. EVANS, and E. MEYER, SENKFOR, A. J., and C. VAN PETTEN, 1996. ERP measures of
1993. Dissociation of human mid-dorsolateral from poster- source and item memory in young and elderly subjects. Psy-
ior dorsolateral frontal cortex in memory processing. Proc. chophysiology S71.
Natl. Acad. Sci. USA 90(3):873-877. SENKFOR, A. J., and C. VAN PETTEN, 1998. Who said what? An
POLICH, J., 1991. P300 in the evaluation of aging and demen- event-related potential investigation of source and item
tia. In Event-Related Brain Research (EEG Suppl. 42), C. H. M. memory. J. Exp. Psychol. Learn. Mem. Cogn. 24(4): 1005-1025.
FABIANI AND WEE: WORKING MEMORY AND FRONTAL LOBE FUNCTION 487
SHIMAMURA, A. P., 1995. Memory and frontal lobe function. with frontal lesions: Evidence from word list learning. Psy-
In The Cognitive Neurosciences, M. S. Gazzaniga, ed. Cam- chol. Aging 11 (3) :387-395.
bridge, Mass.: MIT Press, pp. 803-813. STUSS, D. T, G. A. ESKES, and J. K. FOSTER, 1994. Experimental
SHIMONY, J. S., R. C. MCKINSTRY, E. AKBUDAK, J. A. ARONOV- neuropsychological studies of frontal lobe function. In
ITZ, A. Z. SNYDER, N. F. LORI, T. S. CULL, and T. E. CONTURO, Handbook of Neuropsychology, Vol. 9, F. Boiler and J. Grafman,
1999. Quantitative diffusion-tensor anisotropy brain MR im- eds. Amsterdam: Elsevier, pp. 149-185.
aging: Normative human data and anatomic analysis. Radi- SUTTON, S., M. BRAREN, J. ZUBIN, and E. R. JOHN, 1965.
ology 212(3):770-784. Evoked-potential correlates of stimulus uncertainty. Science
SMITH, E. E., J. JONIDES, and R. A. KOEPPE, 1996. Dissociating 150:1187-1188.
verbal and spatial working memory using PET. Cereb. Cortex TOGA, A. W., and J. C. MAZZIOTTA (EDS.) , 1996. Brain Mapping.
6(1):11-20. The Methods. San Diego: Academic Press.
SMITH, E. E., J. JONIDES, and R. A. KOEPPE, 1998. Dissociating TROTT, C. T., D. FRIEDMAN, W. RITTER, and M. FABIANI, 1997.
verbal and spatial working memory using PET. Corrigen- Item and source memory: Differential age effects revealed
dum. Cereb. Cortex 8 (8) :762. by event-related potentials. NeuroReport 8(15):3373-3378.
SMITH, E. E., J. JONIDES, R. A. KOEPPE, E. AWH, E. H. SCHU- TROTT, C. T, D. FRIEDMAN, W. RITTER, M. FABIANI, and J. G.
MACHER, and S. MINOSHIMA, 1995. Spatial versus object SNODGRASS, 1999. Episodic priming and memory for tem-
working memory: PET investigations. J. Cogn. Neurosci. poral source: Event-related potentials reveal age-related
7(3):337-356. differences in prefrontal functioning. Psychol. Aging
SNOWDON, D. A., L. H. GREINER, J. A. MORTIMER, K P. RILEY, 14(3):390-413.
P. A. GREINER, and W. R. MARKESBERY, 1997. Brain infarc- TULVING, E., 1985. Memory and consciousness. Can. J. Psychol.
tion and the clinical expression of Alzheimer disease. The 26:1-12.
Nun Study [see comments]. JAMA 277(10):813-817. WEE, E., and M. FABIANI, 1999a. Differential effects of aging
SNOWDON, D. A., S. J. KEMPER, J. A. MORTIMER, L. H. GREINER, on encoding and processing of object/spatial stimuli in a
D. R. WEKSTEIN, and W. R. MARKESBERY, 1996. Linguistic
modified two-back task. Psychophysiology 36:S123.
ability in early life and cognitive function and Alzheimer's
WEE, E., and M. FABIANI, 1999b. Aging effects on behavior and
disease in late life. Findings from the Nun Study [see corn-
event-related brain potentials in a modified two-back task.
ments]. JAMA 275(7):528-532.
J. Cogn. Neurosci. 82(suppl.):21.
SPIELER, D. H., D. A. BALOTA, and M. E. FAUST, 1996. Stroop
WEST, R., and C. ALAIN, 2000. Age-related decline in inhibi-
performance in healthy younger and older adults and in
individuals with dementia of the Alzheimer's type. J. Exp. tory control contributes to the increased Stroop effect ob-
Psychol. Hum. Percept. Perform. 22(2):461-479. served in older adults. Psychophysiology 37:179-189.
STERNBERG, S., 1966. High-speed scanning in human memory. WEST, R. L., 1996. An application of prefrontal cortex function
Science 153(3736) :652-654. theory to cognitive aging. Psychol. Bull. 120(2):272-292.
STRAYER, D. L., and A. F. KRAMER, 1994. Aging and skill ac- WOODRUFF-PAK, D. S., 1997. The Neuropsychology of Aging, Vol.
quisition: Learning-performance distinctions. Psychol. Aging 352. Cambridge, Mass.: Blackwell Publishers.
9(4):589-605. WOODS, D. L., 1992. Auditory selective attention in middle-
STRAYER, D. L., C. D. WICKENS, and R. BRAUNE, 1987. Adult aged and elderly subjects: An event-related brain potential
age differences in the speed and capacity of information study. Electroencephalogr. Clin. Neurophysiol. 84(5):456-468.
processing: II. An electrophysiological approach. Psychol. YAMAGUCHI, S., and R. T. KNIGHT, 1991a. P300 generation by
Aging 2(2):99-110. novel somatosensory stimuli. Electroencephalogr. Clin. Neuro-
STROOP, J. R., 1935. Studies of interference in serial verbal physiol. 78(l):50-55.
reactions. J. Exp. Psychol. 18:643-662. YAMAGUCHI, S., and R. T. KNIGHT, 1991b. Age effects on the
STUSS, D. T., F. I. CRAIK, L. SAYER, D. FRANCHI, and M. P. P300 to novel somatosensory stimuli. Electroencephalogr. Clin.
ALEXANDER, 1996. Comparison of older people and patients Neurophysiol. 78(4):297-301.
488 COGNITION
VII
NEURO-
DEVELOPMENTAL
ASPECTS OF
CLINICAL
DISORDERS
This page intentionally left blank
31 The Role of Nutrition
in Cognitive Development
MICHAEL K. GEORGIEFF
AND RAGHAVENDRA RAO
ABSTRACT Nutrients are critical for optimal brain develop- provides energy for storage (as adipose tissue), but can
ment and function. Although maintenance of adequate intake also be utilized more slowly than glucose to provide
of all nutrients throughout life is important for brain health acute energy to the entire body, including the brain.
and function, certain nutrients have a more profound effect
on brain development than others. The timing of nutrient Certain fats and lipoproteins are important for normal
supplementation or deprivation also has an important effect neuronal cell membrane integrity and myelination. For
on brain development and function. Nutrient intakes may example, cholesterol, phosphatidyl choline, and certain
vary widely; yet, for certain nutrients, there must be a steady fatty acids are essential for cell membrane synthesis and
flow across the blood-brain barrier to maintain brain devel- integrity. Linolenic acid, linoleic acid, arachidonic acid,
opment and functional homeostasis. Storage and subsequent
mobilization of nutrients from body pools ensure that the
and docosohexaenoic acid are essential for normal
brain receives neither too much nor too little of each nutrient. brain myelination. During starvation or periods of ill-
This chapter reviews the evidence for the role of nutri- ness, protein can also be used as an energy substrate.
tion in brain development, concentrating especially on nutri- When protein is utilized in such a manner, it is not avail-
ents that are important in the development of cognitive able for structural tissue (including brain) synthesis.
function. A short review of nutrient categories is followed by There is an extensive literature in both humans and
a section on how individual nutrients directly and indirectly
affect developing brain structure and function. A detailed animal models on the isolated and combined adverse
analysis of the nutrients that most profoundly affect brain de- effects of protein and energy malnutrition on the de-
velopment ensues, with emphasis on the importance of the veloping brain (Pollitt, 1996; Pollitt and Gorman, 1994;
timing of nutrient deprivation or supplementation. Pollitt, Watkins, Husaini, 1997; Pollitt et al., 1993, 1995;
Winick and Nobel, 1966; Winick and Rosso, 1969a,b).
Pollitt and Gorman (1994) have pointed out that other
Nutrient categories and growth factors nutrient deficiencies usually coexist with protein-en-
ergy malnutrition (PEM) in free-living populations.
MACRONUTRIENTS Protein and energy are considered
the two macronutrients. Protein is typically used by the
body for somatic (tissue) protein and serum protein MINERALS The major minerals of the body are sodium,
synthesis. Proteins also include all enzymes found potassium (with their usual accompanying dietary an-
within organs. Energy in the form of fat or carbohy- ion, chloride), calcium, and phosphorus. Minerals are
drates is the metabolic fuel on which the body depends. not classically considered essential for brain develop-
The body preferentially uses carbohydrates (typically ment, but deficiencies in these nutrients will lead to
glucose) as the substrate for cellular metabolism to gen- abnormal brain function, mostly through altering neu-
erate adenosine triphosphate (ATP). Neuronal metab- ronal electrical function.
olism is particularly dependent on glucose availability
and is very sensitive to periods of carbohydrate depri- MlCRONUTRIENTS AND TRACE ELEMENTS This class of
vation (Volpe, 1995). Hypoglycemia has a particularly nutrients contains elements that are required in trace
profound negative effect on the developing hippocam- quantities by the body and are used, for the most part,
pus (Barks et al., 1995). Fat, with its higher caloric value, in intermediary cellular metabolism. Members of this
category include magnesium, manganese, iodine, zinc,
copper, molybdenum, cobalt, selenium, fluoride, and
MICHAEL K. GEORGIEFF Professor of Pediatrics and Child De- iron. As with the major minerals, these elements are not
velopment, School of Medicine, Institute of Child Develop-
ment, University of Minnesota, Minneapolis, Minnesota. classically considered to be uniquely important for nor-
RAGHAVENDRA RAO Instructor of Pediatrics, School of Med- mal brain development except as their deficiencies
icine, University of Minnesota, Minneapolis, Minnesota. affect cellular (including neuronal) function. Two
The biochemical and neuroanatomic bases of neu- prioritizes protein and energy during deficiency states.
rodevelopmental impairment from early PEM can be Animal models of IUGR support the human findings
found in human autopsy studies and in investigations of lighter brain weights; reduced neuronal DNA and
using animal models. The human studies show signifi- RNA content, mRNA for neuronal and glial structural
cant reductions in brain DNA, RNA, and protein con- proteins, synapse number, and neurotransmitter pep-
tent (Winick and Nobel, 1966; Winick and Rosso, tide production; and synaptic junction ultrastructural
1969a,b). IUGR infants have lower brain cell number, changes (Bass, Netsky, and Young, 1970; Cragg, 1972;
smaller cell size, and smaller head circumferences. Cer- Jones and Dyson, 1976; Wiggins, Fuller, and Enna,
tain areas (e.g., the cerebellum, cerebral cortex, hip- 1984). Fatty acid profiles are profoundly altered, with
pocampus) demonstrate more profound effects than subsequent reductions in myelination, brain lipid com-
others, suggesting that the developing brain in some way position, and learning ability in fat-restricted rats
(Yamamoto et al., 1987). Malnutrition also downregu- year. Important regions that become functionally sig-
lates CNS growth factors critical for normal brain de- nificant during this year include motor cortex, myeli-
velopment (Nishijima, 1986). The study by Lee and nating motor tracts, rapid cerebellar development, and
colleagues (1999) in transgenic mice emphasizes the hippocampal-prefrontal connections. Infants in the
importance of maintaining IGF-I levels during periods United States and Canada are typically fed either hu-
of malnutrition in order to spare regions of the brain man milk or infant formula derived from cow's milk or
important for cognition. soy plant products during this time. The significant,
It will be important in future research on IUGR in- positive relationship between breastfeeding and cogni-
fants to link specific regional neuropathologic or neu- tive development is intriguing, since it is unclear
rochemical findings (elucidated from controlled whether the positive effect is related to nutritional fac-
animal models of PEM) with deficits in behaviors tor (s) present in human milk but not in formula, to
known to be based in those regions. Thus, one should positive maternal-infant interactions (including the
be able to relate the reduction in visual recognition propensity of higher-IQ mothers to choose breastfeed-
memory processing in IUGR infants (Gottleib, Biasini, ing), or both.
and Bray, 1988) to either reduced hippocampal volume There is good reason to believe that multiple factors
or metabolism on MRI/MRI spectroscopy. found in breast milk promote normal CNS develop-
ment, and that deficiencies of these nutrients in cow's
Effects of postnatal PEM The postnatal brain grows rap- milk or soy-based formulas are responsible for slower
idly during the first six postnatal months and then slows rates of cognitive development (Lucas, 1997; Morrow-
considerably during the last six months of the first Tlucak, Haude, and Ernhart, 1988; Wang and Wu,
ABSTRACT The developmental cognitive neuroscience ap- Our clinical experience with hundreds of patients
proach offers an important framework for understanding the identified as FAS or FAE leads us to believe that child
complexities of fetal alcohol syndrome and other prenatal al- and adult psychopathology is an important outcome of
cohol effects. While the genesis of this main dysfunction lies
in alcohol exposure during the prenatal period, the neuro- alcohol teratogenesis, thus relating this work to the
behavioral sequelae last throughout postnatal life, during broader fields of psychology, psychiatry, and develop-
which adverse social experiences can interact with prenatal mental disabilities (Streissguth, 1997; Streissguth and
brain damage to produce secondary disabilities. Animal, clini- Kanter, 1997; Streissguth and O'Malley, 2000). Concep-
cal, and epidemiologic studies have confirmed many types of tualizing alcohol teratogenesis within the constructs of
neurodevelopmental deficits caused by prenatal alcohol ex-
posure. The challenge for the future is to integrate these find- developmental cognitive neuroscience should provide
ings into a meaningful body of knowledge in order to identify a theoretical framework supporting the types of re-
and intervene with affected children, reduce secondary dis- search that need to be undertaken in order to under-
abilities, and promote healthy and productive lives. The mag- stand these phenomena scientifically and to meet the
nitude of the problem of fetal alcohol-affected youth makes
this a challenge worth undertaking.
clinical needs of individuals of all ages who are them-
selves affected by prenatal alcohol exposure.
Cognitive neuroscience involves the union of cogni-
Establishing the developmental cognitive tive psychology (the study of perception, attention, lan-
neuroscience context for fetal alcohol syndrome guage, memory and the organization of action) with
neural science (the localization within the brain of vari-
Fetal alcohol syndrome (FAS) is a birth defect caused ous cognitive capabilities) (Kandel and Squire, 1992).
by prenatal alcohol exposure, which produces a spec- Developmental cognitive neuroscience provides the
trum of lifelong effects on offspring depending on the lifespan context in which the processes of neural sci-
dose, timing, and conditions of exposure. In this chap- ence produce the outcomes of cognitive psychology.
ter, we review and explore the developmental cognitive This approach should help us understand child psycho-
neuroscience implications of FAS and other fetal alco- pathology as it emerges from insults to the brain during
hol effects (FAE). Although prenatal alcohol increases various stages of development. For children with FAS/
the rate of major congenital malformations, producing FAE, the relevant period of primary damage is prior to
a variety of physical abnormalities including character- birth, during the period of gestational development
istic facial features in childhood, it is the consequences and organogenesis when the developing brain is par-
of brain maldevelopment that are of primary focus ticularly vulnerable to the teratogenic effects of alcohol.
here. The reverberations of this damage persist throughout
childhood, adolescence, and into adulthood.
As Harris (1998) notes, "The developmental perspec-
ANN P. STREISSGUTH AND PAUL D. CONNOR Fetal Alcohol and
Drug Unit, Department of Psychiatry and Behavioral Sciences, tive emphasizes maturation of the brain in the context of
University of Washington School of Medicine, Seattle, social experience, the environmental interface" (p. 166).
Washington. For FAS/FAE, there is a vast literature documenting
ABSTRACT In utero drug exposure can lead to specific, tar- tial impact that an altered neurochemical environment
geted changes in brain structure and function through direct may have on histogenesis. In this chapter we describe
and indirect modification of developing neurotransmitter sys- research findings that implicate monoamine systems,
tems and intracellular messengers. This chapter describes as-
pects of cortical development that can be modulated by
especially dopamine, in regulating certain aspects of
developing neurotransmitter systems, which, in turn can be neural development. These neurotransmitters are par-
affected by prenatal exposure to cocaine and other drugs of ticularly susceptible to modifications by use of cocaine
abuse. Also reviewed are data from animal models of gesta- and other drugs of abuse during pregnancy.
tional exposure to cocaine demonstrating permanent defects
in the structure and function of limbic cortical areas known
to play a role in the regulation of attention in drug-exposed Fundamentals of cerebral cortical development
offspring. We propose that similar changes in pattern forma-
tion likely underlie the complex cognitive and behavioral def- The cerebral cortex mediates higher cognitive func-
icits that have been described in children exposed to cocaine tions and is responsible for the integration of complex
and/or other drugs of abuse during prenatal development. sensory, motor, and homeostatic information. Defects
in cortical development, therefore, can have a pro-
The development of the brain relies on the spatial found impact on mature brain functions. It has been
and temporal regulation of cell-cell interactions that suggested that developmental anomalies in cortical de-
are controlled by contact-mediated and diffusible effec- velopment underlie certain types of psychopathology,
tor substances. Both the availability of molecular signals such as schizophrenia (Bloom, 1993; Raedler, Knable,
and the ability of developing cells to respond to those and Weinberger, 1998; Elvevag and Weinberger, this vol-
signals, by expression of specific receptors, are essential ume), and forms of mental retardation and autism
to ontogenetic processes (Goodman and Shatz, 1993; (Charman, 1999; Wong et al., 1995). The molecular and
Levitt, 1999; Lillien, 1998; McConnell, 1995a). Tran- cellular bases that tie developmental defects to cortical
scription factors, cell adhesion molecules, and neuro- dysfunction in these disorders remain unknown, but we
trophic factors all have been established as mediators know that influences on cell-cell interactions that me-
of tissue patterning, histotypic organization, and circuit diate specific developmental events are likely targets.
formation (Aplin et al., 1998; Fields and Itoh, 1996; This also seems to be true for nongenetic alterations in
Katz and Shatz, 1996; Segal and Greenberg, 1996). It is development, such as prenatal exposure to stress (Tak-
clear, however, that many molecules exhibit pleio- ahashi, 1998) and drugs of abuse (Ferriero and Demp-
trophic activities, serving as regulators of distinct cel- sey, 1999; Levitt, 1998; Malanga and Kosofsky, 1999).
lular functions at different times in development and
adulthood (Levitt, 1999). Neurotransmitters and neu- THE BASICS or BRAIN DEVELOPMENT Temporally over-
romodulators are now recognized as exhibiting multi- lapping events, grouped in five major categories, con-
ple activities, performing very different roles in cellular tribute to the formation of all brain structures from the
communication in the mature brain and during devel- neural tube (figure 33.1; see color plate 21). Progenitor
opment (Lauder, 1993; Levitt, Rakic, and Goldman- cells of the germinal matrix, situated along the forerun-
Rakic, 1997; Mattson, 1988). Here, we focus on several ner of the ventricular system, give rise to all neurons
features of brain development to understand the poten- and macroglia in a well-controlled proliferative process.
Postmitotic neurons migrate subsequently in a directed
GREGG D. STANWOOD AND PAT LEVITT Department of Neu- fashion from their place of origin to their final resting
robiology, University of Pittsburgh School of Medicine, Pitts- position. Neuronal differentiation involves the expres-
burgh, Pennsylvania. sion of specific gene products which, together with the
appearance of unique structural features (axons and CEREBRAL CORTICAL HISTOGENESIS The neocortex is
dendrites), contribute to the remarkable phenotypic di- a six-layered structure that exhibits very similar features
versity of the nervous system. Glial cells differentiate across its tangential extent. Thus, all functional areas
early to form specialized migration guides, radial glia, have repeated laminar and columnar organization that
and later to form astrocytes and oligodendrocytes. Pro- is assembled during development in well-defined tem-
genitor cells, neurons, and glia undergo naturally oc- poral and spatial patterns. The first neurons produced
curring cell death, a complex process that appears to be actually are not located in the forerunner of the cortex,
a normal developmental mechanism to establish appro- the cortical plate, but rather form a structure called the
priate quantitative relationships between projection and preplate, which eventually is split into a subplate and a
target neurons and between neurons and glia. Synapto- supraplate in the marginal zone by the first neurons
genesis is a temporally extended developmental event, destined for the cortical plate. The process of splitting
beginning prenatally and lasting through adolescence the preplate is an important first step in establishing
in the central nervous system (CNS) of all mammals, appropriate migration patterns of neurons. In the mu-
and which is critical in the formation of synapses be- tant mouse reeler, in which the protein reelin is defec-
tween specific target populations of neurons. tive, this process is abnormal and the cortex is
directed behaviors (Hyman, 1996; Kiyatkin, 1995; Koob, ONTOGENY OF THE MESOCORTICAL DOPAMINE SYSTEM
1992; Nestler and Aghajanian, 1997). The mesocort- As described earlier, cortical development involves a
ical system provides dopaminergic afferents to the an- complicated series of regulated events. In the rabbit,
terior cingulate (ACC) and medial prefrontal cortices the species in which we have extensively examined the
(MPF). These regions have been implicated in ori- effects of prenatal cocaine exposure, cells begin leaving
enting, emotional, and attentional processes (Devin- the proliferative zones at embryonic day (E)13 and
sky, Morrell, and Vogt, 1995; Goldman-Rakic, 1996; reach the deepest layers of cortex by El5 (Stensaas,
Pardo, Fox, and Raichle, 1991); and disruption of 1967a). The total gestational period in this animal is
mesocortical DA neurotransmission has been associ- approximately 30 days. By El6, approximately equiva-
ated with neuropsychiatric illnesses such as schizo- lent to a human embryo at the beginning of the third
phrenia and stress-related disorders (Deutch, 1993; month of gestation, the cells in the cortical plate with
Kiyatkin, 1995). a pyramidal morphology are evident (Stensaas, 1967b).
lacking functional D1 receptors (unpublished observa- Although evidence for changes in 5-HT or norepineph-
tions) , providing further validation of this hypothesis. rine activity has not been identified in the rabbit model,
The story is undoubtedly more complex, however, as deficits in serotonergic functioning certainly have been
the resulting changes in D1 receptor signaling due to implicated in higher dose models of prenatal cocaine
D1-Gs uncoupling also likely modifies developing glu- exposure (Akbari et al., 1992; Battaglia et al., 1998;
tamate and GABA systems. As mentioned earlier, Dl re- Snyder-Keller and Keller, 1993; Whitaker-Azmitia, 1998;
ceptor stimulation can facilitate NMDA- and ameliorate Zhang, Schrott, and Sparber, 1998).
AMPA/kainate-receptor responsiveness (Cepeda and
Levine, 1998; Yang and Seamans, 1996) and regulate Structural and functional deficits observed in
the phosphorylation state of the NMDA-receptor (Sny- humans exposed to cocaine in utero
der et al., 1998, 2000). Prenatal cocaine-induced in-
creases in parvalbumin-containing dendrites of cortical Reports of the impact of prenatal cocaine exposure on
interneurons may indicate a greater need for calcium newborns have been confusing, as some reports suggest
buffering following an increase in excitatory drive. Sim- gross physical malformations, others observe specific
ilarly, DA can both promote and inhibit GABA release deficits in cognitive and emotional development, and
based on the complement of activated receptor sub- yet others observe no effects whatsoever (for review, see
types (Cameron and Williams, 1993; Harsing and Gingras et al., 1992; Hawley, 1994; Mayes et al., 1998).
Zigmond, 1997), and the increases in GABA-immuno- These variable outcomes are likely the result of differ-
reactive neurons in the rabbit model suggest cocaine- ences in the experimental designs of these studies,
induced disruption of DA-GABA relationships in the some of which have been poorly controlled for impor-
developing limbic cortex. tant variables such as the timing and amount of cocaine
Further interactions between dopamine and the use during the pregnancy, polydrug use, and extent of
monoamines norepinephrine and 5-HT are also likely pre- and postnatal care.
to play a role. Cocaine blocks the reuptake of norepi- Teratogenic effects of cocaine exposure on birth
nephrine and 5-HT and these monoamines also influ- weight, body length, and head circumference have
ence excitatory and inhibitory activity within the cortex. been suggested by several studies (Chasnoff et al., 1985,
ABSTRACT This chapter summarizes research on the neu- terests and behaviors. In the social domain, the symp-
robiology and cognition of autism. Neurobiological investi- toms listed in the DSM-IV include impaired use of
gations have yielded inconsistent and often contradictory nonverbal behaviors (e.g., eye contact, facial expres-
results. No "signature" autistic anomaly has yet been discov-
ered, nor is a single one expected, given the phenotypic het- sion, gestures) to regulate social interaction, failure to
erogeneity of the disorder. It is suggested that rigorous studies develop age-appropriate peer relationships, little spon-
at the cognitive level that define the core neuropsychological taneous seeking to share enjoyment or interests with
features of autism will prove informative to future neurobio- other people, and limited social-emotional reciprocity.
logical investigations. Previous studies of cognition in autism, DSM-IV symptoms of communicative dysfunction in-
especially executive function, are summarized, and future di-
rections are suggested. Specifically, it appears critical that fu- clude delay in or absence of spoken language, difficulty
ture studies employ a component process approach to any initiating or sustaining conversations, idiosyncratic or
area of cognition investigated, as well as attend carefully to repetitive language, and imitation and pretend play def-
seemingly irrelevant, but potentially critical, task dimensions, icits. In the behaviors and interests domain, there is of-
such as administration format, response mode, and feedback ten an encompassing preoccupation or unusual interest
provision. It is hoped that these steps will eventually bring us
closer to understanding the developmental cognitive neuro- that is abnormal in intensity, inflexible adherence to
science of autism. nonfunctional routines, stereotyped body movements,
and preoccupation with parts or sensory qualities of
Autism is a lifelong developmental disorder typified objects (APA, 1994). In order to meet criteria for a
by social difficulties, communicative limitations, and a diagnosis of Autistic disorder, an individual must dem-
restricted range of interests and behaviors. Onset of the onstrate at least six of the twelve symptoms listed, with
disorder occurs before 36 months of age, although it is at least two coming from the social domain and one
not always recognized at this time. Boys are more often each from the communication and restricted behav-
affected than girls, with gender ratio estimates ranging iors/interests categories. Additionally, at least one
from 2 to 4 males for every female with the disorder symptom must have been present before 36 months of
(Lord and Schopler, 1987). While once considered a age (APA, 1994). Even among individuals meeting full
rare condition, recent prevalence estimates suggest that criteria for Autistic disorder, symptoms can vary widely
autism occurs in approximately 1 in 1000 individuals in severity, spanning a continuum from easily recog-
(Bryson, Clark, and Smith, 1988). It is currently sus- nized deviant behaviors to much milder presentations
pected that higher prevalence estimates do not indicate (Wing, 1988). There is also a wide range of cognitive
an actual increase in the occurrence of autism, but ability associated with Autistic disorder. Approximately
rather, better detection of the disorder, particularly in 75% of individuals with autism are diagnosed with
its milder forms (Gillberg, Steffenburg, and Schau- cornorbid mental retardation, but the other 25%, so-
mann, 1991). called "high-functioning" individuals, test in the bor-
The most commonly used diagnostic system in the derline intellectual range or above (e.g., full scale IQ>70).
United States is the Diagnostic and Statistical Manual of 70).
Mental Disorders, 4th edition DSM-IV), published by the High-functioning autism must be distinguished from
American Psychiatric Association (APA, 1994). As spec- Asperger disorder, a condition sharing the social dis-
ified in the DSM-IV, all individuals with autism dem- abilities and restricted, repetitive behaviors of autism,
onstrate evidence of difficulties with social relatedness, but in which language abilities are well-developed and
verbal and nonverbal communication, and range of in- cognitive functioning falls in the nonretarded range.
Although described more than 50 years ago by Austrian
SALLY OZONOFF Department of Psychology, University of pediatrician Hans Asperger (1944), the Asperger dis-
Utah, Salt Lake City, Utah. order was not included in the DSM until the manual's
"The motor centres and motor faculties, besides furnishing the condi- Association, 1994). Tics are sudden habitual move-
tions and possibilities of multiple and varied voluntary movements ments, gestures, or utterances that typically mimic some
. . . enormously widen the field of sensory experience and complicate fragment of normal behavior. This chapter focuses on
its results."
—David Ferrier, 1886
neurobiological substrates of tics and reviews concep-
tual advances derived from the integration of develop-
"From bad habit. . . take care my dear to guard against bad habits. " mental psychopathology with insights from basic and
—Samuel Johnson, discussing his tic symptoms with a curious clinical neuroscience research. Points of emphasis
young woman (quoted by Murray, 1982)
include a review of the neural substrates of habit for-
"What makes a habit a habit?" mation, the characterization of the clinical and neuro-
—Anonymous psychological phenotype of tic disorders, the likely role
of specific neurobiological circuits in the expression
ABSTRACT Tourette's syndrome (TS) is a model develop- and inhibition of tic symptoms, and the potential value
mental psychopathological disorder that has the potential to of nonlinear dynamic systems to understand the timing
illuminate aspects of how the brain selects and maintains per- of tics and the natural history of TS. Given the limits of
ceptual-emotive-cognitive-action sequences. This chapter re-
views aspects of the phenomenology, neuropsychology, and
space, the symptoms of OCD and ADHD will be ad-
natural history of TS in light of emerging knowledge concern- dressed only in passing. Interested readers are referred
ing the role of the basal ganglia and related cortical and tha- to more comprehensive reviews elsewhere (Leckman
lamic sites in the formation and timing of habits and other and Cohen, 1998).
goal-directed behaviors. Advances in developmental and sys-
tems neuroscience combined with the advent of noninvasive
functional brain imaging techniques make this a propitious
Neural substrates of habit formation
time to study TS.
Habits are assembled routines that link sensory cues
Tourette's syndrome (TS) is a chronic neuropsychi- with motor action. They allow us to act without think-
atric disorder of childhood onset that is characterized ing, as in riding a bicycle or driving a car. As such, they
by tics that wax and wane in severity and by an array of are enormously adaptive and part of a common evolu-
behavioral problems that include some forms of obses- tionary heritage that we share with other vertebrates
sive compulsive disorder (OCD) and attention deficit and probably with all other species capable of goal-
hyperactivity disorder (ADHD) (American Psychiatric directed behavior. Repetitive sensory-cognitive-emotive-
motor ensembles, as occur when we witness a gifted
pianist playing a sonata or a surgeon suturing, have
JAMES F. LECKMAN Child Study Center; Children's Clinical
Research Center; Departments of Pediatrics, Psychiatry, and
habitual elements. When we do things over and over
Psychology, Yale University School of Medicine, New Haven, again, we get better at it. There is less thinking about
Connecticut. the action, and we can respond in a more nuanced man-
BRADLEY s. PETERSON Child Study Center, Yale University ner to other internal cues or to perturbations in the
School of Medicine, New Haven, Connecticut. external world. How does the brain manage these mar-
ROBERT T. SCHULTZ Child Study Center, Yale University
vels? Ann Graybiel, following the lead of George Miller,
School of Medicine, New Haven, Connecticut.
DONALD J. COHEN Child Study Center; Departments of Pe- believes that when we do something over and over, our
diatrics, Psychiatry, and Psychology, Yale University School of brains compress the relevant information into "chunks"
Medicine, New Haven, Connecticut. coordinating bits of behavior into action sequences
pamine pathways originating in the substantia nigra Initially, a large array of neurons fired rapidly in re-
pars compacta (SNc) play a crucial role in this learning sponse to the cue. But many fewer fired once the rats
process (Aosaki, Graybiel, and Kimura, 1994). learned the task. Graybiel proposes that the cells that
Ensemble recordings, in which the activity of multi- do fire may represent a chunk of behavior. Further, it
ple striatal neurons are recorded simultaneously, fur- has been proposed that distinct, individually coordi-
ther clarify the role of the striatum and related brain nated, motor and cognitive action chunks are combined
circuits in the learning and production of habitual or and implemented as more elaborate goal-directed be-
"automatic" behavioral responses (Jog et al., 1996). Re- havior. When performed repeatedly, activity in these
cently, Graybiel and her colleagues Yasuo Kubota and ensembles becomes more efficient, and through still
Juan Canales have recorded from ensembles of elec- unknown mechanisms, habits form.
trodes in the sensorimotor areas of rat striatum during Other investigators have focused on the role of
cued learning tasks (personal communication, 1999). the STN as a regulator of striatal output. Given the
CSTC
Component Sensorimotor (SM) Orbitofrontal (OF) Association (AS) Limbic (L)
Cortical afferents Somatosensory Orbitofrontal Dorsolateral Anterior
Primary motor Superior temporal prefrontal cingulate
Supplementary gyrus Posterior parietal Hippocampal
motor area Inferior temporal Arcuate premotor cortex
gyrus Entorhinal cortex
Anterior cingulate Superior
temporal gyrus
Inferior temporal
gyrus
Striatum Dorsolateral Ventral caudate Dorsolateral Ventral caudate
putamen Ventral putamen caudate Ventral putamen
Dorsolateral caudate Nucleus
Dorsolateral accumbens
subthalamic Olfactory tubercle
Nucleus
Pallidum/SNr Ventrolateral GPi Dorsomedial GPi Dorsomedial GPi Rostrolateral GPi
Caudolateral SNr Rostromedial SNr Rostrolateral SNr Ventral pallidum
Rostrodorsal SNr
Thalamus Ventrolateral Medial dorsal Ventral anterior Medial dorsal
nucleus nucleus nucleus nucleus
Centromedian (parvocellular) (parvocellular) (posteromedial)
intralaminar
nucleus
Cortical
projections Supplementary Orbitofrontal Dorsolateral Anterior
motor area prefrontal cingulate
Abbreviations: GPi, globus pallidus, internal segment; GPe, globus pallidus, external segment; SNr, substania nigra, pars retic-
ulata.
Adapted from Peterson et al., 1998a.
excitatory nature of the STN neurons, their threshold, (Bliss, 1980; Leckman, Walker, and Cohen, 1993; Leck-
and peak firing rates, a simple model of neuron re- man et al., 1994; Miguel et al., 1997). These perceptual
sponses has revealed that large regions of this highly cues include faint premonitory feelings or urges that
interconnected nucleus respond to excitatory input in are relieved with the performance of tics (figure 35.2),
the form of a widespread uniform pulse (Gillies and and a need to perform tics or compulsions until they
Willshaw, 1998). These pulses of activity may act as a are felt ineffably to be "just right." Although the neural
braking signal that resets the basal ganglia output nu- mechanisms that conspire to produce tics have yet to
clei, thus interrupting habits and setting the stage for be elucidated, preliminary evidence suggests that they
the emergence of new goal-directed emotive-cognitive- involve the same structures that underlie habit
behavioral sets. formation.
Tics as fragments of normal behavior'? Tics and the disregulation of the neural circuits
underlying habit formation
If habits are coordinated chunks of thought and action,
then conceptually tics may be best seen as those pre- The basal ganglia and the neural loops that they par-
wired bits of behavior that are available to be chunked ticipate in have long been a focus of TS research (Bal-
together to produce habits (Leckman and Riddle, thasar, 1956; Peterson et al., 1993; Young and Penney,
2000). Like habits, tic action sequences often arise from 1984). Advances in neuroimaging and neurophysiolog-
a heightened and selective sensitivity to environmental ical techniques have made it possible to examine the
cues from within the body or from the outside world activity of these circuits in living subjects.
CSTC CIRCUITS In TS there is preliminary evidence fMRI techniques have been developed that will greatly
that voluntary tic suppression involves activation of enhance the temporal resolution of these studies
regions of the prefrontal cortex and the right caudate (Bruckner et al., 1996). Work in progress suggests that
nucleus and the bilateral deactivation of the putamen it should be possible to monitor the progression of in-
and globus pallidus (figure 35.3; see color plate 24; Pe- dividual tics in the magnet. These studies should permit
terson, Skudlarski, Anderson et al., 1998b). If repli- investigators to begin to define the temporal sequence
cated, these findings are consistent with the well-known of activity within different portions of these cortical-
finding that chemical or electrical stimulation of inputs subcortical loops. In this regard, it will be intriguing to
into the putamen can provoke motor and vocal re- study the involvement of the supplementary motor area
sponses that resemble tics (MacLean and Delgado, (SMA; table 35.1, figure 35.1) as electrical stimulation
1953). They also suggest that prefrontal cortex—basal of the SMA elicits a variety of bodily sensations that in-
ganglia circuits participate in shaping the inhibitory in- clude a premonitory sensations or "urges" to perform
fluence of the output neurons in the internal segment a movement or a sense of anticipation that a movement
of the globus pallidus and the pars reticulata of the sub- is about to occur (Fried et al., 1991).
stantia nigra.
Most functional magnetic resonance imaging (fMRI) DOPAMINERGIC MECHANISMS Although controversial,
studies to date have employed a block design in which age-dependent changes in central dopaminergic path-
the activation/deactivation signal reflects a presumed ways that innervate the basal ganglia may also be char-
continuous mental state. More recently, event-related acteristic of many children with TS (Ernst et al., 1999).
Developmental shifts in the balance of tonic-phasic gests that the most consistently observed deficits occur
dopaminergic tone are likely and may influence the nat- on tasks requiring the accurate copy of geometric de-
ural history of TS (Grace, 1995; Seeman et al., 1987). signs, i.e., "visual-motor integration" or "visual-graphic"
Although future ligand-based functional imaging stud- ability. The ability to copy simple designs accurately has
ies in child and adolescent samples, complemented by been investigated in 12 studies, excluding case reports
neuropathological studies, hold considerable promise (see Schultz et al., 1998b, for a review). For example,
to elucidate these mechanisms, ethical concerns and we studied the Beery test of visual-motor integration
logistic difficulties may limit these avenues of investi- (VMI; Beery and Buktenica, 1989) performance of 50
gation, which in turn points to the need to develop suit- children with TS (34 TS with ADHD, and 16 TS without
able animal models for TS and related disorders. ADHD) and 23 age-matched controls (Schultz et al.,
1998a). The VMI requires subjects to copy simple de-
Neuropsychological findings implicate brain signs as accurately as possible. Consistent with prior
regions involved in visual-motor integration studies, we obtained significant group differences; chil-
dren with TS scored approximately 1 standard deviation
Although motor and phonic tics constitute the core ele- (SD) below the controls on the VMI. We found no evi-
ments of the diagnostic criteria for TS, perceptual and dence to suggest that comorbid ADHD or depressive
cognitive difficulties are also common. These neuropsy- symptomatology could account for the observed group
chological symptoms are potentially informative about differences.
the pathobiology of the disorder. Moreover, these as- As part of this study, a four-component process model
sociated difficulties can be more problematic for school of visual-motor integration (fine motor skill, visual-
and social adjustment than the primary motor symp- perceptual skill, response inhibition, and sustained at-
toms (Dykens et al., 1998; Walkup et al., 1998). tention) was also assessed. Three putative component
Neuropsychological studies of TS have focused on a processes—fine motor skill, visual-perceptual ability,
broad array of functions. Review of the literature sug- and response inhibition—were also significant areas of
FIGURE 35.4 First return maps. (A) and (B): Probability den- in both images, indicate a strong tendency for the periodicity
sity maps for inter-tic intervals recorded for single facial tics if intermediate duration inter-tic intervals. Very short inter-tic
monitored in two men with TS aged 36 and 30 years. Each intervals are typically followed by either very short inter-tic
videotaped recording was 3 hours in length and was uninter- intervals, labeled "b" in both images, or by very long inter-
rupted. Inter-tic intervals (TI(t)) on the x-axis have been di- tic intervals, labeled "c" in both images. Reciprocally, very long
vided into five equal-sized bins (quintiles). The y- axis plots the inter-tic intervals are typically followed by very short inter-tic
very next inter-tic interval (TI (t+1) ). If the succession of inter- intervals, labeled "d" in both images. The remaining regions
tic intervals were random, these joint probability density maps "e" and "f" indicate regions where the joint probability is low.
would be of uniform color. Islands of positivity, labeled "a" For further details, see Peterson and Leckman (1998).
ABSTRACT The goal of the chapter is to explore different that attention emerges out of the synchronized activity
meanings of attention in the context of two neurodevelop- of large-scale neuronal networks distributed across the
mental disorders. Although it has been conceptualized as a whole brain (Dehaene, Kerszberg, and Changeux,
unitary construct, attention refers to a variety of cognitive pro-
cesses. The chapter begins with an overview of research on 1998; Lumer, Edelman, and Tononi, 1997; Niebur and
three commonly used meanings of attention in the cognitive Koch, 1998).
neuroscience literature: selective, executive, and sustained at- Other chapters in this volume (Johnson, Mareschal,
tention. Next, a brief summary of research on these aspects of and Csibra; Rothbart and Posner; Richards) address at-
attention in schizophrenia and attention-deficit/hyperactivity tention and its development in healthy individuals. In
disorder (ADHD) is presented. The final section concludes
with a consideration of potential contributions of cognitive
this chapter, we explore the different meanings of at-
neuroscience to an understanding of disorders of attention tention within the context of two developmental disor-
and potential contributions of research on developmental dis- ders. The first section is a brief overview of three
orders of attention to an understanding of the cognitive and common meanings of attention in cognitive neurosci-
neurobiological substrates of attentional development. ence: selective, executive, and sustained attention (Par-
asuraman, 1998; Perry and Hodges, 1999). All three
The first point to make about attention is that there types of attention are mediated by widespread neuronal
is no one definition of attention. As Treisman (1998) networks that are vulnerable to many kinds of damage
has noted, "Attention need not be a unitary process sim- and dysfunction. Therefore, it is not surprising that at-
ply because a single word is used in everyday language" tentional impairments play a significant role in many
(p. 1305). What researchers mean by attention is inevi- psychiatric disorders. The discussion that follows sum-
tably shaped by their perspective, the methods and pop- marizes research on two of these disorders—schizo-
ulations they are working with, and the specific aspect phrenia and attention-deficit/hyperactivity disorder
of attention they are studying. (ADHD)—and considers potential contributions of the
As a behavioral descriptor, attention can mean being cognitive neuroscience of attention to research on
able to sustain a high level of effort on cognitively de- these and other disorders of attention. Missing to some
manding tasks (e.g., homework), refraining from being extent in both of these research literatures is a genu-
distracted by irrelevant stimuli (e.g., a noise in the back- inely developmental perspective. Thus, the chapter con-
ground), and attending to inappropriate stimuli (e.g., cludes by highlighting developmental questions that
internal or intrusive thoughts, hallucinations). can be investigated in disorders of attention and their
In cognitive neuroscience, some researchers concep- potential for illuminating our understanding of the cog-
tualize attention as an enhancement of processing of nitive and neurobiological substrates of the develop-
relevant stimuli (Posner and Dehaene, 1994), whereas ment of attention.
others view it as akin to an executive controlling differ-
ent parts of the brain (Baddeley, 1996) or a pool of
resources (Kahneman, 1973). Neuropsychological stud-
ies of patients with frontal lesions emphasize the role Meanings of attention in cognitive neuroscience
of the frontal lobes in attention (Swick and Knight,
1998), whereas studies of patients with subcortical le- SELECTIVE ATTENTION The first meaning of attention
sions highlight the interactions of the frontal lobes with refers to selectivity, the ability to attend to relevant stim-
the basal ganglia, the thalamus, and the cerebellum uli and ignore irrelevant stimuli. Selective attention can
(Dubois et al., 1995). Finally, some researchers suggest be controlled in a bottom-up (e.g., by physically salient
stimuli such as a bright light) or a top-down manner
CANAN KARATEKIN Institute of Child Development, Univer- (e.g., by expectations and goals that determine what is
sity of Minnesota, Minneapolis, Minnesota. "relevant").
ABSTRACT In this chapter we explore ways in which the neu- developmental aspects of schizophrenia. We suggest
ropsychological profile of schizophrenia, namely problems in that the core cognitive deficits in schizophrenia have
attention, working memory, episodic memory, and semantic specific physiological correlates and thus provide cru-
memory, can inform us of the neurodevelopmental trajectory
of the illness. A large number of studies have demonstrated cial clues about the neurodevelopmental trajectory of
that schizophrenia is associated with neurocognitive weakness the illness. Second, we ask why brain development is of
which is present prior to the onset of the illness and probably interest in schizophrenia research. We argue that there
is due to subtle development abnormalities in critical circuitry. is compelling evidence that schizophrenia is associated
One of the challenging questions is why abnormal brain de- with neurocognitive weakness that is present prior to
velopment might result specifically in schizophrenia rather
than some other neurodevelopmental disorder. We suggest
the onset of the illness. This is most probably due to an
that a combination of age and specific sites of brain insult underlying abnormality of cortical development. Third,
(especially the temporal-limbic and dorsolateral prefrontal we ask why any abnormality of brain development
cortices) produces a unique neurodevelopmental trajectory might result specifically in schizophrenia rather than
that leads to the specific neurocognitive profile associated with some other neurodevelopmental disorder. We present
schizophrenia.
data demonstrating that it is possible for early brain in-
jury involving the neural circuitry implicated in schizo-
Schizophrenia is a devastating human illness involv-
phrenia to have substantial and long-lasting effects on
ing changes in perception, cognition, and comport-
specific and also syndromally relevant neural systems
ment. Research in unraveling the mystery of this
that are quite distant from the site of original dam-
disorder has met with surprisingly limited success, de-
age. Moreover, we contend that since the cognitive
spite considerable commitment of people and money.
deficit profiles of schizophrenia are qualitatively dis-
Recently, the research community has shown increasing
tinct from other disorders, it is highly probable that
enthusiasm for the notion that neuropsychological def-
there is a unique neurodevelopmental profile asso-
icits are core features of the illness that arise from ab-
ciated with schizophrenia.
normalities in the development of the brain and that
predict long-term disability and outcome. In this chap-
What compromised neurocognitive systems
ter we address three questions that are fundamental to
the exploration of the nature of neuropsychological def- are implicated in schizophrenia ?
icits in patients with schizophrenia, and particularly The cardinal symptoms of schizophrenia are abnormal
how the neurodevelopmental model of schizophrenia ideas (such as delusions); abnormal perceptions (such
might shed light on the neuropsychological course of as hallucinations); formal thought disorder (as evi-
the schizophrenia illness. First, we ask what affected or denced by disorganized speech); motor, volitional, and
compromised neurocognitive systems are implicated in behavioral disorders; and emotional disorders (such as
schizophrenia, and what this can tell us about neuro- affective flattening or inappropriateness) (American
Psychiatric Association, 1994). In addition to this ex-
BRITA ELVEVAG Clinical Brain Disorders Branch, National ceedingly wide range of symptoms and behaviors, it has
Institute of Mental Health/National Institutes of Health,
Bethesda, Maryland.
become increasingly apparent that schizophrenia is, to
DANIEL R. WEINBERGER Chief, Clinical Brain Disorders varying degrees, accompanied by a broad spectrum of
Branch, National Institute of Mental Health/National Insti- cognitive deficits (for a review, see Elvevag and Gold-
tutes of Health, Bethesda, Maryland. berg, 2001). If the cognitive deficits of schizophrenia
Affective disorder A case could be made that, from a cog- HUNTINGTON'S DISEASE Both Huntington's disease
nitive perspective, affective disorder is just a muted and schizophrenia have been considered to be subcor-
form of the schizophrenia illness, and that neuropsy- tical disorders with cognitive slowing and executive-type
chological differences between groups might deter- problems. Nonetheless, the cognitive profiles of these
mine, to some extent, the outcome in these disorders— two groups are in fact quite distinct. In a direct com-
schizophrenia being characterized by a chronic course parison of patients, Goldberg and colleagues (Goldberg
and affective disorder being more episodic and less dis- et al., 1990) found that patients with schizophrenia per-
abling. We suggest that these disorders do present with formed better on visuospatial tasks after the two groups
different profiles, in that the cognitive deficit profile is had been matched on a putative measure of frontal lobe
not only more severe in patients with schizophrenia function (the WCST), a finding taken to suggest that
compared to bipolar or depressed patients but also patients with Huntington's disease and those with
qualitatively different (for a review, see Goldberg, 1999; schizophrenia present with different cognitive perfor-
but see an alternative view by Zihl, Gron, and Brun- mance profiles.
nauer, 1998). In comparisons with patients with affec-
tive disorders, patients with schizophrenia have been Alzheimer's disease Comparison of patients with Alzhei-
shown to have worse performance on a degraded ver- mer's disease with schizophrenic patients have dem-
sion of the vigilance task—the Continuous Perfor- onstrated that the former group are more impaired
mance Task (Nuechterlein et al., 1991)—a decline in cognitively, especially as the Alzheimer's disease pro-
IQ from premorbid levels, and decreased visual mem- gresses. As discussed earlier, the cognitive courses of the
ory and executive function (Goldberg et al., 1993a; illnesses are very distinct, the cognitive declines being
Gourovitch et al., 1999). Such evidence is consistent small in schizophrenia relative to that of Alzheimer's
with the observations that functional outcomes are disease. A study by Heaton and colleagues (Heaton et
worse for patients with schizophrenia (Marneros, al., 1994) found that patients with schizophrenia did
Rohde, and Deister, 1995, 1998). Importantly, since af- not show an increased rate of forgetting, unlike the pa-
fective illness is more episodic, it is also less disabling, tients with Alzheimer's disease.
with patients generally recovering their cognitive
function once an episode has resolved (i.e., state- Attention deficit/hyperactivity disorder One of the main
dependent). This is quite unlike the cognitive deficits cognitive deficits observed in both schizophrenic
ABSTRACT New laboratory research has begun to reveal a tions, there is evidence that attachment experiences in
network of simple behavioral, physiological, and neural pro- infancy strongly affect characteristics of the attachment
cesses that underlie the psychological constructs of attach- pattern of the mother with her infant in the next gen-
ment theory. It has become apparent that the unique features
of early infant attachment reflect certain unique features of eration. Several extensive reviews of this field have re-
early infant sensory and motor integration, early learning, cently been published, covering human and nonhuman
communication, motivation, and the regulation of biobehav- primate research (Cassidy and Shaver, 1999; Goldberg,
ioral systems by the mother-infant interaction. This chapter Muir, and Kerr, 1995; Suomi, 1997) and the growing
is organized around the three major questions that gave rise field of neurobiological studies (Carter, Lederhandler,
to the concept of attachment: How does the infant find its
own mother and stay close to her? Why does separation of the
and Kirkpatrick, 1997; Insel, 1992; Kraemer, 1992; Wil-
infant from its mother produce such severe physiologic and son and Sullivan, 1994).
behavioral responses? How can individual differences in adult It is curious that most of the research on attachment
offspring and especially in their maternal behavior be related subsequent to Bowl by's landmark volume defining the
to the patterns of their early life with their parents? In each field (Bowlby, 1969) has focused on the different pat-
of these cases, we review the recent research that has given us
new answers to these questions at the level of early behavioral
terns of early attachment and their later developmental
and cognitive processes and their neurobiological substrates. correlates, leaving far less studied the developmental
Attachment remains useful as a concept, like hunger, that de- processes through which attachment is initially formed
scribes the output of subprocesses that work together within in altricial (slow-developing) mammals and the behav-
the frame of a vital biological function. ioral mechanisms underlying the dramatic responses of
infants to separation. Meanwhile, others interested in
The word attachment has come to refer to a broad
processes underlying the early development of motor
range of behavioral processes and mental states unified
and sensory systems, perception, attention, learning,
by a single central concept. Attachment theory envis-
memory, communication, motivation, and emotion
ages a unique motivational system with evolutionary
have developed new methods (Shair, Barr, and Hofer,
survival value, on a par with hunger and sex, that
1991) and a knowledge base in each of these areas
is organized to maintain physical proximity to the
(Michel and Moore, 1995) without relation to the con-
mother/caretaker soon after birth and psychological
cept of attachment or its field of research. To these re-
proximity, or closeness, later in development (Bowlby,
searchers, the range of associated phenomena did not
1969). The special qualities of this bond reflect the na-
appear to function as a unitary system and attachment
ture of the infant's and the caretaker's mental repre-
concepts were too global to generate incisive research
sentations of the behavioral dynamics and patterns of
questions that could lead to a deeper understanding of
their many previous interactions. Much current re-
search is concerned with identifying developmental the phenomena described.
Now that a knowledge base has been established in
continuities between individual differences in the pat-
the various research areas just described, at least for one
terns of parent-infant attachment and the later cogni-
mammalian species, the laboratory rat, we can revisit
tive and emotional characteristics of the child and
adult. In an extension of these longitudinal correla- the basic observations on which attachment theory is
based, and describe some of the underlying processes
responsible for them in terms that can be related to
MYRON A. HOFER Department of Developmental Psychobi- neural structure and function. When one does this, it
ology, New York State Psychiatric Institute; Department of Psy-
chiatry, Columbia University, New York, New York. becomes apparent that the unique features of early at-
REGINA M. SULLIVAN Department of Zoology, University of tachment phenomena can be shown to be the result
Oklahoma, Norman, Oklahoma. of certain unique features of early sensory and motor
Moore (Moore, Jordan, and Wong, 1996) and Elliot factory bulb anatomical changes reflected in enlarged
Blass (Pillion and Blass, 1986) demonstrated that adult glomemli within these foci (Woo, Coopersmith, and
male rats exhibited enhanced sexual performance Leon, 1987). As with the behavioral changes in attach-
when exposed to the natural and artificial odors ment, these neural changes are retained into adult-
learned in infancy. These results support observations hood, but their acquisition is dependent upon
on the role of early experience on adult mate prefer- experiences during infancy (Pager, 1974; Woo and
ence in other species, such as occurs in imprinting. Leon, 1988).
Many neurotransmitters have a role in early olfactory
NEURAL PLASTICITY UNDERLYING OLFACTORY PREF- learning in neonatal rats—5-HT (McLean et al., 1993),
ERENCE LEARNING The development of a specific DA (Barr and Wang, 1992; Weldon, Travis, and Ken-
olfactory-based attachment system in the rat pup dur-
nedy, 1991), GABA (Okutani, Yagi, and Kaba, 1999),
ing the first week and a half of life is associated with the
glutamate (Lincoln et al., 1988; Mickley et al., 1998),
acquisition of olfactory bulb neural changes. We found
and opiates (Barr and Rossi, 1992; Kehoe and Blass,
that rat pups express this modified olfactory bulb re-
sponse to both natural maternal and artificial odors ex- 1986). Roth and Sullivan (2000) have recently shown
perienced in the nest (Sullivan et al., 1990), as well as that PN8 pups treated with systemic naltrexone, a non-
to odors in controlled learning experiments (Johnson specific opioid antagonist, do not show a subsequent
et al., 1995; Sullivan and Leon, 1986; Wilson and Sulli- odor preference after odor-shock classical condition-
van, 1991; Wilson, Sullivan, and Leon, 1987). The mod- ing, suggesting an important role for opioids in early
ified olfactory bulb response is characterized by preference formation to noxious stimuli. However, the
enhanced immediate-early gene activity (c-fos) and en- action of norepinephrine (NE) appears particularly
hanced 2-deoxyglucose (2-DG) uptake in focal, odor- important in neural plasticity during early develop-
specific glomerular regions in response to the ment and in the form of olfactory (Brennen and
conditioned odor. Within the underlying neural sub- Keverne, 1997; Wilson and Sullivan, 1994) and so-
strate, modified single-unit response patterns of mi- matosensory (Landers and Sullivan, 1999b) learning-
tral/tufted cells near the enhanced glomerular foci induced plasticity used in early attachment.
were found (Wilson and Leon, 1988b; Wilson and Sul- Norepinephrine input to the olfactory bulb is wide-
livan, 1990; Wilson, Sullivan, and Leon, 1987) and ol- spread throughout the granule cell layer as early as the
FIGURE 38.2 The adult olfactory pathway. Although many of learning, their role in infant learning has not been deter-
these brain areas have been implicated in adult olfactory mined.
ABSTRACT This chapter briefly summarizes the state of our United States, Canada, and Western Europe. Because
current understanding of the development of children from these children offer a dramatic case in point, in consid-
deprived, orphanage backgrounds, considering, where rele- ering the impact of early deprivation in this chapter, I
vant, the results of animal studies. Three levels of deprivation
are contrasted that differ in whether health and nutrition,
will focus on information obtained in studies of or-
social and physical stimulation, and relationship needs are phanage-reared children—in particular, those fostered
met. Emphasis is placed on relating the relatively large liter- or adopted into families after spending some or all of
ature on deprivation and behavioral development to the more their first two years in institutions. The term "orphan-
limited literature on deprivation and neurobiological devel- age," while evocative, is a misnomer, as most of these
opment in humans and other primates. children are not orphaned, but abandoned or removed
from families who cannot care for them adequately. As
The effects of early deprivation have seen renewed
yet, there have been few neurobiological studies of
interest recently. As will be discussed, deprivation can
these children (Chugani et al., 2001). Nonetheless,
be multifaceted, involving various failures in the care-
there is a long history of behavioral research on or-
giving environment: failure to provide adequate nutri-
phanage rearing. There is also a related body of non-
tion, medical care, stimulation, and/or enduring and human primate research, some of which includes
supportive relationships. Such failures constitute ne- neurobiological data (Kraemer et al., 1989). This ear-
glect, and there is a mounting belief that neglect or lier work should serve as a starting point for future neu-
deprivation may confer as significant a risk for psycho- robiological studies. With this in mind, the goal of this
pathology as physical and sexual abuse (National Re- chapter is to briefly summarize the state of our current
search Council, 1993). Our renewed interest and understanding of the development of children from de-
deepening concern about the effects of early depriva- prived, orphanage backgrounds, considering, where
tion comes at a time when the tools of neuroscience relevant, the results of animal, and in particular, pri-
offer promise that we may be able to understand dep- mate studies.
rivation effects in ways not previously possible. None-
theless, the impact of early deprivation is difficult to
History of orphanage-rearing studies
isolate unless the period of deprivation is circum-
scribed. In most cases, deprivation early in life is fol- and conceptual issues
lowed by life events that also threaten healthy physical, Current studies of orphanage-reared children focus on
emotional, and cognitive development, making it diffi- the wave of Romanian children adopted into families
cult to disentangle early from later threats to normal following the fall of the Romanian communist regime
brain development. Children adopted from orphan- in the early 1990s (Benoit et al., 1996; Groze and Ileana,
ages offer a dramatic contrast; their early deprivation is 1996; Morison, Ames, and Chisholm, 1995; Rutter,
typically followed by development within often en- 1998). Future studies are likely to deal with the thou-
riched environments (Rutter, 1981). The last decade sands of children arriving yearly from orphanages in
also has seen an increasing flow of these children from Russia, the former Soviet Republics, Eastern Europe,
orphanages in the developing world and Eastern Eu- and China (Albers et al., 1997). The work on the Ro-
rope into middle- and upper-middle class families in the manian children adds to information from research on
orphanage rearing that spanned the decades of the
MEGAN R. GUNNAR Institute of Child Development, Univer- 1930s through the 1970s (Rutter, 1981). This work is
sity of Minnesota, Minneapolis, Minnesota. credited with the closing of most orphanages in western
ABSTRACT Temperament refers to a style or pattern of be- the work of Gray. Gray (1982, 1987) articulated a
havior that characterizes young infants' ongoing interaction motivation-based view of temperament and proposed
with the world. Individual differences in this behavioral style three fundamental motivation systems, each denned by
are thought to reflect biologically based dispositions (Gold-
smith et al., 1987). The precise nature of these biological dif-
a set of behavioral input-output relations and each as-
ferences has, however, only recently been investigated. In part, sociated with a particular subsystem in the brain. These
current research on the biological bases of temperament three systems are the behavioral inhibition system, the
stems from an articulation of models of individual differences fight-flight system, and the behavioral approach sys-
that emphasize the role of the nervous system in the expres- tem. There are a number of comprehensive reviews of
sion of behavioral styles (Kagan, 1994; Rothbart and Derry-
berry, 1981). This work reflects a developing synthesis of
the comparative behavioral evidence and the underly-
research in the neurosciences and developmental psychology ing neural bases for each system (Gray, 1991). Much of
toward an understanding of the biology of behavior. This the behavioral work examining Gray's model and its re-
chapter provides the reader with a view of the different models lation to temperament/personality has centered on the
of temperament that have guided the field of developmental use of Eysenck's model of personality (Eysenck and
research. We present our own conceptualization of infant tem-
Eysenck, 1985). This model makes specific predictions
perament, one that focuses on the concepts of reactivity and
regulation but emphasizes as well the overlay of emotional about the susceptibility to conditioning and arousal of
valence in the expression of individual differences in reactiv- introverts and extroverts. Gray's model predicts the ef-
ity. The final section of this chapter presents a brief review of ficacy of certain types of reinforcement for introverts
the work on behavioral inhibition as an example of the new and extroverts.
synthesis of biological and behavioral research toward an un- Mary Rothbart expanded on the ideas of Pavlov and
derstanding of individual differences in temperament.
the Eastern European personality psychologists as well
as the notions of activation and inhibition of Gray to
Psychobiological models of temperament articulate a psychobiological model of infant tempera-
ment. Rothbart's theory of temperament has as its foun-
The notion of a biological basis to individual differences dation the notion that infants differ early in life in the
in personality appears as early as the Greek philoso- manner in which they respond to sensory stimulation
phers, with their division of temperament into types (Rothbart and Derryberry, 1981). In addition, infants
(Kagan, 1994). The twentieth-century neurophysiolo- differ in their ability to return to homeostasis following
gist Pavlov noted differences in speed of conditioning a reactive response.
and suggested that these effects were a result of individ- Growth of interest in individual differences in infant
ual differences in strength of the nervous system (Pav- temperament has been paralleled by research in the
lov, 1927). Eastern European personality psychologists development of emotions as well as individual differ-
subsequently adopted the notion of strength of the ner- ences in expression and control of emotion (Fox,
vous system and constructed typologies of personality 1994a). Goldsmith and Campos (1986), for example,
based upon this notion of individual differences (e.g., posit that temperament may be viewed as individual dif-
Strelau, 1983). ferences among infants in the expression of specific
A second influential stream of work on the biology emotions in reaction to specific contexts. For example,
of individual differences in personality emerged from some infants may be predisposed to respond with the
NATHAN A. FOX, HEATHER A. HENDERSON, AND PETER J.
emotion of fear to novelty or uncertainty. The work on
MARSHALL Department of Human Development, University behavioral inhibition has utilized an emotion-based
of Maryland, College Park, Maryland. temperament approach toward understanding the
ABSTRACT Much of human behavior is organized around in neuropsychological assessment that when motivation
stimuli that we conceptualize as worthy of approach (e.g., posi- is insufficient, individuals demonstrate inconsistent pat-
tive reinforcers) or emotional attachment (natural rewards). terns of performance, sometimes doing better on more
Two separable emotional systems have been described that
contribute to these reward-related processes. One of these is
difficult (presumably more demanding) task items or
a system involved in approach to distal rewards, and the other otherwise demonstrating a performance pattern that is
mediates consummatory aspects of reward acquisition. In ad- not easily accounted for by a task's cognitive demands
dition to their emotional functions, each of these systems is (Lezak, 1983). In most instances of adult assessment, it
proposed to guide specific cognitive processes. Spatial work- is a fundamental assumption that a test taker's motiva-
ing memory is viewed as the cognitive extension of an emo-
tional system underlying anticipatory reward and approach to
tional resources are appropriately engaged and that
incentive stimuli. Both aspects of behavior share a common task performance can be attributed more or less solely
neurobiological substrate in the functioning of ascending to cognitive abilities. That is, there is often an assumed
brain dopamine systems. In contrast, the type of affective state equivalence between performance and overall ability.
that accompanies reward acquisition is consummatory in na- When we test children, psychiatric patients, or individ-
ture and serves to keep the organism close to objects and uals with neurological injuries, this basic assumption
places of attachment. When this system is activated, behaviors
promoted by the approach system, including spatial working may be particularly flawed because integration between
memory, are inhibited. The consummatory system is medi- the cortical areas that perform distinct cognitive com-
ated, at least in part, by endogenous opiates. In this chapter, putations and other brain regions that mediate intrinsic
the implications of a DA-based incentive system and an opiate- motivational states and sensitivity to external rewards is
based consummatory attachment system are discussed in re- not maximally adaptive.
lation to the developmental assessment of spatial working
memory skills as exemplified by Piaget's A-not-B task. Specif-
The general objective of this chapter is to consider
ically, it is proposed that young infants are predominantly cognition from this expanded behavioral perspective,
guided by an opiate-based attachment system, leading to the by emphasizing the motivational conditions under
A-not-B error and conditioned place preferences when they which it occurs, and using spatial working memory as
perform spatial memory tasks. Hence, the resolution of the A- an exemplar. Herein, spatial working memory is under-
not-B error may represent an affective, rather than a cognitive, stood as a component of dopamine-driven reward-
milestone reflecting increasing maturity of brain dopamine
systems underlying agency and approach to positive stimuli. seeking behavior and will be discussed in that context
(Depue and lacono, 1989; Depue and Collins, 1999; De-
One critical challenge in deciphering what is re- pue et al., 1994; Luciana, 1994; Luciana and Collins,
flected in discrete measures of cognitive skill is to ac- 1997; Luciana et al., 1992). I will begin by describing
count for the manner in which motivational variables the phenomenological structure of reward-seeking
can influence performance. It is a longstanding tenet behavior and dopamine's role in promoting positive in-
centive motivation. Cognitive extensions of this emo-
MONICA LUCIANA Department of Psychology, University of tional repertoire, including spatial working memory
Minnesota, Minneapolis, Minnesota. and its modulation by prefrontal dopamine, will then
s/he will once again become reactivated to begin the the object is hidden but must inhibit prepotent re-
search anew. Neuromodulatory systems would be de- sponding in order to correctly guide that memory. This
signed such that activity in one branch of this overall review extends Diamond's observations by offering an
reward system is accompanied by decreased activity in explanation of the affective processing mechanism un-
the other branch, and that as activity in one branch was derlying the infants' prepotent responding. When the
decreasing, the other would begin a gradual ascent. adult encounters a reversal contingency on the Wiscon-
However, if one or the other piece of this overall reward sin Card Sort, s/he presumably does not have other
system is "locked in" or otherwise biased, as may be the reward associations that have been guiding his/her
case within discrete developmental periods, this bal- responses. A perseverative error that occurs after one
ance of behavior will be altered to favor one repertoire has been clearly told "Wrong" is particularly hard to
or the other. fathom. Similarly, it has been assumed that when the
With respect to cognition, organisms in a high-DA infant makes the A-not-B error, s/he has behaved illog-
state will be pulled by search processes that maximize ically. The perseverative response is viewed as an autom-
novelty and demand representational thought for an- atism (Diamond, 1988). However, what has perhaps
ticipated events. On the other hand, if the organism is been overlooked is that the infant may be drawn by a
dominated by a high attachment-consummatory state, legitimate reward association—one that, until correctly
s/he will perform better on tasks that maximize re- updated, links the desired object with a specific loca-
sponses to familiarity, habituation, and contact comfort. tion. The observation that the AB error occurs only at
This state is one of inherent contentment and well- increasingly long delay intervals depending on the age
being, whereby behavioral activation is at a relative low. of the child might again relate to how behavior is mod-
With respect to DA-modulated behaviors, individuals in ulated under conditions of uncertainty, and the re-
this state would perform relatively poorly. It is suggested sponse of the developing DA system under such
that infants are affectively biased toward attachment- conditions. A delay interval, however small, puts reward
consummatory activation and that, until the DA system acquisition into question. If the DA-modulated seeking
is more mature, this bias has predictable implications system is immature, cellular firing patterns in limbic
for A-not-B task performance. and cortical structures that hold incentive-guided mo-
tor behaviors on-line will fail. Under such conditions,
The A-not-B error one's reward-respondent behavior might be biased by
and consummatory-attachment behavior attachment-based principles (Bowlby, 1982). On a neu-
ral level, this would mean that the neurochemical
The A-not-B error has been described as a perseverative environment is more conducive to attachment-
response, similar to that made by frontal patients on the consummatory behavior, which as described above, may
Wisconsin Card Sort Task or other measures that re- increase the likelihood that responses will be condi-
quire shifting of response sets in response to external tioned as the task unfolds to discrete locations.
feedback. Diamond (1990; this volume) discusses these Infants' performance on the A-not-B task can be re-
tasks in terms of their requirements for inhibitory con- interpreted from this framework, whereby the gradual
trol. Her argument is that in order to be correct on each elimination of the A-not-B error is a marker of an
A-to-B trial, the infant must not only remember where affective milestone in which the infant transitions from
RICHARD N. ASLIN, PH.D. Department of Brain and Cog- GERGELY CSIBRA, PH.D. Centre for Brain and Cognitive
nitive Sciences, University of Rochester, Rochester, New Development, Department of Psychology, Birkbeck College,
York University of London, London, England
JOCELYNE BACHEVALIER, PH.D. Department of Neurobiol- JAMES L. DANNEMILLER, PH.D. Department of Psychology,
ogy and Anatomy, University of Texas Health Science Cen- University of Wisconsin at Madison, Madison, Wisconsin
ter, Houston, Texas MICHELLE DE HAAN, PH.D. Cognitive Neuroscience Unit,
ELIZABETH BATES, PH.D. Center for Research in Lan- Institute of Child Health and Great Ormond Street Hospi-
guage, University of California at San Diego, La Jolla, tal, University College London Medical School, The Wolf-
California son Centre, London, England
GAIL C. BEDI, PH.D. Director, Manhattan Neuropsychol- ADELE DIAMOND, PH.D. Director, Center for Developmen-
ogy; Assistant Clinical Professor, Mount Sinai School of tal Cognitive Neuroscience, Eunice Kennedy Shriver Cen-
Medicine, New York, New York ter; Professor, Department of Psychiatry, University of
FRANCINE M. BENES, M.D., PH.D. Laboratory for Struc- Massachusetts School of Medicine, Waltham, Massachusetts
tural Neuroscience, McClean Hospital, Belmont, Massachu- THOMAS ELBERT, PH.D. Department of Psychology, Uni-
setts; Program in Neuroscience and Department of versity of Konstanz, Konstanz, Germany
Psychiatry, Harvard Medical School, Boston, Massachusetts BRITA ELVEVAG, PH.D. Clinical Brain Disorders Branch,
J.-P. BOURGEOIS, PH.D. Laboratoire de Recepteurs et Cog- National Institute of Mental Health/National Institutes of
nition, Departement des Biotechnologies, Institut Pasteur, Health, Bethesda, Maryland
Paris, France MONICA FABIANI, PH.D. Department of Psychology, Uni-
JOHN BRUER, PH.D. James S. McDonnell Foundation, St. versity of Missouri, Columbia, Missouri
Louis, Missouri NATHAN A. Fox, Pn.D. Department of Human Develop-
ment, University of Maryland, College Park, Maryland
JUDY L. CAMERON, PH.D. Departments of Psychiatry, Neu-
MICHAEL K. GEORGIEFF, M.D. School of Medicine, Insti-
roscience, and Cell Biology & Physiology, University of Pitts-
tute of Child Development, Department of Pediatrics, Uni-
burgh, Pittsburgh Pennsylvania; Oregon Regional Primate
versity of Minnesota, Minneapolis, Minnesota
Research Center, Oregon Health Sciences University, Bea-
ROBBIN GIBB, M.Sc. Departments of Psychology and Neu-
verton, Oregon
roscience, University of Lethbridge, Lethbridge, Canada
SUSAN CAREY, PH.D. Department of Psychology, New York ELIZABETH GOULD, PH.D. Department of Psychology,
University, New York, New York Princeton University, Princeton, New Jersey
B. J. CASEY, PH.D. Sackler Institute for Developmental Psy- MEGAN R. GUNNAR, PH.D. Institute of Child Development,
chobiology, Joan and Sanford I. Weill Medical College of University of Minnesota, Minneapolis, Minnnesota
Cornell University, New York, New York NICHOLAS B. HASTINGS, PH.D. Department of Psychology,
V. S. CAVINESS, JR., M.D., PH.D. Department of Neurology, Princeton University, Princeton, New Jersey
Massachusetts General Hospital, Harvard Medical School, SABINE HEIM, PH.D. Department of Psychology, University
Boston, Massachusetts of Konstanz, Konstanz, Germany
MELISSA W. CLEARFIELD, B.A. Department of Psychology, HEATHER A. HENDERSON, B.A. Department of Human De-
Indiana University, Bloomington, Indiana velopment, University of Maryland, College Park, Maryland
DONALD J. COHEN, M.D. Child Study Center; Departments MYRON A. HOFER, M.D. Department of Developmental
of Pediatrics, Psychiatry, and Psychology, Yale University Psychobiology, New York State Psychiatric Institute; Depart-
School of Medicine, New Haven, Connecticut ment of Psychiatry, Columbia University, New York, New
PAUL D. CONNER, PH.D. Fetal Alcohol and Drug Unit, De- York
partment of Psychiatry and Behavioral Sciences, University RUSKIN H. HUNT, B.A. Department of Brain and Cognitive
of Washington School of Medicine, Seattle, Washington Sciences, University of Rochester, Rochester, New York
CONTRIBUTORS 663
MARK H.JOHNSON, PH.D. Centre for Brain and Cognitive RAGHAVENDRA RAO, MD Department of Pediatrics, School
Development, Department of Psychology, Birkbeck College, of Medicine, University of Minnesota, Minneapolis,
University of London, London, England Minnesota
CANAN KARATEKIN, PH.D. Institute of Child Development, JOHN E. RICHARDS, PH.D. Department of Psychology, Uni-
University of Minnesota, Minneapolis, Minnesota versity of South Carolina, Columbia, South Carolina
BRYAN KOLB, PH.D. Departments of Psychology and Neu- BRIGITTE ROCKSTROH, PH.D. Department of Psychology,
roscience, University of Lethbridge, Lethbridge, Canada University of Konstanz, Konstanz, Germany
JAMES F. LECKMAN, M.D. Child Study Center; Children's KATHERINE ROE, B.A. Center for Research in Language,
Clinical Research Center; Departments of Pediatrics, Psy- University of California at San Diego, Lajolla, California
chiatry, and Psychology, Yale University School of Medicine, MARY K. ROTHBART, PH.D. Department of Psychology,
New Haven, Connecticut University of Oregon, Eugene, Oregon
PAT LEVITT, Pn.D. Department of Neurobiology, Univer- RICARDO C. SAMPAIO, M.D. Department of Radiology,
sity of Pittsburgh School of Medicine, Pittsburgh, University of Minnesota, Minneapolis, Minnesota
Pennsylvania ROBERT T. SCHULTZ, PH.D. Child Study Center, Yale Uni-
TERRI L. LEWIS, PH.D. Department of Psychology, McMas- versity School of Medicine, New Haven, Connecticut
ter University, Hamilton, Ontario; Department of Ophthal- LASZLO SERESS, M.D., PH.D. Central Electron Microscopic
mology, The Hospital for Sick Children, Toronto; Laboratory, Faculty of Medicine, University of Pecs, Pecs,
Department of Opthalmology, University of Toronto, Szigeti, Hungary
Toronto, Canada GREGG D. STANWOOD, PH.D. Department of Neurobiol-
MONICA LUCIANA, PH.D. Department of Psychology, Uni- ogy, University of Pittsburgh School of Medicine, Pitts-
versity of Minnesota, Minneapolis, Minnesota burgh, Pennsylvania
DENIS MARESCHAL, PH.D. Centre for Brain and Cognitve JENNIFER MERVA STEDRON, B.A. Department of Psychol-
Development, Department of Psychology, Birkbeck College, ogy, University of Denver, Denver, Colorado
University of London, London, England JOAN STILES, PH.D. Department of Cognitive Science, Uni-
PETER J. MARSHALL, PH.D. Department of Human Devel- versity of California at San Diego, Lajolla, California
opment, University of Maryland, College Park, Maryland ANN P. STREISSGUTH, PH.D. Fetal Alcohol and Drug Unit,
DAPHNE MAURER, PH.D. Department of Psychology, Mc-
Department of Psychiatry and Behavioral Sciences, Univer-
Master University, Hamilton, Ontario; Department of Oph-
sity of Washington School of Medicine, Seattle, Washington
thalmology, The Hospital for Sick Children, Toronto,
REGINA M. SULLIVAN, PH.D. Department of Zoology, Uni-
Canada
versity of Oklahoma, Norman, Oklahoma
BRUCE MCCANDLISS, PH.D. Sackler Institute for Develop-
T. TAKAHASHI, M.D. Department of Neurology, Massachu-
mental Psychobiology, Joan and Sanford I. Weill Medical
setts General Hospital, Harvard Medical School, Boston,
College of Cornell University, New York, New York
Massachusetts; Department of Pediatrics, Keio University
CHRISTOPHER S. MONK, PH.D. Institute of Child Devel-
opment, University of Minnesota, Minneapolis, Minnesota School of Medicine, Tokyo, Japan
YUKO MUNAKATA, PH.D. Department of Psychology, Uni- PATIMA TANAPAT, PH.D. Department of Psychology,
versity of Denver, Denver, Colorado Princeton University, Princeton, New Jersey
CHARLES A. NELSON, PH.D. Institute of Child Develop- ESTHER THELEN, PH.D. Department of Psychology, Indi-
ment and Department of Pediatrics, University of Minne- ana University, Bloomington, Indiana
sota, Minneapolis, Minnesota KATHLEEN M. THOMAS, PH.D. Sackler Institute for Devel-
R. S. NOWAKOWSKI, PH.D. Department of Neuroscience opmental Psychobiology, Joan and Sanford I. Weill Medical
and Cell Biology, UMDNJ-Robert Wood Johnson Medical College of Cornell University, New York, New York
School, Piscataway, New Jersey CHARLES L. TRUWIT, M.D. Department of Radiology, Uni-
WILLIAM H. OVERMAN, PH.D. Psychology Department, versity of Minnesota, Minneapolis, Minnesota
University of North Carolina at Wilmington, Wilmington, ATHENA VOULOUMANOS, B.Sc. Department of Psychology,
North Carolina University of British Columbia, Vancouver, British Colum-
SALLY OZONOFF, PH.D. Department of Psychology, Uni- bia, Canada
versity of Utah, Salt Lake City, Utah EMILY WEE, M.A., M.H.S. Department of Psychology, Uni-
BRUCE F. PENNINGTON, PH.D. Department of Psychology, versity of Missouri, Columbia, Missouri
University of Denver, Denver, Colorado DANIEL R. WEINBERGER, M.D. Chief, Clinical Brain Dis-
BRADLEY S. PETERSON, M.D. Child Study Center, Yale Uni- orders Branch, National Institute of Mental Health/Na-
versity School of Medicine, New Haven, Connecticut tional Institutes of Health, Bethesda, Maryland
MICHAEL I. POSNER, PH.D. Sackler Institute for Develop- JANET F. WERKER, PH.D. Department of Psychology, Uni-
mental Psychobiology, Weill Medical College of Cornell versity of British Columbia, Vancouver, British Columbia,
University, New York, New York Canada
664 CONTRIBUTORS
INDEX
Note: Page numbers followed by letters f and t and myelination milestones, 37* aromatase activity in, 63
refer to figures and tables, respectively. and synaptic density, 24-25, 28-29, 82 and attachment, 606, 608, 610
Aggression and attention, 322
A early deprivation/orphanage rearing and, 623- and declarative learning/memory, 111, 369
624 dopamine in, 522
ABR, in auditory assessment, 206-207 effortful control and, 361 early deprivation/orphanage rearing and, 624
Absolute thresholds Aging estrogen receptors in, 59-60
auditory, 206-207 and cognition in fear system, 632-633
visual, 222-223 generalized slowing of function with, 484 in isolation call of infants, 608
Abuse, and attachment, 603, 604 f individual differences in, 474-475 mother-infant interactions and, 610
Accumulator model, of analog-magnitude physical fitness and, 474 nutrition and, 497-498
representations, 421-422 use-it-or-lose-it hypothesis of, 474-475 in processing of emotional expressions, 389
Acetylcholine and EEC, 484 in schizophrenia, 580
and attention, 322-323, 323f, 325 and event-related potentials sex hormones and, 59-61, 63, 72, 138
as neurotransmitter, maturation of, 84, 86 mismatch negativity amplitude, 475-476, sexual dimorphism in, 66
and recovery from early brain injury, 175, 185- 476f-477f and temperament, 632, 640-641
186, 186f 188 novelty P3, 479-481, 481f-482f volume of, developmental changes in, 138
Acetylcholinesterase, developmental changes P300 amplitude and scalp distribution, 479 Analog-magnitude representations, 421-425, 427-
in, 84 and inhibitory control, 479-483 429
Activation functions, in neural network models, and interpretation of brain imaging data, 475 accumulator model of, 421-422
161-162, 162f and memory difficulties with, 422
Activational effects, 60 for contextual detail, 483-484 numerical properties of, 422
of sex steroid hormones, 60-62 frontal cortex activation patterns in, 479, 480/ versus other numerical representations, 422
Acuity Card Procedure, 237 frontal lobe changes and, 473-484 Analytic style, in language development, 303
Acuity, visual. See Visual acuity future directions for study of, 484 Androgen (s). See also Testosterone
Adaptive behavior, fetal alcohol syndrome and, item, 473, 483 aromatization of, to estrogen, 62-63, 62f-63f
511 number of operations to be performed and, 66-67
Adenosine triphosphate, synthesis of, 491 478-479 5a-reductase pathway for, 62f 63-64, 65f
ADHD. See Attention-deficit/hyperactivity disorder sensory, 475-476 metabolism of, and effects on brain, 62-64, 62f-
Adoption studies, 151-152, 152t source, 473-474, 483-484 63f 65f
Adrenal steroids type of material to be processed and, 479 and sexual dimorphism, 66
and neuronogenesis, 97-98, 103 working, 473-484 Androgen receptors, in brain, 60, 138
sexual dimorphism in, 67-69 working memory load and, 476-478 Anemia, and brain development, 498-499
B-Adrenergic receptors, developmental changes methodological issues in studies of, 474-475 Anger, early deprivation/orphanage rearing and,
in, 85 research, use of neuropsychological testing in, 623-624
Adrenocorticotropic hormone, infant levels of, 475 Angular gyrus, development of, 79, 80t
mother-child interactions and, 609-610 and susceptibility to distractors, 479-483 Animal cognition tests, application to children,
Adult(s) A-invisible displacement task, 449t 109-121
aphasia in, with brain injury, 281-282, 315-316 Alcohol caveats to, 120-121
neuronogenesis in, 93-103 prenatal effects of, 505-513 concurrent discrimination, 112-115, 120*
discovery of, in mammalian brain, 94-96, neurodevelopmental consequences, 509-511 declarative learning in, 111-112
94f-95f as teratogen, verification of, 509 delayed-nonmatch-to sample tasks, 111-112,
functional significance of, 100-103 Alcohol-related birth defects, nomenclature for, 120t
hippocampal, 93-103 508-509 inferences about functional development of
new method for detection of, 94f 95-96 Alcohol-related neurodevelopmental disorder neural systems via, 109-121
significance of, 93 (ARND), 508-509 inferences about ontogenesis of sex-specific
reaching skills in, 254, 255f Alerting, attention network of, 354, 355f cognition from, 112-115, 114f
Adult-directed speech, versus infant-directed Alleles, 149 inferences from, summary of, 120t, 120-121
speech, 209 Alzheimer's disease logic of approach, 109-121
Affect cognitive deficits of schizophrenia versus, 587 Morris search task, 115, 117-118, 119f 120
A-not-B task and, 653 prediction of, from earlier functioning, 474 object reversal discrimination task, 112-115,
regulation of, 611 Aminobutyric acid, gamma. See GABA 114f, 120t
cognitive prerequisites for, 638 Ammon's horn oddity task, 120t, 120-121
early deprivation/orphanage rearing and, cell migration in, 49-52, 56 procedural learning in, 111-112
623-626 estrogen receptors in, 60 radial-arm maze, 115-117, 116f 120t
fetal alcohol syndrome and, 506 neurons of tasks developed for monkeys, 109-111
and temperament, 632-634, 636-640 connections of, development of, 52, 57 tests developed for rodents, 115-120
and reward-seeking behavior, 648, 653, 658 formation of, 45-47, 48f-50f 56 verbal instructions for, 110, 119-120
Affective bias, 635-636 principal and nonprincipal, development of, Wisconsin General Testing Apparatus, 109-115,
Affective disorder, cognitive deficits of 52-54, 54f 56 110f ,120
schizophrenia versus, 587 sex hormones and, 68 A-not-B task, 449t
Affective permanence, 652, 654 synapse formation in, 52, 54-56 affective demands of, 653
Age Amputees, cortical plasticity in, 193, 193f 195, components of, 654t
and cognitive deficits in schizophrenia, 585 195f in dopamine-prefrontal cortex studies, 435-440,
and functional outcome of early brain injury, Amygdala 438f-439f 441t, 447-448, 449t, 451f 459,
179, 180f-182f 296-299 androgen receptors in, 60, 138 459f 653-655
INDEX 665
A-not-B task (continued) interactions of, with cognitive seeking, 656- posterior system of, 322, 324, 404
error 657 prefrontal cortex and, 436, 563
attachment-based, 654-655 opiates and, 655-656, 658 as resource, 564-565
and consummatory attachment, 657-658 early deprivation/orphanage rearing and, 624- in schizophrenia, 561, 565-567, 578-579, 587-
possible behavioral explanations for, 654-655 626 588
in number representation research, 417, 428 emotion in, emergence of, 607-609 selective, 323-324, 561-562, 565
PKU treatment and, 448, 451f and infant development, 609-610 feature integration theory of, 562
in reaching skills assessment, 260-263 and maternal behavior in next generation, 609- in modification of blink reflex, 327-329, 328f
in spatial localization assessment, 402-404 610 in schizophrenia, 566-567
in working memory assessment, 372-373, 375t maternal, parallel processes in, 605 and self-regulation, 353-354, 358-360, 638-639
Anterior cingulate cortex maternal separation and, 607—609 and soothing behavior, 358-359
and attention, magnetic resonance imaging neurobiology of, 599-612 spatial, development of, 404-405, 410
studies of, 138-139, 139f olfactory preference learning and, 604-605 supervisory, 562-564
in attention-deficit/hyperactivity disorder, 529, perinatal transition in, 601 sustained, 324-325, 327-329, 437, 561, 564-565
567 from physiological regulation to mental top-down control of, 561-562
cocaine and, 527-529 representation in, 608-609 and working memory, 563-564
dopamine in, 523, 525 postnatal learning and, 602-604 in young infants
dorsolateral prefrontal cortex connection to, prenatal origins of, 600-601 developmental psychophysiological
435 specific, initial formation of, 600-607 perspective on, 321-336
in go/no-go task, functional magnetic system, in newborns, 601-602 direct measures of, 326-327, 334-335
resonance imaging of, 141, 141f and temperament, 633 indirect measures of, 326
in schizophrenia, 565 Attachment-based error, in A-not-B task, 654-655 marker tasks of, 326-327
and self-regulation, 638-639 Attention psychophysiological measures of, 321, 324-
Anterior orbital gyrus, development of, 80t arousal system of, 321-323 327
Apathy and complex cognition, 333—334 and recognition memory, 330-334, 332f
and growth, 622-623 developmental changes in, 327-334 334f
in orphanage-reared children, 618 neuroanatomical model of, 321-322, 322f Attention-deficit/hyperactivity disorder
Aphasia, 281 neurochemical model of, 322-323, 323f attention in, 561, 567-571
in adults, with brain injury, 281-282, 315-316 in attention-deficit/hyperactivity disorder, 561, autism versus, 540
lack of, in children with brain injury, 281—282 567-571 cognitive deficits of schizophrenia versus, 587-
treatment of, training based on experience- in autism, 540 588
induced plasticity for, 316 and behavioral responses to sound, 206 diagnosis of, 567, 569
Appearance-reality tasks, in dorsolateral prefrontal bottom-up control of, 561 dopamine in, 156-157, 525, 567
cortex studies, 441 brain systems involved in, 321-324 executive function in, 542, 568
Arachidonic acid Broadbent's model of, 353 genetic studies of, 151-152, 156-157, 567
and brain development, 491 cocaine exposure and, 529-530 neurobiology of, 567, 570-571
nutritional need for, 496 cognitive neuroscience of, contribution of pharmacologic magnetic resonance imaging of,
Arcuate fasciculus, sex hormones and, 72 research in, to study of attention disorders, with Ritalin treatment, 144
Arcuate nucleus 568-569 prenatal drug exposure and, 529
androgen receptors in, 60 definitions of, 561 prevalence of, 567
progesterone receptors in, 60 in cognitive neuroscience, 561-565 with Tourette's syndrome, 549
ARND. See Alcohol-related neurodevelopmental developmental disorders of, 561-571 working memory in, 568
disorder research in, contributions of, to cognitive Auditory localization
Aromatase activity, 62-63, 63f neuroscience of attention, 569-571 in human infants, 212-213, 216
Aromatic acid decarboxylase, and dopamine, 522, early deprivation/orphanage rearing and, 621- in nonhumans, 210-211
523f 623 Auditory system
Aromatization hypothesis, 62 in emotional versus cognitive tasks, 355, 356f and attachment, 600-602
Aromatization, of androgens to estrogen, 62-63, event-related potentials in, 128-129, 326-327, best areas of, 211-212
62f-63f 66-67 332-336, 334f-335f 562 development of, 205-216
Arousal attention system, 321-323 eye movement control systems and, 329-330, absolute thresholds and intensity
and complex cognition, 333-334 330/-331/ 356-358 discrimination in, 206-207
developmental changes in, 327-334 and face processing, 383-385, 393 early listening preferences in, 208-209, 216,
and heart rate, 325 fetal alcohol syndrome and, 510-511, 513 271
neuroanatomical model of, 321-322, 322f filter theories of, 569 frequency discrimination, resolution and
neurochemical model of, 322-323, 323f heart rate in, 321, 324f 324-325, 327-332, 332f masking in, 207
and selective attentional modification of blink 334-335 and language disorders, 198-199, 208, 309-
reflex, 327-329 individuality/temperament and, 353-354, 360- 311
Asperger disorder, 537-538 361 maps of, 211
Association analysis, 152, 153t, 154-156 involuntary biological versus voluntary social, melody perception in, 209-210
ethnic stratification in, 155-156 353-354 neural specializations in, 210-216
within-family methods in, 153t, 156 and learning, 359 onset of hearing, 206
Association cortex, in habit formation, 550-552, magnetic resonance imaging studies of, 138- perception of complex stimuli in, 208-210,
551f, 552t 139, 139f 216
Associationism, and schizophrenia, 569 networks of, 353-361 fundamental capacities in infants, 205-208
Associative pathways, myelination of, 35-36 alerting, 354, 355f myelination of, 35-36, 42-43
Attachment development of, 356-360 plasticity of, 195-196, 196f 205, 210-216
abuse and, 603, 604f executive control, 354-355, 356f 359-360, in humans, 215-216
behavioral system of, 601-602 561-564 in nonhumans, 213-215
and biological regulation, 607-612, 609t organization of, 355f prenatal alcohol exposure and, 507, 513
brain regions involved in, 605-606, 606f orienting, 354, 355f 356-357, 360 principles of, 216
consummatory, 657t novelty and, 330-334, 357-358 sequence learning in, 210
A-not-B error and, 657-658 object-based, 427 techniques for assessing, 206
cognitive extensions of, 656 phases of, 324-325 temporal resolution in, 207-208
666 INDEX
in utero experiences and, 208-209, 216 implications for fundamental issues in, 156— genetics and, 149-157
visual system and, 211-216 157 backward strategy for studying, 150-151
Autism, 537 methods for studying, 151-156 biometric (indirect) studies of, 151-154, 152t
attention problems in, 540 molecular (direct) studies of, 151, 153t, 154- classes of effects in, 157
versus attention-deficit/hyperactivity disorder, 156 forward strategy for studying, 150-151
540 magnetic resonance imaging of, 138-139 implications for fundamental issues in, 156-
and cognition, 539-545 functional, 141, 141f 157
cognitive neuroscience of, advances in, 537-545 Behavioral inhibition, 631-632, 640-642 methods for studying, 151-156
DSM-IV diagnostic criteria for, 537 Behavioral relevance, and plasticity, 194-195, 312 molecular (direct) studies of, 151, 153t, 154-
executive dysfunction in, 538, 540-545 Behavioral science, neuroscience and, 125 156
component process approach to, 541-543 Behavioral therapy, for early brain injury, 183- iodine and, 491-492, 495t, 500
computer facilitation in studies of, 543-545 185, 185t, 188 iron and, 491-492, 494t, 497-499
and social behavior, 543-545 BESA. See Brain electrical source analysis magnetic resonance imaging of, 36-41, 137-
face processing in, 392-393 Best areas, auditory, 211-212 145, 138f
frontal lobe theory of, 538-539 Bicep muscles, in reaching skills, 258, 258f functional, 140-141, 141f
gender ratio in, 537 Biconditional discrimination task, in relational myelination in, 35-43
genetic studies of, 152 in normal brain, 175-178
memory assessment, 370-371, 375t
glucose metabolism in, 539 nutrients, selected, and, 493-500, 494t-495*
Binocular correlation detection, development of,
high-functioning, versus Asperger disorder, 537- nutrition and, 491-501
229
538 critical periods for, 497, 500-501
Biology, and psychology
macrocephaly in, 157 role of covariates in studies of, 497
bridging gap between, 415-417
medial temporal hypothesis of, 538-539 role of timing in deprivation and subsequent
diagnostic marker research on, 417
neurobiology of, 519, 538-539 repletion of nutrients, 500-501
structural analogy between, 415-417
neuropsychological profile of, 540 plasticity and, 191-199
Biometric methods, of studying genetic influences
onset of, 537 protein-energy malnutrition and, 491, 493-498,
on brain/behavioral development, 151-154,
prevalence of, 537 494t, 501
152t
working memory in, 541-543 and schizophrenia, 577, 581-585
Bipolar disorder, genetic studies of, 152
Autistic-like behavior, in orphanage-reared selenium and, 491, 495t, 500
children, 626 Bird song system, sexual dimorphism in, 66-67
zinc and, 491, 494t, 499-500
Autonomic responsivity, vagal tone as index of, Bivariate extension, of DF method, 153-154
Brain electrical source analysis (BESA), of event-
634 Blind
related potentials, 127
Awkward reach, 438-440 cortical plasticity in, 194, 215-216
Brain injury
Axons sensory compensation in, 215-216
and adult aphasia, 281-282, 315-316
cortical development of, 521 Blink reflex, selective attentional modification of,
diagnostic markers of, 417
myelination of, 35-36 327-329, 328f
early, 175-188
Body-centered frames of reference, 344, 345f 348-
age at injury and functional outcome of, 179,
B 349
180f-182f
Bootstrapping mechanism
Backpropagation algorithm, 162 behavioral sequelae of, 178-179
for integer-list representation, 426-429
Backward strategy, for studying genetic influence behavioral therapy for, 183-185, 185t, 188
properties of, 426
on brain/behavioral development, 150-151 brain development after, 179-183, 183t, 187
Bottleneck developmental theories, of visual
Balance scale task, neural network model of, 165- changes in cortical connectivity with, 180, 187
development, 231-232
166 consequences of recovery from, in adulthood,
Braille readers, cortical plasticity in, 194
Balint syndrome, 427 181-183
Brain
Barn owls, sensory deprivation experiments with, experience and, 175, 183-185, 185t, 188
arousal system of, 321-323
211-212 gonadal hormones and, 175, 186-188, 187f
developmental changes in, 327-334
Basal ganglia Hebb hypothesis of, 178-179
and heart rate, 325 Kennard principle of, 178-180
in attention, 354-356, 359, 561, 563
neuroanatomical model of, 321-322, 322f manipulation of endogenous changes in,
in attention-deficit/hyperactivity disorder, 567,
neurochemical model of, 322-323, 323f 183-187
570-571
as dynamic system, 192-193 neuromodulators and, 175, 185-186, 186f
Ritalin and, 144
equipotentiality of, 282-283 188
dopamine in, 522
face-responsive activation of, 382 and neuronogenesis, 176, 180-183, 182f 187
in fetal alcohol syndrome, 512
sex hormones and, 59-72, 138 and plasticity, 175-188, 281-282, 297-299
in habit formation, 550-552, 551f
general principles of actions, 72 and schizophrenia, 577
myelination of, magnetic resonance imaging of,
organizational versus activational effects of, and spatial attention, 405
38f,41
60-62 and synaptic space, 179-180
in obsessive compulsive disorder, magnetic
resonance imaging of, 139-140 sex steroid receptors in, 59-60 and neglect, 405-406, 410, 427
in procedural learning, 111 sexual dimorphism in, 59, 61, 65-71 perinatal, and face processing, 392
in schizophrenia, 565 examples of, 66-71 and spatial analysis, 406-410
in Tourette's syndrome, 552-553, 555-558 size/volume of unilateral, in children
volume of, developmental changes in, 138 developmental changes in, 138 age of onset and outcome of, 296-299
Basilar membrane, in auditory development, 213 early brain injury and, 179, 187-188 crowding hypothesis of, 296
Bats, echolocation in, 210-211 experience and, 177 developmental sensitivities in studies of, 299
BDNF. See Brain-derived neurotrophic factor genetics and, 157 emergentist view of, 282, 297, 299-304
Behavior, early brain injury and, 175-188 Brain development equipotentiality in, 282-283
Behavior problems, early deprivation/orphanage alcohol and, 505-513 fresh-start hypothesis of, 298
rearing and, 623 basics of, 519-520, 520f IQ data from studies of, 284t-287t, 296-298,
Behavioral approach system, 631 breastfeeding and, 495-498 300
Behavioral development cocaine and, 519-530 irreversible determinism in, 282-297
genetics and, 149-157 early brain injury and, 179-183, 183t, 187 language assessment data on, 288t-295t
backward strategy for studying, 150-151 early deprivation/orphanage rearing and, 623 and language development, 281-305, 316
biometric (indirect) studies of, 151-154, I52t experience-dependent changes in, 177-178, lesion etiology and outcome of, 298
forward strategy for studying, 150-151 178f 187, 191-192 lesion neurocorrelates and outcome of, 298
INDEX 667
Brain injury (continued) in obsessive compulsive disorder, magnetic behaviorally inhibited, 631-632, 640-642
unilateral, in children (continued) resonance imaging of, 139-140, 140f cognitive development of, dorsolateral
methodological confounds in studying, 282, in Tourette's syndrome, 553, 554f 555 prefrontal cortex in, 440-444, 443f
297-299 CD. See Concurrent discrimination tasks effortful control in, 360-361
prospective studies of, 299 cdk2, in neuronogenesis, 14-15 face processing in, 389-391
retrospective studies of, 299 cdk4, in neuronogenesis, 14 development changes during, 390-391
sample size in studies of, 298-299 cdk6, in neuronogenesis, 14 mental representation of facedness in, 389-
seizures with, 296, 298 Cell cycle 390
timing of language testing for, 299 and neuronogenesis, 7, 9f 10, llf 15-18, 17f recognition of identity in, 390-391
type, site and size of lesion and outcome of, output mode of, 9f,10 imaging of, difficulties with, 315
296, 298 time mode of, 9f,10 orphanage-reared
Brain-behavior relationships Cell death attachment of, 624-626
in fetal alcohol syndrome, 512-513 cortical, 519-520, 520f behavior problems in, 623
in visual development, 221-232 hippocampus, evidence of, 47-49, 51f catch-up growth in, 620-623
Brain-derived neurotrophic factor (BDNF), 492, neocortical, and structural variation, 18 cognition in, 620-622
521 sex hormones and, 65 domains of development in, 620-626
BrdU (5-Bromo-2'-deoxyuridine) labeling, 45, 94f Cell migration effects of early deprivation in, 617-627
95-96
alcohol and, 511-512 exemplar studies of, 618-619
Breastfeeding, and cognitive development, 495-
cortical, 80, 519-521, 520f general intelligence in, 620-621
498
genetics and, 157 growth failure in, 622-623
Broadbent, Donald, 353
hippocampal, 49-52, 51f, 56, 369 language development in, 622
Broca's area, sexual dimorphism in, 70f 70-71
5-Bromo-2'-deoxyuridine (BrdU) labeling, 45, 94f sex hormones and, 65 levels of privation for, 618
95-96 Central executive, 562-564 peer problems in, 624
Brownian movement, and diffusion-weighted Central nucleus socio-emotional development in, 623-626
imaging, 37 and fear system, 632, 640 stimulation of, framing questions about, 619
Brown-Peterson task, 478 and temperament, 632, 640-641 studies of, history and conceptual issues of,
Central sulcus, 434, 435f 617-620
c Centrum semiovale, myelination of, magnetic
resonance imaging of, 37t
plasticity in
capacity for, versus adult capacity for, 194, 316
CA1 pyramidal cells Cerebellar vault, in fetal alcohol syndrome, 512 in treatment of specific language impairment
development of, 52-54 Cerebellar vermis and dyslexia, 314-316
sex hormones and, 68 in attention-deficit/hyperactivity disorder, 567 with unilateral brain injury
CA3 pyramidal cells in autism, 538 age of onset and outcome of, 296-299
development of, 52-54, 56 Cerebellum crowding hypothesis of, 296
sex hormones and, 68 alcohol and, 511-512 developmental sensitivities in studies of, 299
Calbindin, 96 in attention, 563 emergentist view of, 282, 297, 299-304
Calcium
in attention-deficit/hyperactivity disorder, 567 equipotentiality in, 282-283
and brain function, 491
in autism, 538-539 fresh-start hypothesis of, 298
as second messenger, estrogen and, 64-65
maturation of, 367-368 IQdata from studies of, 284f-287t, 296-298,
Callosomarginal gyrus, development of, 80t
in memory, 366-368, 374 300
Cambridge Neuropsychological Automated
myelination of, 42 irreversible determinism in, 282-297
Battery (CANTAB), 542-545, 580
cAMP. See Cyclic adenosine monophosphate magnetic resonance imaging of, 37t, 39-40 language assessment data on, 288t-295t
Canaries, sexual dimorphism in song system of, neuronogenesis in, 93, 367-368 language development in, 281-305, 316
66-67 nutrition and, 492-495, 499 lesion etiology and outcome of, 298
CANTAB. See Cambridge Neuropsychological in schizophrenia, 565-566 lesion neurocorrelates and outcome of, 298
Automated Battery sexual dimorphism in, 66 methodological confounds in studying, 282,
Case-control studies, 153t, 155-156 Cerebral cortex. See Cortex 297-299
Castration, and brain development, 60, 62, 66 Change, mechanisms of, neural network models prospective studies of, 299
Cataracts, 240 of, 163-166 retrospective studies of, 299
congenital Children sample size in studies of, 298-299
and competitive interaction between eyes, animal cognition tests applied to, 109-121 seizures with, 296, 298
243-244, 247-248 caveats to, 120-121 timing of language testing for, 299
end of visual deprivation from, visual acuity concurrent discrimination, 112-115, 120t type, site and size of lesion and outcome of,
immediately after, 240-241 declarative learning in, 111-112 296, 298
mismatch of pre- and postsynaptic activity delayed-nonmatch-to sample tasks, 111-112, Chloride, and brain function, 491
with, 247 120t Choice experiment, on number representation,
and nondeprived eye, 244 inferences about functional development of 419, 420f, 422, 424
permanent deficits from, mechanisms neural systems via, 109-121 Cholesterol, and brain development, 491
underlying, 245-248 inferences about ontogenesis of sex-specific Choline acetyltransferase, activity of,
during sensitive period, 244-245, 245f cognition from, 112-115, 114f developmental changes in, 84, 368
treatment for, development of visual acuity
inferences from, summary of, 120t, 120-121 Cholinergic system
after, 241f, 241-244
logic of approach, 109-121 and attention, 322-323, 323f, 325
and visual development, 240-248, 349
Morris search task, 115, 117-118, 119f 120 maturation of, 84, 86
Catch-up growth, after early deprivation, 620-623
object reversal discrimination task, 112-115, in recovery from early brain injury, 175, 185-
Categorical perception, 388
of emotional expressions, 388-389 114f 120t 186, 186f, 188
autism and, 392-393 oddity task, 120t, 120-121 Chomsky, Noam, 283
Caudate nucleus procedural learning in, 111-112 CHP. See Conditioned headturning procedure
in attention, 355 radial-arm maze, 115-117, 116f, 120t Cingulate cortex
development of, 367-368 tasks developed for monkeys, 109-111 in attention, 322, 322f,
dorsolateral prefrontal cortex connection to, tests developed for rodents, 115-120 magnetic resonance imaging studies of, 138-
435 verbal instructions for, 110, 119-120 139, 139f,
in fetal alcohol syndrome, 512 Wisconsin General Testing Apparatus, 109- in attention-deficit/hyperactivity disorder, 529,
in habit formation, 550-552, 551f, 552t 115, 110f, 120 567
668 INDEX
cocaine and, 527-529 Cognitive development, 191-192 Continuous TV-back task, 478, 579-580
dopamine in, 523, 525 breastfeeding and, 495-498 Continuous performance test
dorsolateral prefrontal cortex connection to, dorsolateral prefrontal cortex in, 440-444 of attention, 564-565
435 event-related potentials in study of, 125-133 in attention-deficit/hyperactivity disorder, 568
in go/no-go task, functional magnetic genetics and, 149-157 in schizophrenia, 566-567, 578-579
resonance imaging of, 141, 141f integration of, with systems neuroscience, 427- Contrast sensitivity
in schizophrenia, 565 429 congenital cataracts and, 240-248
and self-regulation, 638-639 intrauterine growth restriction and, 493-495 development of, 223-224, 229-231, 237-248,
Cingulate gyrus neural network models of, 159-170 238f
and attention, 354-355, 356f applications and contributions of, 163-166 mechanisms underlying, 239-240
development of, 79-80, 80t benefits of, 159-160 visual input and, 240-248
Cingulum bundle, myelination of, 82 criticism of, 166-168 PKU treatment and, 456-461, 457f-458f
Classic conditioning response, in procedural elements of, 160-163 Copper, and brain development, 491
memory assessment, 366-368 nutrition and, 491-501 Corona radiata, myelination of, magnetic
CNV. See Contingent negative variation plasticity and, 196-199, 410-411 resonance imaging of, 38f
Cobalt, and brain development, 491 spatial, 399-411 Corpus callosum
Cocaine associated with dorsal visual stream, 401-406 development of, alcohol and, 511-512
and developing nervous system, 519-530 associated with ventral visual pathway, 406- myelination of, 43
and dopaminergic systems, 519, 522 410 magnetic resonance imaging of, 37t, 38f-40f
humans exposed to, in utero, structural and Cognitive dysmetria, 566 41
functional deficits observed in, 528-529 Cognitive neuroscience sexual dimorphism in, 71
prenatal exposure to, in animals developmental Corsi-Milner test, 450t, 453
alterations in brain development and function advances in, 339 Cortex. See also specific cortical specializations
due to, 525-528, 528f fetal alcohol syndrome studies in, 505-513 androgen receptors in, 60
summary of studies of, 525-527, 526t goals of, 149, 417 in attention, 321-323, 322f-323f, 325, 354, 355f
Cognition neural network models in, 159-170, 339 in attention-deficit/hyperactivity disorder, 567,
aging and, 474-475 number representations in, 418-425 570-571
generalized slowing of function with, 484 understanding of mind in, 415, 418 in autism, 519, 538-539
individual differences in, 474-475 genetics and,149-150 cytoarchitectural maturation of, 80-82, 81f
physical fitness and, 474 meanings of attention in, 561-565 519-520, 520f
use-it-or-lose-it hypothesis of, 474-475 Colliculi, myelination of, 42 development of
animal tests of, applied to children, 109-121 Color vision, development of, 228-231, 340, 418 alcohol and, 511-512
caveats to, 120-121 Component process approach, to executive cocaine and, 525-528
concurrent discrimination, 112-115, 120t function, 541-543 fundamentals of, 519-521, 520f
declarative learning in, 111-112 Composite effect, in face processing, 390-391 nutrition and, 493-495, 499
delayed-nonmatch-to sample tasks, 111-112, Computer facilitation, in executive function stages of, 80, 519-520, 520f
120f studies, in autism, 543-545 developmental abnormalities in
inferences about functional development of Concurrent discrimination (CD) tasks disorders associated with, 519
neural systems via, 109-121 for children, 112-115, 114f, 120t and schizophrenia, 565-566, 577, 582
inferences about ontogenesis of sex-specific gonadal hormones and, 113 early injury to, 175-188
cognition from, 112-115, 114f neural basis of, 113 age at injury and functional outcome of, 179,
inferences from, summary of, 120-121, 120t sex specificity of, 112-115, 114f 180f-182f
logic of approach, 109-121 Conditioned headturning procedure (CHP), 206- behavioral sequelae of, 178-179
Morris search task, 115, 117-118, 119f, 120 209, 212-213 behavioral therapy for, 183-185, 185t, 188
object reversal discrimination task, 112-115, Conditioned place preference (CPP) brain development after, 179-183, 183t, 187
114f, 120t dopamine and, 658-659 changes in connectivity with, 180, 187
oddity task, 120-121, 120t opiates and, 656 experience and, 175, 183, 185, 185t, 188
procedural learning in, 111-112 Cones, visual, development of, 221, 223-224, 239 manipulation of endogenous changes in,
radial-arm maze, 115-117, 116f ,120t Configural encoding, in face processing, 390-391 183-187
tasks developed for monkeys, 109-111 Conflict resolution, executive control in, 359-360 neuromodulators and, 175, 185-186, 186f
tests developed for rodents, 115-120 Conflict tasks, aging and, 481-483 188
verbal instructions for, 110, 119-120 Connectionist modeling, 159. See also Neural and neuronogenesis, 180-183, 182f, 187
Wisconsin General Testing Apparatus, 109— network models and .synaptic space, 179-180
115, 110f,120 Conscience, effortful control and, 361 estrogen receptors in, 60
autism and, 539-545 Conspec/Conlern hypothesis, of face processing, event-related potentials from, 125-126
complex, arousal attention system and, 333-334 383 in face processing, 381-382, 388-389, 393
developmental neuroscience and, bridging gap Constructionism, 156 in habit formation, 550-552, 551f, 552t
between, 415-429 Constructivist hypothesis, of synaptogenesis, 29 histogenesis of, 520-521
dorsolateral prefrontal cortex in Consummatory attachment, 657t in learning, 111-113
in childhood, improvement of, 440-444, 443f A-not-B error and, 657-658 myelination of, 82
dopamine and, 433-463 cognitive extensions of, 656 neuropil increase in, 80, 81f
in infancy, evidence of, 435-440, 438f-440f interactions of, with cognitive seeking, 656-657 neurotransmitters in
441t opiates and, 655-656, 658 convergence of, 88-89
early deprivation/orphanage rearing and, 620- Contention scheduling system, 563 developmental similarities among systems of,
622 Context-sensitive sequences, attention and, 359 85-86
menopause and, 72 Contextual detail, memory for, aging and, 483- extrinsic, 84-88
phenylalanine levels and, linear relationship 484 interactions of, 79, 86-88
with, 450-452 Contingent negative variation (CNV), in event- intrinsic, 83-84, 86-88
PKU treatment, early and continuous, and, 434, related potentials, 130 maturation of, 79, 82-85
444-463 Continuity hypothesis, of number representation, partial functioning of, 404, 410
schizophrenia and, 577-588 419 plasticity of, 191-199
sexual differences in, 69-71, 70f 112-115 Wynn's ease of learning argument against, 419- adaptive and maladaptive, 194
spatial, 399 421 in amputees, 193, 193f 195, 195f
INDEX 669
Cortex (continued) Deafferentation, and cortical plasticity, 193-194 and general intelligence, 620-621
plasticity of (continued) Declarative learning and growth failure, 622-623
and auditory development, 213-216 inferences about, from animal cognition tests and language development, 622
behavioral relevance and, 194-195, 312 applied to children, 111-112 levels of privation in, 618
deafferentation and, 193-194 neural basis for, 111 orphanage-reared studies of, history and
in musicians, 193-196, 196f Declarative memory, 365-366 conceptual issues of, 617-620
principles of, 194-195 adult abilities in, emergence of, 368-371 and peer problems, 624
in representational zones, 191-196 development of, 368-371, 374 and socio-emotional development, 623-626
sensitive periods for, 192, 199 in monkeys and humans, 375t Depth of processing hypothesis, of face
temporal dynamics of, 194 inferences about, from animal cognition tests processing, 390-391
in treatment of specific language impairment applied to children, 111-112 Depth vision, development of, 229
and dyslexia, 314-316 neural circuits mediating, maturation of, 369- Development. See also Brain development;
use-dependent, 193-195 371, 374 Cognitive development
sexual dimorphism in, 66, 69-71, 70f Delay of gratification paradigm, in dorsolateral
synaptic connections in, formation of, 82, 519— plasticity and, 191-199
prefrontal cortex studies, 442 Developmental cognitive neuroscience
521, 520f Delayed alternation task
in temperament, 633 advances in, 339
in animal model of cognitive effects of PKU, fetal alcohol syndrome studies in, 505-513
in Tourette's syndrome, 553, 554f
455, 456f goals of, 149,417
Cortical face areas hypothesis, 382
in working memory assessment, 371-374, 375t neural network models in, 159-170, 339
Cortical parcellation hypothesis, of visual
Delayed response task number representations in, 418-425
development, 231
affective demands of, 653 understanding of mind in, 415, 418
Cortical-striatal-thalamo-cortical (CSTC) circuits
as diagnostic marker of dorsolateral prefrontal Developmental neuroscience
in habit formation, 550-552, 551f 552t
in Tourette's syndrome, 553, 554f cortex injury, 417 and cognition, bridging gap between, 415-429
Corticofugal pathways, early brain injury and, 180 in dopamine-dorsolateral prefrontal cortex structural analogy in, 415-417
Corticolimbic pathways, myelination of, 82 studies, 435-440, 438f-439f, 441t, 459
DF method, 153-154
Corticopontine tracts, myelination of, 43 performed by children treated for PKU, 454,
Diagnostic and Statistical Manual of Mental Disorders,
Corticospinal tracts, myelination of, magnetic 455t
fourth edition (DSM-IV), diagnostic criteria in
resonance imaging of, 38f in working memory assessment, 371-374, 375t
for Asperger disorder, 537-538
Corticosterone, infant levels of, mother-child Delayed-nonmatch-to-sample (DNMS) tasks, 449t
for attention-deficit/hyperactivity disorder, 569
interactions and, 609-610 for children, 111-112, 120t
for autism, 537
Corticostriatal system, in procedural learning, in recognition memory testing, 369-370, 375t
Diagnostic markers, 417
111-112 Deltoid muscles, in reaching skills, 258, 258f-259f
Diencephalic dopamine, 649
Corticotropin-releasing factor, opiates and, 655 Dendrite(s)
Diencephalon
Corticotropin-releasing hormone, mother-infant development of
development of, nutrition and, 493
interactions and, 610 in cortex, 80, 521
in fetal alcohol syndrome, 512
Cortisol levels early brain injury and, 179-180, 186-188,
medial, in declarative memory, 368
in behaviorally inhibited children, 640-641 187f
early deprivation/orphanage rearing and, 622 Differential maturation hypothesis
experience-dependent changes in, 177-178,
as measure of reactivity, 634-635 subtleties of, 230-232
178f
CPP. See Conditioned place preference of visual development, 226-232
in hippocampal formation, 52-54, 53f-54f
CPT. See Continuous performance test Differential regression to mean, 153
119
Cranial nerves, myelination of, 42 Diffusion anisotropy, 37, 41
in normal brain, 176-177
Critical periods Diffusion tensor imaging, 484
nutrition and, 492, 499
developmental, neural network model of, 164— Diffusion-weighted imaging
and visual development, 226, 227f 230
165 Brownian movement and, 37
orientation of, and event-related potentials, 126,
for language development, 165, 282 of myelination, 37, 39, 43
127f
for nutrition, in brain development, 497, 500- Digit span task, in schizophrenia, 579
sex hormones and, 65
501 Digits, cortical representational zones for
Dentate gyrus
in synaptogenesis phase 3, 27-28 fusion of, 193-194
cell death in, evidence of, 47-49, 51f
for visual development, 211-212 plasticity of, 193-194
Crowding hypothesis, of language development in cell migration in, 49-52, 51f 56
development of, 96f 96-97, 369, 371 Dihydrotestosterone, testosterone transformation
children with unilateral brain injury, 296 to, by 5a-reductase, 62f 63-64, 65f
CSTC circuits. See Cortical-striatal-thalamo-cortical early deprivation/orphanage rearing and, 621
estrogen receptors in, 60 Dipole, in event-related potentials, 126, 127f
circuits
hormones and, 68-69, 93, 97 Directional selectivity, visual, development of, 229
Cue(s)
in learning and memory, 56-57 Directional Stroop task, in dopamine-dorsolateral
in radial-arm maze testing, 116-117
in spatial-relational learning, 115 neuronal connections of, development of, 52, prefrontal cortex studies, 462t, 462-463,
Cuneus gyrus, development of, 80t 56-57 463f-464f
Cyclic adenosine monophosphate (cAMP), as neuronal development in, of principal and Discriminant validity, 540
second messenger, estrogen and, 64 nonprincipal neurons, 52-54, 53f, 56 Disparity sensitivity, visual, development of, 229
Cyclin D, in neuronogenesis, 14 neuronogenesis in, 45-47, 48f-50f 369, 371 Distal pathways, myelination of, 35-36
Cyclin E, in neuronogenesis, 14-15 in adults, 93-103 Distinctiveness effects, in face processing, 390
Cyclin-dependent kinases, in neuronogenesis, synapse formation in, 52, 54-56, 55f Distractors, susceptibility to, aging and, 479-483
14-15 2-Deoxyglucose metabolic labeling, in dorsolateral Distress keeper, in infants, 358
prefrontal cortex studies, 436 Divided visual field studies, of face processing,
D Depression, infantile, and growth, 622-623 382, 388
Depression, long-term, synaptic, 192, 247 DL-PFC. See Dorsolateral prefrontal cortex
Darwinian medicine, 648
Deprivation, early Docosohexaenoic acid
Day-night Stroop-like task, 449t
in dorsolateral prefrontal cortex studies, 442- and attachment, 624-626 and brain development, 491
443, 443f, 448, 451f and behavior problems, 623 nutritional need for, 496
PKU treatment and, 448, 451f catch-up growth after, 620-623 Dopamine
Deaf and domains of development, 620-626 in A-not-B task, 435-440, 438f-439f, 4411, 447-
cortical plasticity in, 215-216 effects of, 617-627 448, 449t, 451f, 459, 459f, 653-655
sensory compensation in, 215-216 exemplar studies of, 618-619 in appetitive motivation, 649-650
670 INDEX
in attention, 322-323, 323f, 325 computational model of object processing in, Ease of learning argument, against continuity
in attention-deficit/hyperactivity disorder, 156- 346/ 346-348 hypothesis of number representation, 419-
157, 525, 567 development of, 341-348 421
brain systems dissociation of, with ventral pathway, behavioral Echoes, suppression of, 213
abnormalities in, disorders associated with, evidence of, 344-346, 349 Echolocation, 210-211
522 and eye movement control, 339-340, 342f-343f Echopractic errors, 444
anatomical organization of, 522-523 342-344, 349, 401 E2F, in neuronogenesis, 14
neuropharmacology of, 521-525 function of, 399 Effortful control, 562-563
sex dimorphism in, 67 high-density ERP studies of, 342-344, 343f-344f during childhood, 360-361
cocaine and, 519, 522, 525-528 neurocomputational properties of, 339-341 Elderly
in conditioned place preference, 658-659 plasticity of, 348-349 brain imaging data from, interpretation of, 475
in consummatory attachment, 655-656 processing in, 340-341 event-related potentials in
in cortical development, 521 spatial processes associated with, development mismatch negativity amplitude in, 475-476,
developmental roles of, 524-525 of, 401-406 476f-477f
diencephalic, 649 in visual processing, 339-341 novelty P3 in, 479-481, 481f-482/
functional activity of, individual differences in, as where or action pathway, 340, 399 P300 amplitude and scalp distribution in, 479
652-653 Dorsolateral prefrontal cortex individual differences and cognitive function in,
functions of, 522 anatomy of, 434-435, 435f 474-475
and glutamate, interactions between, 525 in attention, 436 inhibitory control in, 479-483
innervation, development of, 524 in cognition memory in
iron and, 492, 498 in childhood, improvement of, 440-444, 443f for contextual detail, 483-484
major cell groups of, 649 in infancy, evidence of, 435-440, 438f-440f frontal cortex activation patterns in, during
mesencephalic, 649 441t working memory task, 479, 480f
mesocortical, 521-525, 649, 651-652 phenylalanine levels and, 450-452 frontal lobe function and, 473-484
ontogeny of, 523-524 in conflict tasks, 482 future directions for study of, 484
mesolimbic, 649 connections and interactions of, 435 item, 473, 483
metabolism, 522, 523f number of operations to be performed in
damage
as neurotransmitter, 522 working memory and, 478-479
cognitive effects of, 436-440
interaction and convergence of, with sensory, 474-475
diagnostic marker for, 417
source, 473-474, 483-484
serotonin, 79, 88-89 dopamine in
type of material to be processed in working
interaction with GABA, 86-89, 87f-88f and cognition, 435-456, 459-463
memory and, 479
maturation of, 79, 85-86, 86f early and continuous treatment of PKU
working, 473-484
opiates and, 655-656 model of, 434, 444-463, 447f
working memory load and, 476-478
pharmacology, 522, 523f evidence from visual psychophysics
methodological issues in studies of, 474-475
in prefrontal cortex confirming hypothesis of, 456-459, 457f- susceptibility to distractors in, 479-483
and cognition, 435-456, 459-463 458f Electrode (s)
early and continuous treatment of PKU role of, 433-463, 651-652 10/20 system of, 126
model of, 434, 444-463, 447f selective deficit in, with PKU treatment, 444- high-density arrays of, 126-127
evidence from visual psychophysics 447, 445f for recording event-related potentials, 126
confirming hypothesis of, 456-459, 457f- tyrosine levels and, 446-447, 447f 460 Electroencephalography (EEC)
458/ functions of, 433, 459 in affective bias studies, 635-636
and inhibitory control, 436-440, 448-450 in inhibitory control, 436-440, 448-454, 459- aging and, 484
role of, 433-463, 523, 525, 647-659 463, 479 in attention studies, 326-327, 334-336
selective deficit in, with PKU treatment, 444- maturation of, 373-374, 417, 433 event-related potentials in, 125-133. See also
447, 445f in schizophrenia, 586-587 Event-related potentials
and progesterone receptors, in sexual behavior, in self-regulation, 639 Electrooculogram, of infant eye movements, 329
65,72 size of, evolutionary increase in, 434-435 Emergentist view, of children with unilateral brain
in retinal function, 456-461, 457f-4S8/ in spatial localization, 401-404 injury, 282, 297, 299-304
in reward-seeking behavior, 647-659 in spatial working memory, 650-652 Emotion(s)
in schizophrenia, 523, 525, 565-566, 586-587, in working memory, 366, 371-374, 436, 440- A-not-B task and, 653
652 444, 448-454, 459-463, 473 in attachment, emergence of, 607-609
synthesis of, 522, 523f Down syndrome hemisphere specialization in, 635
in Tourette's syndrome, 553-555 genetic studies of, 150-151 perception of, in face processing
tyrosine levels and, 446-447, 447f 460, 522, 523f microcephaly in, 157 autism and, 392-393
in working memory, 436, 440-444, 448-454, Drawing tasks, in Tourette's syndrome, 555-556 categorical, 388-389, 392-393
459-463 Dwarfism, psychosocial, 622-623 in infants, 388-389
spatial, 650-652 Dyscalculia, 422 in newborns, 384
Dopamine receptors, 87, 373, 522 Dyslexia, 310 regulation of
in attention-deficit/hyperactivity disorder, 156- auditory/speech development and, 198-199, attention and, 353-354, 358-360, 638-639
157, 525 309-311 cognitive prerequisites for, 638
development of, 524 functional magnetic resonance imaging in, 141- early deprivation/orphanage rearing and,
pharmacological activation of, and cognition, 143 623-626
436 genetic studies of, 151-152, 154-156 fetal alcohol syndrome and, 506
in prefrontal cortex, 434 temporal processing deficits in, 198-199, 309- and temperament, 632-634, 636-640
subtypes of, 522 311 and reward-seeking behavior, 648, 653, 658
in ventral tegmental area, 436, 446-447 treatment of, 309 and temperament, 631-632
Dorsal putamen, myelination of, magnetic experience-driven plasticity in, 314-316 Emotional development, early deprivation/
resonance imaging of, 40 orphanage rearing and, 623-626
Dorsal visual pathway, 339-349, 399, 400f E Emotional reactivity
anatomy of, 399-401, 400f hypothalamic-pituitary-adrenal axis in, 634-635
and body-centered frames of reference, 344, Early brain injury. See Brain injury, early and temperament, 632-635
345f, 348-349 Early deprivation, effects of, 617-627 vagal tone as measure of, 634
INDEX 671
Empathy, effortful control and, 361 mismatch negativity amplitude in, 475-476, Eye movement control
Encoding switch hypothesis, of face processing, 476f-477f and attention, 329-330, 330f-331f, 356-358
390-391 novelty P3 in, 479-481, 481f-482f dorsal visual pathway and, 339-340, 342f-343f
Energy P300 amplitude and scalp distribution in, 479 342-344, 349, 401
attention as, 564-565 working memory load and, 478 Eyes, preferential looking at, 385
and brain development, 491, 493-498, 494t in phonetic perception, 272-274 Eysenck's model of personality, 631
Entorhinal cortex recording of, 126
cell migration in, 49-52, 56 in sensory memory, aging and, 475-476 F
in declarative learning, 111 signal, genesis of, 125-126,127f
in hippocampus, 54-56 in speech processing, 270-271 Face inversion effect, 382, 385, 386f-387f 390-391
in memory, 368 topography and source localization in, 127 Face processing
neurons of in word learning, 275-276 in adults, 381-383
development of, 53 Evocation,121 in autism, 392-393
formation of, 45-47, 48f-50f, 56 of rules, 121 in childhood, 389-391
Environment E-wave, 130 development changes during, 390-391
early deprivation in, effects of, 617-627 Excitatory input, and event-related potentials, mental representation of facedness in, 389-
and recovery from early brain injury, 183-185, 125-126,127f 390
185t Executive control/function, 540 recognition of identity in, 390-391
and synaptogenesis, 24f, 27-29, 31-32, 150 attention network of, 354-355, 356f 359-360, composite effect in, 390-391
Eph tyrosine kinases, and dopamine, 524 561-564 cortical face areas hypothesis of, 382
Epidermal growth factor, 492 in attention-deficit/hyperactivity disorder, 542, cortical/hemisphere specialization for, 381-382,
Epigenesis 568 388-389, 393
probabilistic, 150, 156 in autism, 538, 540-545 depth of processing hypothesis of, 390-391
and synaptogenesis, 23, 31-32 and behavioral inhibition, 641 development of, 231, 342, 393-394
Epilepsy, temporal lobe, cognitive deficits of component process approach to, 541-543 normative, 383-391
schizophrenia versus, 587 development of, 359-360 developmental disorders of, 391-393
Episodic memory, 580 discriminant validity in studies of, 540 distinctiveness effects in, 390
in schizophrenia, 577, 580-581 early deprivation/orphanage rearing and, 621- encoding switch hypothesis of, 390-391
Equipotentiality, in children with unilateral brain 622 event-related potentials in, 385-386, 386f-387f
injury, 282-283 fetal alcohol syndrome and, 506, 513 390
EROS. See Event-related optical signal hierarchical relation of, with orienting, 360 experience and, 383-384, 393-394
Error-driven learning algorithm, 162 individuality/temperament and, 360-361 in infants, 383-389
Estradiol. See Estrogen measurement imprecision in studies of, 540-541 mental representation of facedness in, 383-
Estrogen prefrontal cortex in, 434 386
aromatization of androgens to, 62-63, 62f-63f in schizophrenia, 566, 579-580, 584 perception of emotion in, 388-389
66-67 and self-regulation, 638-639 recognition of identity in, 387-388
and brain development, 59, 61-69, 138 and social behavior, 543-545 neuropsychology of, 381-394
and dopamine, 67 and temperament, 633 in newborns, 383-384
in learning and memory, 68-69, 102-103 Expectation, event-related potentials in, 130 Conspec/Conlern hypothesis of, 383
as neural growth factor, 65 Experience externality effect in, 384
and neuronogenesis, 93, 97, 98f, 102-103 and brain development, 191-192 limitations of, 384
organizational and activational effects of, 61-62 normal, 177-178, 178f 187 mental representation of facedness in, 383
and verbal abilities, 70 early deprivation of, effects of, 617-627 perception of emotion in, 384
Estrogen receptors and face processing, 383-384, 393-394 recognition of identity in, 383-384
in brain, 59-60, 64, 68, 138 and neuronogenesis, 93, 98-100,103, 177-178 sensory hypothesis of, 383
ER-a, 59, 64 and recovery from early brain injury, 175, 183- social hypothesis of, 383
ER-B, 59-60, 64, 68 185, 185t, 188 norm-based encoding hypothesis of, 390-391
Ethnic stratification, in association studies, 155- and speech/language processing, 276-277 versus object processing, 381-383
156 and visual development, 192, 240-248, 619 other race/other species effect in, 389-390
Event-related optical signal (EROS), 482-484 Experience-dependent synaptogenesis, 24f 27-28, perinatal unilateral lesions and, 392
Event-related potentials 177-178 relational/configural/holistic encoding
in attention, 128-129, 326-327, 332-336, 334f- Experience-driven plasticity, 311-312 hypothesis of, 382
335f, 562 after speech perception training, 313 right hemisphere development hypothesis of,
in cognitive development, 125-133 characteristics of training necessary to induce, 390-391
components of 312-313, 315 right hemisphere hypothesis of, 382
development of, 128-132,131f-134f in treatment of specific language impairment special nature of, 381-383
labeling of, 126 and dyslexia, 314-316 ventral visual pathway and, 342, 381
negative, 128-130 Experience-expectant synaptogenesis, 24f, 27 windows of opportunity for, 393-394
positive, 130-131 Experience-expectant view, of face processing, Face-responsive brain activation, in humans, 382
primary, 128-132 393-394 Face-responsive neurons, in monkeys, 382
in conflict tasks, 482 Experience-independent synaptogenesis, 24f, 27 Faces, preferential looking at, 383-385, 387, 416
cross-registering of, to other imaging methods, Experience-induced plasticity, for aphasia Failure to thrive, of orphanage-reared children,
133 treatment, 316 618
electrode arrangements for, 126 Explicit learning, 111. See also Declarative learning False belief tasks, in dorsolateral prefrontal cortex
excitatory input and, 125-126, 127f Externality effect, in face processing, 384 studies, 441-442
in face processing, 385-386, 386f-387f 390 Extinction, 405 False memory, 484
high-density studies with, of dorsal and ventral Extraversion, 648 Family studies, 152, 152t
visual pathways, 342-344, 343f-344f, 348 Extrinsic neurotransmitters FAS. See Fetal alcohol syndrome
inhibitory input and, 125-126,127f developmental changes in, 84-85 Fasciculus habenulointerpeduncularis, myelination
in language studies, 197-198, 270-276, 313 interactions of, with intrinsic neurotransmitters, of, 42
in older adults 86-88 Fats, and brain development, 491
interpretation of, 475 Extrinsic thalamic afferents, myelination of, 42 Fatty acids, and brain development, 491
in memory research, 475-476, 478-484 Extroverts, 631 Fear system, and temperament, 632-633, 640-641
672 INDEX
Feature integration theory, of selective attention, in attention-deficit/hyperactivity disorder, 567 Genicostriate visual system, 399
562 in autism, 538-539 Gerstmann's syndrome, 428
Feedback weights, in neural network models, 160, damage, cognitive effects of, 442-444 Glia cells, development of
161f early deprivation/orphanage rearing and, 621- early brain injury and, 187
Feedforward weights, in neural network models, 622, 626-627 experience and, 177-178
160, 161f in schizophrenia, 565, 585-586 in normal brain, 175-176, 176f 519-520, 520f
Fetal alcohol effects, 505-513 in self-regulation, 633, 636-638 Glial growth factor, 492
estimated prevalence of, 507 in spatial working memory, 650-652 Global-local task, 449t-450t
explanation of terminology, 509 Frontal pole, myelination of, 35-36 Globus pallidus
Fetal alcohol syndrome (FAS), 505-513 Frontal white matter, myelination of, magnetic in habit formation, 550-552, 551f 552t
associating structural changes in, with function, resonance imaging of, 37t in obsessive compulsive disorder, magnetic
512-513 Functional magnetic resonance imaging (fMRI), resonance imaging of, 139-140, 140f
and attention problems, 510-511 137, 140-143 in Tourette's syndrome, 553, 554f
CNS characteristics of, 507-509 of behavioral development, 141, 141f Glucocorticoid (s)
developmental cognitive neuroscience approach of brain development, 140-141, 141f and neuronogenesis, 97-98
to, relevance of, 513 of children, difficulties with, 315 sex dimorphism in, 67-69
developmental cognitive neuroscience context combination of, with event-related potentials, Glucocorticoid receptors, early deprivation/
for, establishment of, 505-507 133 orphanage rearing and, 621
discovering and diagnosing, 507-509 in face processing, 382, 385 Glucose metabolism
facial characteristics of, 507-508, 507f-508f in older adults, interpretation of, 475 in autism, 539
growth characteristics of, 507-508 in phonetic perception, 273-274 and brain development, 491
incidence of, 507, 509 in reading disorders, 141-143, 143f-144f Glutamate
necropsy and brain studies of, in humans and in remediation/intervention, 141-143 in cortical development, 521
animals, 511-512 in schizophrenia, 581 dopamine and, 525
neurodevelopmental consequences of, in in Tourette's syndrome, 553, 554f as neurotransmitter, maturation of, 83
humans and laboratory animals, 509-511 Fundamental frequency, in speech, 270 and olfactory preference learning, 604
postnatal rearing conditions and, 506-507, 513 Fusiform gyrus, in face processing, 385-386, 391 in schizophrenia, 565-566
Fetus Future-oriented behavior, event-related potentials Glutamate acid decarboxylase (GAD), activity,
auditory development in, 208-209 in, 130 developmental changes in, 83-84, 368
origins of attachment in, 600-601 Glutamate receptors, in long-term potentiation, 83
Fight-flight system, 631 G Gonadal hormones. See Sex hormone (s)
Filter theories, of attention, 569 Go/no-go task
Gl restriction point, of neuronogenesis, 9f, 14-15 in autism assessment, 541
Fingers, cortical representational zones for
GABA in dorsolateral prefrontal cortex studies, 442
fusion of, 193-194
as neurotransmitter functional magnetic resonance imaging in, 140-
plasticity of, 193-194
interactions with dopamine, 86-89, 87f-88f 141, 141f
5ot-reductase, and effects of sex hormones on
maturation of, 83f, 83-84 Gramezy, Norman, 565
brain, 62f, 63-64, 65f
and olfactory preference learning, 604 Granule neuron (s)
Flexibility
zinc and, 499 death of, 47-49, 51f
in acquisition of skilled movement, 253-265 GABA receptors, GABAA and progesterone, 64
versus stability, 255, 262 development of, 52-54, 53f 57, 119, 367
GABAergic neurons migration of, 49-52, 51f 56
Fluoride, and brain development, 491 cocaine and, 527-528, 528f
fMRI. See Functional magnetic resonance imaging production of, 45-47, 56, 68-69
development of in adults, 93-103 .
Focal hand dystonia, 194 in cortex, 521
Foldiak algorithm, 346-347 experience and, 93, 98-100, 103
in hippocampal formation, 52-54 learning and, 93, 99-100, 99f-101f, 103
Folk acid, and brain/neural development, 492 in habit formation, 550-552, 551f, 552t
Food behaviors, early deprivation/orphanage steroid hormones and, 93, 97
GAD. See Glutamate acid decarboxylase stress and, 93, 97-98, 99f, 103
rearing and, 623 Gamma aminobutyric acid. See GABA
Forced-choice testing, with radial-arm maze, 116 Grating acuity, development of, 237-248, 238f
Gelman and Gallistel's continuity hypothesis, of mechanisms underlying, 239-240
Forceps major, myelination of, 43 number representation, 419 visual input and, 240-248
Foreign language, difficulties with processing, 269, Wynn's ease of learning argument against, 419- Gray matter, volume of, developmental changes in,
272-274 421
Foreign listening syndrome, 269 138
Gene(s) Graybiel, Ann, 549-551
Formants, in speech, 270 polymorphic, 149 Growth, catch-up, after early deprivation, 620-623
Formula, nutritional deficiencies in, and brain unique to humans, 156-157 Growth factors, 492
development, 495-498 Generalization, in neural network models, 167- and brain development, 492-493
Fornix, myelination of, 82 168 Growth failure, early deprivation/orphanage
Forward strategy, for studying genetic influence Genetic determinism, 150, 156 rearing and, 622-623
on brain/behavioral development, 150-151 Genetics Growth hormone
Fos expression, and visual development, 241-242 animal models of, 151 early deprivation/orphanage rearing and, 622-
Fragile X syndrome and brain size, 157 623
genetic studies of, 150-151 and brain/behavioral development, 149-157 mother-infant interactions and, 607
macrocephaly in, 157 backward strategy for studying, 150-151 Growth hormone-releasing hormone, early
Free-choice testing, with radial-arm maze, 116 biometric (indirect) studies of, 151-154, 152t deprivation/orphanage rearing and, 623
Frequency discrimination, 207 classes of effects in, 157 Guilt, effortful control and, 361
Fresh-start hypothesis, of brain injury, 298 forward strategy for studying, 150-151 Gyrus (gyri), 434. See also specific gyri
Friendliness, indiscriminate, in orphanage-reared implications for fundamental issues in, 156- development of, 79-80, 80t
children, 625-626 157 Gyrus rectus, development of, 80t
Frontal cortex, 434, 435f methods for studying, 151-156
age-related changes in molecular (direct) studies of, 151, 153t, 154- H
activation patterns during working memory 156
task, 479, 480f confounding variables in studies of, 149-150 Habit(s),549
and working memory, 473-484 controversy over studies in psychology, 150 establishment of, perceptually cued learning
in attention, 354, 355f, 561, 563 weaknesses in studies of, 151 and, 550-552
INDEX 673
Habit(s) (continued) learning and, 93, 99-100, 99/-101f, 103 Indiscriminate friendliness, in orphanage-reared
formation of, 549-558 neuroendocrine factors and, 93, 97, 103 children, 625-626
disregulation of neural substrates underlying, stress and, 93, 97-98, 99f, 102-103 Infant-directed speech, 197-199, 209, 271
tics and, 552-554 neurotransmitters in, glutamate, maturation Infants
neural substrates of, 549-552 of, 83 attachment of, neurobiology of, 599-612
neuroanatomic circuits in, 550, 551f ,552t nutrition and, 491-495, 499 attention in
Habit learning, 111. See also Procedural learning in opiate reward pathway, 656 arousal system of, 321-323, 327-334
Hamilton Perserveration Test, 373 in schizophrenia, 580-581 development of, 327-334, 356-360
Hand dystonia, focal, 194 serotonin receptors in, 85 developmental psychophysiological
Hand trajectories, and reaching skills sex hormones and, 60, 64-65, 67-69, 72, 93, 97, perspective on, 321-336
in adults, 254, 255f 102-103, 138 direct measures of, 326-327, 334-335
perseverative, in infants, 261, 262f-263f sexual dimorphism in, 66-69 event-related potentials in, 326-327, 332-336,
Haplotype relative risk (HRR), 153t, 156 synapse formation in, 52, 54-57, 55f, 68, 371 334f-335f
Head-sparing, 492-493 volume of, developmental changes in, 138 eye movement control and, 329-330, 330f-
Headturning responses, in auditory assessment, Holistic style, in language development, 303 331f, 356-358
206-207, 212-213 Homunculus, 192-193 heart rate in, 321, 324f 324-325, 327-332,
Hearing. See also Auditory system Hormone replacement therapy, and cognitive 332f 334-335
onset of, 206 functions, 72 indirect measures of, 326
Heart rate, as psychophysiological measure of Hormone(s), sex. SeeSex hormone(s) and learning, 359
infant attention, 321, 324f 324-325, 327-332, HPA. See Hypothalamic-pituitary-adrenal axis marker tasks of, 326-327
332f 334-335 HRR. See Haplotype relative risk and modification of blink reflex, 327-329,
Hebb hypothesis, of early brain injury, 178-179 5-HT. See 5-Hydroxytryptamine 328f
3
Hebbian algorithm, 162 H-thymidine autoradiography, 94-95 psychophysiological measures of, 321, 324-
Hebbian mechanism, in plasticity, 192 Human Genome Project, 157 327
Hemispatial neglect, 405-406, 410, 427 Huntington's disease, cognitive deficits of selective, 323-324, 327-329, 328f
Hemisphere specialization schizophrenia versus, 587 and soothing behavior, 358-359
in emotions, 635 3a-hydroxysteroid oxidoreductase, 64, 65f auditory development in, 205-216
in face processing, 382, 388-389, 393 3b-hydroxysteroid oxidoreductase, 64, 65f absolute thresholds and intensity
in language development, 281-282, 304-305 5-Hydroxytryptamine (5-HT). See also Serotonin discrimination in, 206-207
sex differences in, 70 cocaine and, 528 frequency discrimination, resolution and
in speech processing, 270-271 developmental changes in, 85 masking in, 207
in visual-motor integration, 555-556 and olfactory preference learning, 604 perception of complex stimuli in, 208-210,
Heritable markers, 94 Hyperactivity 216
Heschl's gyrus, myelination of, 35-36, 43 in attention-deficit/hyperactivity disorder, 567 sequence learning in, 210
Heterochronic epigenesis hypothesis, of fetal alcohol syndrome and, 510-513 techniques for assessing, 206
synaptogenesis, 32 Hyperkinetic reaction of childhood, 567 temporal resolution in, 207-208
Hidden regulators, in mother-infant interaction, Hyperphagia, early deprivation/orphanage in utero experiences and, 208-209, 216
607-612, 609< rearing and, 623 auditory localization in, 212-213, 216
Hidden units, in neural network models, 160, Hypoglycemia, and brain development, 491 breastfeeding of, and cognitive development,
161f Hypothalamic-pituitary-adrenal axis (HPA) 495-498
High-density arrays of electrodes, for recording activation of, and neuronogenesis, 98 cognition, dorsolateral prefrontal cortex and,
event-related potentials, 126-127 in behaviorally inhibited children, 640-641 435-440, 438f,-440f, 441t
Hippocampal trisynaptic circuit, 52, 55-57 cortisol levels in, as measure of reactivity, 634- event-related potentials in, 131-132, 131f
Hippocampus 635 face processing in, 231, 342, 383-389
development of, mother-infant interaction and, mental representation of facedness in, 383-
alcohol and, 512
androgen receptors in, 60 609-610, 612 386
opiates and, 655 perception of emotion in, 388-389
in attachment, 606
Hypothalamic-pituitary-ovarian axis, regulation recognition of identity in, 387-388
in attention, 322
of, 59 hearing onset in, 206
autopsy specimens for study of, 46, 47t
Hypothalamus isolation call of, 607-608
cell death in, evidence of, 47-49, 51f
androgen receptors in, 60 listening preferences in, 208-209, 216, 271
cell migration in, 49-52, 51f, 56, 369
aromatase activity in, 63, 63f maternal regulators of biological functions in,
composition of, 68
early deprivation/orphanage rearing and, 622- 607-612, 609t
early deprivation/orphanage rearing and, 621-
623 melody perception in, 209-210
622 motor skills of, stability and flexibility in
estrogen receptors in, 59
estrogen receptors in, 60 acquisition of, 253-265
in isolation call of infants, 608
event-related potentials from, 125-126 number representation in, 418-425
in opiate reward pathway, 656
in face processing, 383-384, 388 orienting in, 356-357
progesterone receptors in, 60
in isolation call of infants, 608 orphanage-reared, effects of early deprivation
sex hormones and, 59-65
in learning and memory, 45, 56-57, 374 in, 617-627
sexual dimorphism in, 66
adult-generated neurons and, 100-103 reaching skills of, 253-265
Hypothyroidism, and brain development,
declarative, 111-112, 369-371 development of control in, 254-259
492, 500
sex differences and, 67-69 developmental changes in, 260-262
spatial-relational, 115, 119-120 I dorsal pathway and, 340-341, 349, 401-404
working, 372 from early movement (wiggling) to, 255-256
morphological changes of ICA. See Independent component analysis linear synergy constraint on, 256-258, 257f
functional implications of, 56-57 IDED task. See Intradimensional-extradimensional muscle pattern changes at onset of, 258-259,
from midgestation to early childhood, 45-57 task 258f-259f
neuronal connections in, development of, 52 Identity, facial, recognition of perseverative, 260-262, 262f-263f
neuronal development in, of principal and in childhood, 390-391 selecting stable pattern for, 255-256
nonprincipal neurons, 52-54, 53f-54f, 56 in infants, 387-388 speech and language processing in,
neuronogenesis in, 45-47, 48f-50f, 56, 369, 371 in newborns, 383-384 neurocognitive approach to, 269-277
in adults, 93-103 Independent component analysis (ICA), of event- speech perception in, 271-272
experience and, 93, 98-100, 103 related potentials, 127 temperament in, 631-642
674 INDEX
visual capacities of, 228-230 IQ Language organ hypothesis, 281-282
visual development in, 221-232, 225f early deprivation/orphanage rearing and, 620- Language processing
absolute thresholds in, 222-223 621 experience and, 276-277
acuity in, 223-224, 229-230, 237-248 focal brain injury and, 284t-287t, 296-298, 300 foundations of, 269-271
contrast sensitivity in, 223-224, 229-231, intrauterine growth restriction and, 493 in infancy, neurocognitive approach to, 269-277
237-248 PKU treatment and, 446, 461-462 Laplacian analysis, of event-related potentials, 127
cortical, 221, 224-232, 240 schizophrenia and, 582-584 Lateral geniculate nucleus (LGN), 399
differential maturation in, 226-232 Iron neurotransmitters in, glutamate, maturation of,
dorsal and ventral pathways in, 339-349 and brain development, 491-492, 494t, 497-499 83
effort required for tasks in, 231 regionalization effect of, 498-499 visual deprivation and, 246-247
retinal, 221-224, 239-240 Irreversible determinism, in children with in visual development, 226-227, 228f, 229-231,
word learning in, 274-276 unilateral brain injury, 282-297 239-240
Inferior cerebellar peduncle, myelination of, 42 Isolation call, of infants, 607-608 Lateral orbital gyrus, development of, 80t
magnetic resonance imaging of, 37t, 39 Item memory, aging and, 473, 483 Lateral weights, in neural network models, 160,
Inferior colliculi, brachia of, myelination of, 42 161f
Inferior temporal gyrus, development of, 80t K LCPUFA. See Long-chain polyunsaturated fatty
Infundibular nucleus, progesterone receptors
Kanner, Leo, 538 acids
in, 60
Kennard principle, 178-180 Learned helplessness, 620
Inhibition, behavioral, 631-632, 640-642
Ki-67, as proliferation marker, in hippocampus, Learning
Inhibition of return, 357-358, 404-405
46-47, 47t, 48f-50f adult-generated neurons and, 100-103
Inhibitory control, 403
Knowledge, neural network models of, 160, 163- attention and, 359
aging and, 479-483
166 auditory, 205-216
in A-not-B task, 657
autism and, 540
in attention-deficit/hyperactivity disorder, 568
dorsolateral prefrontal cortex in, 479 L declarative
dopamine and, 436-440, 448-454, 459-463 inferences about, from animal cognition tests
Lambda wave, in ERP studies of eye movement, applied to children, 111-112
early deprivation/orphanage rearing and, 621-
343-344 neural basis for, 111
623
Language ease of, in number representation, 419-421
PKU treatment and, 448-450, 449t-450t, 451f
event-related potentials in, 129-130 estrogen and, 68-69, 102-103
in schizophrenia, 566
native versus foreign, 269, 272-274, 276-277 fetal alcohol syndrome and, 511-512
and self-regulation, 638-640
verbal, sexual dimorphism and, 66, 70f, 70-71 functional magnetic resonance imaging in, 142
in Tourette's syndrome, 554-556
Language development hippocampus in, 45, 56-57, 67-69, 100-103,
Inhibitory input, and event-related potentials,
125-126, 127f analytic style in, 303 111-112,115, 119
in children with unilateral brain injury, 281- long-term potentiation in, 68, 83
Input units, in neural network models, 160, 161f
Insulin-like growth factor-I 305, 316 and neuronogenesis, 93, 99-100, 99f-101f
and brain development, 492-493, 495 age of onset and, 296-299 nucleus basalis and, 194
zinc and, 499 crowding hypothesis of, 296 olfactory preference
Integer-list representation developmental sensitivity in studies of, 299 and attachment, 602-605
bootstrapping mechanism for, 426-429 emergentist view of, 282, 297, 299-304 neural plasticity underlying, 604-605
construction of, 421, 425-426 equipotentiality in, 282-283 sensitive period for, closing, 605
as new representational resource, 425 fresh-start hypothesis of, 298 perceptually cued, and habit establishment,
object-file representations versus, 423 IQ data from studies of, 284t-287t, 296-298, 550-552
sketch of one proposal for, 426 300 place, 115
Intelligence. See also IQ irreversible determinism in, 282-297 plasticity and, 199
early deprivation/orphanage rearing and, 620- language assessment data on, 288t-295f
postnatal, and attachment, 602-604
621 lesion etiology and, 298
procedural
Intensity discrimination, auditory, 206-207 lesion neurocorrelates and, 298
attention-deficit/hyperactivity disorder and,
Interaural level difference, 211 methodological confounds in studying, 292,
570
Interaural time difference, 211-212 297-299
inferences about, from animal cognition tests
Internal capsule prospective studies of, 299
applied to children, 111-112
myelination of, 43, 82 retrospective studies of, 299
neural basis for, 111
magnetic resonance imaging of, 37t, 38f, 40f sample size in studies of, 298-299
sexual dimorphism and, 66-69, 112-115
40-41 seizures and, 296, 298
spatial-relational, 115-120
sex hormones and, 72 timing of language testing for, 299
type, site and size of lesion and, 296, 298 neural circuit mediating, maturation of, 371
Internal working models, in attachment, 608, 624
critical period for, 282 stress and, 102
Interval mapping, 155
neural network model of, 165 types of, 111
Intervention, magnetic resonance imaging in,
early deprivation/orphanage rearing and, 622 word, 274-277
139-140
experience and, 192 Learning algorithms
functional, 141-143
holistic style in, 303 error-driven, 162
Intradimensional-extradimensional (IDED) task,
543, 545, 580 left hemisphere as specialized organ for, 281- in neural network models, 162-163
Intrauterine growth restriction, and brain 282, 304 in object processing model, 346-347
development, 492-495 nominal style in, 303 self-organizing, 162
Intrinsic neurotransmitters otitis media and, 304 Left hemisphere
developmental changes in, 84-85 plasticity and, 192, 196-199,276 damage
interactions of, with extrinsic neurotransmitters, pronominal style in, 303 and adult aphasia, 281-282
86-88 schizophrenia and, 570 and language development, 281-305, 300f
Introverts, 631 sensitive period for, 197-198 302f
Iodine, and brain development, 491-492, 495t, vowel triangle in, 197-198, 198f and neglect, 405-406
500 Language disorders, 309-316 versus right hemisphere damage, 282-297,
Ipsilateral corticospinal pathway, early brain injury auditory development and, 198-199, 208, 309- 300f, 300-304
and, 180 311 and spatial analysis, 406-410
INDEX 675
Left hemisphere (continued) of children, difficulties with, 315 for contextual detail, aging and, 483-484
perceptual detail advantage of, 304 combination of, with event-related potentials, declarative, 365-366
in spatial analysis, 406-407, 408f 133 adult abilities in, emergence of, 368-371
as specialized organ for language development, in face processing, 382, 385 development of, 368-371, 374, 375t
281-282,304 in older adults, interpretation of, 475 inferences about, from animal cognition tests
Lesion paradigm, 150 in phonetic perception, 273-274 applied to children, 111-112
LGN. See Lateral geniculate nucleus in reading disorders, 141-143, 143f-144f neural circuits mediating, maturation of, 369-
Light, detection of, absolute thresholds of, 222- in remediation/intervention, 141-143 371, 374
223 in Tourette's syndrome, 553, 554f episodic, 580
Limbic system of myelination, 35-43, 38f-40f in schizophrenia, 577, 580-581
aromatase activity in, 63 appearance in Tl-weighted images, 36 event-related potentials in, 130-131
in attention, 321-322, 322f appearance in T2-weighted images, 36 false, 484
in autism, 538 milestones in, 37t fetal alcohol syndrome and, 511, 513
in temperament, 633 in obsessive compulsive disorder, 139-140, 140f hippocampus in, 45, 56-57, 111, 119-120, 369-
zinc deprivation and, 499 versus other imaging modalities, 137 372, 374
Line bisection task, 450t pharmacologic, 143-145 item, aging and, 473, 483
Linear synergy, and reaching skills, 256-258, 257f of remediation/intervention, 139-140 long-term, 365-366
Lingual gyrus, development of, 80t rescanning, reliability of, 139 long-term potentiation in, 68, 83
Linguistic processing, event-related potentials in, in schizophrenia, 581, 584 neural bases of, insights into, from
129-130 terminology, 36 neuropsychological studies in primates,
Linkage analysis, 152, 153t, 154-156 Magnetoencephalography, in phonetic perception, 365-375
nonparametric, 153t, 154-155 273 plasticity and, 199
parametric, 153t, 154-155 Magnocellular pathway, 226-227, 228f. See also procedural, 365-366
Linkage, versus linkage disequilibrium, 155 Dorsal visual pathway adult abilities in, emergence of, 366-367
Linking propositions, in visual development, 221 development of, 221, 229-232 development of, 366-368, 374, 375t
Linoleic acid, and brain development, 491 visual deprivation and, 246 inferences about, from animal cognition tests
Linolenic acid, and brain development, 491 Malnutrition applied to children, 111-112
Lipoproteins, and brain development, 491 and brain development, 491-498, 501 neural circuits mediating, maturation of, 367-
Liquid conservation task, in dorsolateral protein-energy, 491, 493-498, 494t, 501 368, 374
prefrontal cortex studies, 442 postnatal, effects of, 495-498 recognition
Listening, preferences, in infants, 208-209, 216, prenatal, effects of, 493-495 adult abilities in, emergence of, 368-369
271 weaning, and risk of, 496 aging and, 483-484
Localization Mammilothalamic tract, myelination of, 82 attention and, 330-334, 332f, 334f
auditory Manganese, and brain development, 491 Corsi-Milner test of, 450t
in human infants, 212-213, 216 MAO-A. See Monoamine oxidase A development of, in monkeys and humans,
in nonhumans, 210-211 Marginal zone, 80 375t
spatial, development of, 401-404 Marker tasks, 326 neural circuits mediating, maturation of, 369-
Locus coeruleus of attention in young infants, 326-327 370
in attention, 354 Masking, auditory, 207 visual, 119-120
in closing of sensitive period for olfactory Mass action, principle of, 298 relational
preference learning, 605 Maternal attachment, parallel processes in, 605 adult abilities in, emergence of, 370-371
Long-chain polyunsaturated fatty acids (LCPUFA), Maternal separation responses, 607-609 development of, 375t
nutritional need for, 496 Mean, differential regression to, 153 neural circuit mediating, maturation of, 371
Long-term depression, synaptic, 192, 247 Medial diencephalon, in declarative memory, 368 semantic, in schizophrenia, 577, 581
Long-term potentiation, 192 Medial frontal gyrus, development of, 79 sensory, aging and, 475-476
estrogen and, 68 Medial geniculate nuclei, myelination of, 42 sexual dimorphism and, 66-69
glutamate receptors in, 83 Medial lemniscus, myelination of, 82 short-term, 365-366
Looking Medial longitudinal fasciculus, myelination of, 82
source, aging and, 473-474, 483-484
obligatory, 356 Medial orbital gyrus, development of, 80t
types of, 111
preferential Medial prefrontal cortex
working, 365-366, 541, 579
and attention, 357-358 in attention-deficit/hyperactivity disorder, 529
adult level of proficiency in, emergence of,
at eyes, 385 cocaine and, 527, 529
372-373
at faces, 383-385, 387, 416 dopamine in, 523, 525
age-related changes in frontal cortex and,
in global-local task, 449t Medial preoptic area
473-484
in studies of dorsal and ventral visual androgen receptors in, 60
sexual dimorphism in, 66 aging and, 473-484
pathways, 344-346
Medial temporal lobe attention and, 563-564
in visual assessment, 237
in autism, 538-539 in attention-deficit/hyperactivity disorder, 568
Lordosis, in rats, sex hormones and, 61-62
in declarative learning/memory, 111-112, 365, in autism, 541-543
368-371 development of, 371-374, 375t
M dopamine and, 436, 440-444, 448-454, 459-
in schizophrenia, 580-581
Macrocephaly, 157 Medium-wavelength sensitive cones, development 463
Macronutrients, and brain development, 491 of, 221 dorsolateral prefrontal cortex in, 366, 371-
Magnesium, and brain development, 491 Melody perception, 209-210 374, 436, 440-444, 448-454, 459-463,
Magnetic resonance imaging (MRI), 137-140 Memory 473
of behavioral development, 138-139 aging and PKU treatment and, 448-453, 449t-450t,
of brain development effects of working memory load and, 476-478 451f, 459-463
applications of, 137-145, 138f frontal cortex activation patterns in, 479, 480f in schizophrenia, 566-567, 578-580, 582
postnatal, 36-41 frontal lobe changes and, 473-484 spatial, 647, 650-655
of cognitive functioning, 70 future directions for study of, 484 Memory fields, 651
of fetal alcohol syndrome, 512 number of operations to be performed and, Menopause, behavioral changes in, 71-72
functional, 137, 140-143 478-479 Menstrual synchrony, 611
of behavioral development, 141, 141f type of material to be processed and, 479 Mental retardation, phenylketonuria and, 445
of brain development, 140-141, 141f categories of, 365-366 Mesencephalic dopamine, 649
676 INDEX
Mesencephalic reticular activating system, 321- and infant development, 609-610 phonetic perception and, 272-274
322, 322f, 325, 564 and long-term stress responses, 610 speech perception and, 271-272, 282
Mesocortical dopamine, 521-525, 649, 651-652 and maternal behavior in next generation, 609- Nativism, 156
Mesolimbic dopamine, 649 610 Nature versus nurture, 150, 164
Metachromatic leukodystrophy, as schizophrenia Motivation Nc. See Negative central
model, 585-586 and reward-seeking behavior, 647-659 Nd. See Negative difference
l-Methyl-4-phenyl-l ,2,3,6-tetrahydropyridine. See and temperament, 631 Negative central (Nc), in event-related potentials,
MPTP Motor coordination, fetal alcohol syndrome and, 129
Microcephaly, 157, 492 510 in attention, 326, 332-333, 334f-335f
Micronutrients, and brain development, 491-492 Motor cortex, 434, 435f development of, 131-132, 132f-133f
Middle cerebellar peduncle, myelination of, 42 development of, 262-263 in infants, 131, 131f
magnetic resonance imaging of, 37t, 39-40 Motor pathways, myelination of, 35-36 Negative difference (Nd), in event-related
Middle frontal gyrus, development of, 80t Motor preparation, event-related potentials in, 130 potentials, 128
Middle temporal gyrus, development of, 80t Motor root fibers, myelination of, 42 Negative slow wave. See NSW
Midline diencephalon, in declarative learning, 111 Motor skills Neglect, hemispatial, 405-406, 410, 427
Mind acquisition of Neglect of children, 617-627
theory of neural basis for, in first year, 262-264 Neocerebellum, dorsolateral prefrontal cortex
in dorsolateral prefrontal cortex studies, 441- stability and flexibility in, 253-265 connection to, 435
442 fine Neocortex
early deprivation/orphanage rearing and, 624 attention-deficit/hyperactivity disorder and, architecture of, 3-4, 4f 12-14, 16f 17, 520
understanding of, 415, 418 570 classical view of, 210
Minerals, and brain function, 491 in Tourette's syndrome, 554-556 histogenesis of
Minimum auditory angle, 212-213 schizophrenia and, 570, 583 neuronogenetic algorithm for, 7-19
Mismatch field (MMF), in event-related potentials, visually guided. See Eye movement control; postproliferative regulation of, 18
in phonetic perception, 273 Reaching skills scaling of, across species, 5-6, 8f
Mismatch negativity (MMN), in event-related Movement sequence of, 4-7, 6f
potentials, 128-129 and object perception, 344-345 specification in, 4t, 15
after treatment of specific language impairment skilled magnitude of, 15-16, 16f
and dyslexia, 315 neural basis for, in first year, 262-264 neuronal classes in, 3, 4f, 4t
amplitude, aging and, 475-476, 476f-477f stability and flexibility in acquisition of, 253- neuronogenesis in, 3-19
development of, 132 265 abbreviations related to, 7t
in language studies, 197-198, 270, 273, 313, 315 MPTP (l-Methyl-4-phenyl-l,2,3,6- algorithm for, 7-19
in phonetic perception, 273 tetrahydropyridine), in dopamine-dorsolateral cell cycle and, 7, 9f, 10, llf, 15-18, 17f
in speech processing, 270, 313 prefrontal cortex studies, 436-440, 447 equations in, 10
MMF. See Mismatch field MRI. See Magnetic resonance imaging master integrator of, 14-15
MMN. See Mismatch negativity Muscarinic receptors, developmental changes in, output domain of, 11-12
Model-dependent linkage analysis, 154-155 84 parameters in, 10, llf
Model-free linkage analysis, 154-155 Muscles, activation changes, with onset of reaching in pseudostratified ventricular epithelium, 4-
Molecular methods, of studying genetic influences skills, 258-259, 258f-259f 5,6f,7-19,11f
on brain/behavioral development, 151, 153t, Music, perception of, auditory development and, restriction point of, 9f 14-15
154-156 209-210 and structural variation, 15-18, 16f
Molybdenum, and brain development, 491 Musicians temporospatial domain of, 11-14, 13f
Monkey(s) cortical plasticity in, 193-195, 196f phenotype, variation and regulation of, 5t
face-responsive neurons in, 382 focal hand dystonia in, 194 in procedural learning, 111
Myelin, 35 species-specific variations in, 3-4, 15-18, 16f
hippocampal cell death in, 48
composition of, 35 specification of, 4t, 15
memory development in, 365-375, 375t
formation of, 35 structure of, 3-4, 4f
synaptogenesis model in, 23-31
Myelin basic protein, 35 postproliferative regulation of, 18
testing children with tasks developed for, 109-
Myelination variations in, neuronogenetic determinants
111
CNS, sequences of, 41-43 of, 15-18, 16f
Monoamine(s), developmental roles of, 524-525
of corticolimbic pathways, 82 synaptogenesis in, 18
Monoamine oxidase, 522, 523f
in developing human brain, 35-43 in newborn, 23-32
Monoamine oxidase A (MAO-A), disruption of
diffusion-weighted imaging of, 37, 39, 43 phase 3 of, evolution of, 29-31, 30f, 31t
gene encoding, 524-525
general rules of, 35-36 volume/surface area of, 3-4, 5t, 15
Monoaminergic neurotransmitters. See also
histochemical definition of, 35 Neonatal Behavioral Assessment Scale, 634-635
Dopamine; Serotonin
iron and, 492, 498-499 Neonativism, 156
maturation and interactions of, 79, 89
magnetic resonance imaging of, 35-43, 38f-40f Nerve growth factor, and estrogen, 65
Morphine, and consummatory attachment, 655- Net input, in neural network models, 161-162
appearance in Tl-weighted images, 36
656 Neural network models
appearance in T2-weighted images, 36
Morris search task applications and contributions of, 163-166, 169,
milestones in, 37t
apparatus and procedure in, 117 339
nutrition and, 491-492, 494-495, 497-499
for children, 115, 117-118, 120 challenges to, 167-168
of visual system, 35-36, 40-43, 230
adaptations in, 118, 119f benefits of, 159-160, 169
Myelogenetic cycle, 42
results of, 118, 120t challenges to, 166-167
neural basis of, in animals, 117-118 of change mechanisms, 163-166
N
Mossy cells, development of, 52-54, 53f 56 of cognitive development, 159-170
and spatial-relational learning, 119 N100, in event-related potentials, 128 complexity of, 159, 169
Mossy fiber terminals, hippocampal, development N200, in event-related potentials, 130, 132, 133f control in, 159, 169
of, 52-56, 55f N350, in event-related potentials, 130 criticism of, 166-168
Mother(s), attachment of, 605 N400, in event-related potentials, 129-130, 132 elements of, 160-163, 169
Motherese, 197-199, 209 N-acetyl aspartate, in schizophrenia, 580-581 challenges to, 167
Mother-infant interaction Native language explicitness of, 160, 169
and attachment, 599-612 experience and, 276-277 failure of, insights from, 166-167
hidden regulators in, 607-612, 609t versus foreign language, 269, 272-274 generalization in, 167-168
INDEX 677
Neural network models (continued) developmental cognitive Norm-based encoding hypothesis, of face
learning algorithms in, 162-163 advances in, 339 processing, 390-391
limitations of and improvements to, 169-170 fetal alcohol syndrome studies in, 505-513 Novelty
net input and activation functions in, 161-162, goals of, 149, 417 behavioral inhibition and, 640
162f neural network models in, 159-170, 339 fear of, early deprivation/orphanage rearing
of origin of knowledge, 163-164 number representations in, 418-425 and, 623
understanding behavior from, 159, 169 understanding of mind in, 415, 418 Novelty P3, 131
units and weights in, 160, 161f systems in older adults, 479-481, 481f-482f
Neural plasticity. See Plasticity in closing gap between biology and Novelty preference
Neural tube defects, folic acid deficiency and, 492 psychology, 417 and attention, 330-334, 357-358
Neurogenesis. See Neuronogenesis integrating cognitive development with, 427- in recognition memory testing, 330-334, 368-
Neuromodulators, and recovery from early brain 429 370
injury, 175, 185-186, 186f 188 Neurosteroids, 64, 65f NSW (negative slow wave), in event-related
Neuromotor system, interconnectedness and Neurotransmitter(s) potentials, 129
redundancy in, 254-255, 264-265 activity, pharmacologic magnetic resonance in infants, 131, 131f
Neuron (s) imaging of, 144 Nuclear magnetic resonance (NMR), 137
aromatase activity in, 63 in attention-deficit/hyperactivity disorder, 567 Nucleotides, nutritional need for, 496
changes in, during development, 175-177, 176f convergence of, 88-89 Nucleus accumbens
dopamine in, 522-523, 525, 649-650
connections, development of, in hippocampus, 52 functional implications of, 88-89
in opiate reward pathway, 656
cortical, maturation of, 80-82, 81f cortical, maturation and interactions of, 79,
Nucleus basalis, 194
face-responsive, in monkeys, 382 82-88
Number representation
formation of. See Neuronogenesis developmental changes in, 79, 82-85
analog-magnitude, 421-425, 427-429
neocortical, classes of, 3, 4f, 4t developmental similarities among systems of,
choice experiment on, 419, 420f, 422, 424
number of, genetics and, 157 85-86
disorders of, 422, 427-428
sex hormones and, 61, 63-69 extrinsic
as example of bridging gap between cognition
size of, experience and, 177 developmental changes in, 84-85
and developmental neuroscience, 418-425
substance P-containing, testosterone and, 61 interactions of, with intrinsic
Gelman and Gallistel's continuity hypothesis of,
as units in neural network models, 160, 161f neurotransmitters, 86-88
419
Neuronal specific enolase, 96 genetics and, 157 Wynn's ease of learning argument against,
Neuronogenesis intrinsic 419-421
abbreviations related to, 7t developmental changes in, 83-84 infant, format of, 424-425
adult, 93-103 interactions of, with extrinsic integer-list
discovery of, in mammalian brain, 94f-95f neurotransmitters, 86-88 bootstrapping mechanism for, 426-429
94-96 in schizophrenia, 565-566 construction of, 421, 425-426
functional significance of, 100-103 sex hormones and, 64-65, 72 as new representational resource, 425
new method for detection of, 94f, 95-96 Neurotrophin(s) sketch of one proposal for, 426
in cerebellum, 93, 367-368 in cortical development, 521 object-file, 423-425, 427-429
cortical, 80-82, 81f, 519-520, 520f visual deprivation and, 247 prelinguistic, 418-425
prefrontal, 373 Neurotrophin-3, 521 possible systems for, 421-424
early brain injury and, 176, 180-183, 182f, 187 Newborns. See also Infants structural analogy from, 426-429
experience and, 93, 98-100,103, 177-178 attachment system in, 601-602 Numerons, 419
hippocampal, 45-47, 48f-50f, 56, 369, 371 auditory development in, 205-216 Nutrients
in adults, 93-103 face processing in, 383-384 categories of, 491-493
hormones and, 65-69, 93, 97, 102-103 Conspec/Conlern hypothesis of, 383 role of timing in deprivation and subsequent
learning and, 93, 99-100, 99f-101f, 103 externality effect in, 384 repletion of, 500-501
neocortical, 3-19 limitations of, 384 selected, and brain development, 493-500,
algorithm, 7-19 mental representation of facedness in, 383 494t-495t
cell cycle and, 7, 9f 10, 11f, 15-18, 17f perception of emotion in, 384 Nutrition, and cognitive development, 491-501
equations in, 10 recognition of identity in, 383-384 critical periods for, 497, 500-501
master integrator of, 14-15 sensory hypothesis of, 383 role of covariates in studies of, 497
output domain of, 11-12 social hypothesis of, 383
parameters in, 10, llf listening preferences in, 208-209, 216, 271 O
in pseudostratified ventricular epithelium, 4- neocortical synaptogenesis in, 23-32
Object continuity, neural network model of, 163-
5,6f,7-19, llf Nigrostriatal tract, dopamine in, 522, 525, 649
164, 169
restriction point of, 9f, 14-15 NMDA receptors
Object permanence, 428
and structural variation, 15-18, 16f cocaine and, 528
neural network model of, 164, 169
temporospatial domain of, 11-14, 13f estrogen and, 68
Object processing
in normal brain, 175-176, 176f subunit 1, as marker of neuronogenesis, 96
dorsal and ventral visual pathways in, 339-349
nutrition and, 492 NMR. See Nuclear magnetic resonance computational model of, 346f, 346-348
stress and, 93, 97-98, 99f 102-103 Noise, sensory, in infants, 206-207, 221, 224 versus face processing, 381-383
in visual system, 230 Nominal style, in language development, 303 movement and, 344-345
Neuronogenetic sequence, 6f 8-10 Nonparametric linkage analysis, 153t, 154-155 Object recognition module, in object processing
Neuropeptides, sex hormones and, 64-65, 72 Noradrenaline, and recovery from early brain model, 346f, 346-347
Neuropil, increase of, in cortex, 80, 81f injury, 175, 185-186, 186f, 188 Object retrieval task, 449t
Neuroscience Noradrenergic system in dopamine-dorsolateral prefrontal cortex
cognitive and attention, 322-323, 323f studies, 435-440, 440f, 441t, 447-448, 451f
genetics and, 149-150 maturation of, 84-85 PKU treatment and, 448, 451f
meanings of attention in, 561-565 Norepinephrine Object reversal discrimination task
developmental cocaine and, 528 for children, 112-115, 114f, 120t
and cognition, bridging gap between, 415- as neurotransmitter, maturation of, 84-85 gender specificity in, 112-115, 114f
429 in olfactory preference learning and gonadal hormones and, 113
structural analogy in, 415-417 attachment, 604-605 neural basis of, 113
678 INDEX
Object-based attention, 427 individuality/temperament and, 360 development of, 221, 229-232
Object-file representations, 423-425, 427-429 network of, 354, 355f visual deprivation and, 246
versus integer-list representation, 423 preferential, to faces, 383-384 Pattern vision. See also Contrast sensitivity; Visual
numerical properties of, 423 Orphanage-reared children acuity
Obligatory looking, 356 attachment of, 624-626 development of, 228
OBP. See Observer-based procedure behavior problems in, 623 P3b, in event-related potentials, 130-131
Observer-based procedure (OBP), in auditory catch-up growth in, 620-623 Pediatric autoimmune neuropsychiatric disorders
assessment, 206-207, 212 cognition in, 620-622 associated with streptococcal infections
Obsessive-compulsive disorder domains of development in, 620-626 (PANDAS), magnetic resonance imaging in,
cognitive deficits of schizophrenia versus, 588 effects of early deprivation in, 617-627 139
PET studies of, following pharmacologic exemplar studies of, 618-619 Peer problems, early deprivation/orphanage
treatment, 144 general intelligence in, 620-621 rearing and, 624
progression of, magnetic resonance imaging of, growth failure in, 622-623 PEM. See Protein-energy malnutrition
139-140, 140f language development in, 622 Perception, categorical, 388
with Tourette's syndrome, 549 levels of privation for, 618 of emotional expressions, 388-389, 392-393
Occipital gyrus, development of, 80t peer problems in, 624 Perceptual detail advantage, of left hemisphere,
Occipital poles, myelination of, 35-36 socio-emotional development in, 623-626 304
Occipital white matter, myelination of, magnetic stimulation of, framing questions about, 619 Perceptually cued learning, and habit
resonance imaging of, 37t, 41 studies of, history and conceptual issues of, establishment, 550-552
Ocular dominance columns 617-620 Perirhinal cortex
development of, experience and, 192 Other race/other species effect, in face in declarative learning, 111
visual deprivation and, 246 processing, 389-390 and memory, 368, 370
Oculomotor nerves, myelination of, 42 Otitis media, and language development, 304 Perseveration
Oddball paradigms Output domain, of neuronogenetic operation, in A-not-B task, 657
in event-related potential studies, 128-131, 313, 11-12 executive attention and, 360
332-333 Output mode, of cell cycle, 9f, 10 fetal alcohol syndrome and, 511, 513
novelty, in aging research on working memory, Output units, in neural network models, 160, 161f in reaching skills, 260-262
479-481 Ovarian steroids, and neuronogenesis, 97, 98f development of, 261-262, 263f
Oddity task O-wave, 130 hand trajectories and, 261, 262/-263f
for children, 1201, 120-121 Owls, sensory deprivation experiments with, 211- repetition and, 260-261
in relational memory assessment, 370-371, 375t 212 in working memory assessment, 373
Old age, 474-475 Oxidative metabolism, nutrition and, 492, 498 Personality. See also Temperament
Olfactory bulb, neuronogenesis in, 93-94 Eysenck's model of, 631
Olfactory pathway, 605-606, 606f P PET. See Positron emission tomography
Olfactory preference learning Pharmacologic magnetic resonance imaging, 143-
P300, in event-related potentials, 128, 130-131 145
and attachment, 602-605
development of, 131-133, 133f-134f in attention-deficit/hyperactivity disorder,
neural plasticity underlying, 604-605
in older adults, 479-481, 481f-482f Ritalin and, 144
sensitive period for, closing, 605
p21, in neuronogenesis, 14-16 Phenylalanine
Olfactory tubercle, dopamine in, 522 p27, in neuronogenesis, 14-16
Omega-3 fatty acids, nutritional need for, 496 animal model of mild, chronic plasma
P3a, in event-related potentials, 131 elevations of, 454-456, 456f
Open field, for event-related potentials, 125-126 novelty, in older adults, 479-481, 481f-482f
Opiate (s) and cognitive performance, linear relationship
Paired-comparison test, 449t with, 450-452
and consummatory attachment, 655-656, 658 of recognition memory, 330-334, 368-369, 375t
mu receptor, 655-656 diet low in, for treatment of phenylketonuria,
Pallidum, myelination of, 42 445-446
and olfactory preference learning, 604 PANDAS, 139
Optic nerve, myelination of, 42 metabolism, 444-445. See also Phenylketonuria
Parabrachial nucleus ratio with tyrosine, 446-447, 460
magnetic resonance imaging of, 40-41 androgen receptors in, 60 Phenylketonuria, 444-445
Optic radiations, myelination of, 42-43 estrogen receptors in, 59 genetic studies of, 150-151, 157
magnetic resonance imaging of, 40-41 Paracentral gyrus, development of, 80t treatment
Optic tectum, 211-212 Parahippocampal gyrus, development of, 79, 80t animal model of, 454-456, 456f
Optic tracts, myelination of, 42 Parallel distributed processing modeling, 159. See cognitive deficits with, 434, 444-463
magnetic resonance imaging of, 40-41 also Neural network models developmental delay versus lasting deficits
Optical transfer function, 223 Parametric linkage analysis, 153t, 154-155 with, 452
Optokinetic nystagmus, early asymmetries in, 228, Paraventricular nucleus diet low in phenylalanine for, 445-446
232 androgen receptors in, 60 early and continuous, as model of dopamine
Orbital frontal gyrus, development of, 79-80, 80t estrogen receptors in, 60 function in prefrontal cortex, 434, 444-
Orbital frontal pathways, early deprivation/ mother-infant interactions and, 610 463, 447/
orphanage rearing and, 624 Parentese (motherese), 197-199, 209 four-year longitudinal study of children in,
Orbital prefrontal cortex, in learning, 111-112, Parietal cortex 447-454, 449t-450t, 451 f
114 in attention, 354, 355f, 359, 404-405, 427, 563 limitations of, 445-446
Orbitofrontal cortex in attention-deficit/hyperactivity disorder, 567 mechanism for cognitive effects of, 446-447,
cardioinhibitory centers in, 325 dorsolateral prefrontal cortex connection to, 447f
in go/no-go task, functional magnetic 435 preservation of performance on self-ordered
resonance imaging of, 141, 141f in motor skill acquisition, 263 pointing and temporal order memory
in habit formation, 550-552, 551f, 552t in number representation, 427-428 tasks with, 453-454, 455t
Organizational effects, 60 in schizophrenia, 581 and retinal function, 456-461, 457f-458f
of sex steroid hormones, 60-62 in Tourette's syndrome, 555 selective versus global deficits with, 452-453
Orientation discrimination, visual, development Parietal-directed stream, visual, 226-227. See also unexpected findings in study of, 453-454
of, 229 Magnocellular pathway Phoneme discrimination, 192, 196-199
Orienting, 353-354 Parkinson's disease, dopamine in, 522, 652 Phonetic perception, 272-274, 276
development of, 356-357 Partial cortical functioning, 404, 410 Phonological awareness skills, deficits in, and
hierarchical relation of, with executive control, Parvocellular pathway, 226-227, 228f. See also language/reading disorders, 310-311
360 Ventral visual pathway Phosphorus, and brain function, 491
INDEX 679
Photoreceptors, development of, 222-224 Posterior parietal cortex, dorsolateral prefrontal medial
Physical fitness, and cognitive effects of aging, 474 cortex connection to, 435 androgen receptors in, 60
Piaget, Jean, 402 Post-rolandic gyrus, development of, 80* sexual dimorphism in, 66
PKU. See Phenylketonuria Posture, and reaching skills, 258-259 progesterone receptors in, 60
Place learning, 115 Potassium, and brain function, 491 Preplate, 520-521
Planum temporale, sexual dimorphism in, 70, 70f Precentral gyrus, representation of motor Pre-rolandic gyrus, development of, 80t
Plasticity responses in, 192 Prism studies
acetylcholine and, 185-186 Precentral sulcus, 434, 435f of reaching skills, 254
adaptive and maladaptive, 194 Preferential listening, in infants, 208-209, 216, of sensory development, in owls, 211-212
in amputees, 193, 193f ,195, 195f 271 Privation, levels of, in orphanages, 618
of auditory system, 195-196, 196f, 205, 210-216 Preferential looking Probabilistic epigenesis, 150, 156
in humans, 215-216 and attention, 357-358 Procedural learning
in nonhumans, 213-215 at eyes, 385 attention-deficit/hyperactivity disorder and, 570
basic aspects of, 193-194 at faces, 383-385, 387, 416 inferences about, from animal cognition tests
behavioral relevance and, 194-195, 312 in global-local task, 449t applied to children, 111-112
in blind, 194, 215-216 in studies of dorsal and ventral visual pathways, neural basis for, 111
in children, capacity for, versus adult capacity 344-346 Procedural memory, 365-366
for, 194, 316 in visual assessment, 237 adult abilities in, emergence of, 366-367
in deaf, 215-216 Prefrontal cortex development of, 366-368, 374
deafferentation and, 193-194 in monkeys and humans, 375t
anatomy of, 434-435, 435f
and development, 191-199 inferences about, from animal cognition tests
in attention, 436, 563-564
early brain injury and, 175-188, 281-282, 297- applied to children, 111-112
in attention-deficit/hyperactivity disorder, 567,
299 neural circuits mediating, maturation of, 367-
570-571
early deprivation and, 619 368, 374
Ritalin and, 144
experience-driven or training-induced, 311-312 Processing negativity (PN), in event-related
cocaine and, 527
after speech perception training, 313 potentials, 128
cytoarchitectural maturation of, 80-82, 81f
for aphasia, 316 Progenitor cells
in declarative memory, 370-371
characteristics of training necessary to induce, development of, in normal brain, 175-176,
development of, 79-89 176f, 519-521, 520f
312-313,315
dopamine in early brain injury and, 183
Hebbian mechanisms in, 192
and cognition, 435-456, 438f,-440f, 441t Progesterone
and language development, 196-199, 276
459-463 and brain development, 60-62, 64, 65f,
and learning, 199
early and continuous treatment of PKU 67-68, 72
and memory, 199
model of, 434, 444-463, 447f and dopamine, 67
in musicians, 193-196, 196f
evidence from visual psychophysics organizational and activational effects of, 61 -62
noradrenaline and, 185-186
confirming hypothesis of, 456-459, 457f- and verbal abilities, 70
in normal developing brain, 175-178
458/ Progesterone receptors
in olfactory preference learning, 604-605
and inhibitory control, 436-440, 448-450 in brain, 60
principles of, 194-195
and reward-seeking behavior, 647-659 dopamine and, 65, 72
regulation of, 192-193
role of, 433-463, 523, 525 Projection pathways, myelination of, 35-36
of representational zones, 191-194
selective deficit in, with PKU treatment, 444- Pronominal style, in language development, 303
examples of, 195-196
447, 445f Prepositional memory. See Declarative memory
sensitive periods for, 192, 199
tyrosine levels and, 446-447, 447f, 460 Prosopagnosia, 382
of somatosensory system, 192-193, 193f, 195,
and working memory, 436, 440-444 developmental, 392
195f
in spatial cognitive development, 410-411 dorsolateral. See Dorsolateral prefrontal cortex Prostate, 5a-reductase metabolism of testosterone
functions of, 434 and, 63-64
subcortical, vertical, 228-229
in go/no-go task, magnetic resonance imaging Protein, and brain development, 491, 493-498,
temporal dynamics of, 194
in, 140-141, 141f 494t
use-dependent, 193-195
in learning, 111-113 Protein-energy malnutrition (PEM)
of visual system, 240-248, 348-349
in motor skill acquisition, 262-263 and brain development, 491, 493-498, 494t, 501
Pleiotropy, 156
myelination of, 82 postnatal, effects of, 495-498
PN. See Processing negativity
neurotransmitters in prenatal, effects of, 493-495
Pointing tasks, self-ordered, preservation of
convergence of, 88-89 weaning, and risk of, 496
performance on, in children treated for PKU,
developmental similarities among systems of, Protein-energy status, and brain development,
453-454, 455* 493-498, 494t
Poles, myelination of, 35-36 85-86
extrinsic, 84-85 Proteolipid protein, 35
Polygeny, 156 Proton magnetic resonance spectroscopy, in
Polymorphic genes, 149 interactions of, 79, 86-88
intrinsic, 83-84 schizophrenia, 580-581
Polysialylated form of neural cell adhesion Proximal pathways, myelination of, 35-36
molecule (PSA-NCAM), 96 maturation of, 79, 82-85
Proximity, with mother, and attachment, 601-602,
Positive slow wave. See PSW in schizophrenia, 579
611, 656
Positron emission tomography (PET) in self-regulation, 638-640
Pruning, synaptic, 82, 373, 393
of children, difficulties with, 315 size of, evolutionary increase in, 434-435
Pseudostratified ventricular epithelium (PVE),
in obsessive compulsive disorder, following in spatial working memory, 650-652
7-10
pharmacologic treatment, 144 synaptic connections in, formation of, 82 architecture of, 7-10, 8f
of visual development, 227-228 in temperament, 633 cell cycle of, 7, 9f, 10, 11f, 15-17, 18f
Posterior attention system, 322, 324, 404 in Tourette's syndrome, 553, 554f neuronogenesis in, 4-5, 6f, 7-19
Posterior cerebellar peduncle, myelination of, Pregnenolone, and brain development, 64, 65f algorithm for, 10-19
magnetic resonance imaging of, 37t Premonitory urges, in tic disorders, 552-553, 553f equations in, 10
Posterior cingulate cortex, dorsolateral prefrontal Premotor cortex, dorsolateral prefrontal cortex master integrator of, 14-15
cortex connection to, 435 connection to, 435 output domain of, 11-12
Posterior fossa structures, myelination of, Preoptic area parameters in, 10, 11f
magnetic resonance imaging of, 39-40 aromatase activity in, 63 temporospatial domain of, 11-14, 13f
Posterior orbital gyrus, development of, 80 t estrogen receptors in, 60 operational properties of, 7-10
680 INDEX
positional coding mechanism of, 12-13 stability and flexibility in acquisition of, 253-265 seeking system in, 648-651, 657t
proliferative cells in, neighborhood two-step paradigm of, 254 structure of, 648
relationships of, 8f 9 Reactivity Rey-Osterreith Complex Figure, 409
size of, calculation of, 10 hypothalamic-pituitary-adrenal axis in, 634-635 Rhodopsin, developmental increase in, 223
PSW (positive slow wave), in event-related in temperament, 632-635 Right hemisphere
potentials, 131 vagal tone as measure of, 634 damage
development of, 131-132, 133f Reading disorders. See also Dyslexia and language development, 281-297, 300-
in infants, 131, 131f functional magnetic resonance imaging in, 141- 304, 300f
Psychological short stature, 622-623 143, 143f-144f and neglect, 405-406
Psychology, biology and phonological awareness deficits and, 310-311 and spatial analysis, 406-410
bridging gap between, 415-417 specific language impairment and, 310-311 in face processing, 382, 388-391
diagnostic marker research on, 417 Recognition memory in spatial analysis, 406-407, 408f
structural analogy between, 415-417 adult abilities in, emergence of, 368-369 in spatial attention, 404
Psychosocial dwarfism, 622-623 aging and, 483-484 Ritalin, in attention-deficit/hyperactivity disorder,
Puberty attention and, 330-334, 332f ,334f functional magnetic resonance imaging of
behavioral changes in, 71-72 Corsi-Milner test of, 450t effects of, 144
early-onset, early deprivation/orphanage development of, in monkeys and humans, 375 t Rodents, testing children with tasks developed for,
rearing and, 623 neural circuits mediating, maturation of, 369- 115-120
synaptic decline in, 24-25, 28-29, 82 370 Romanian orphanages, effects of early deprivation
Pupillary dilation visual, 119-120 in, 617-620
in attention assessment, 564 Recurrent weights, in neural network models, 160, Rothbart, Mary, 631
in schizophrenia studies, 566 161f Rule evocation, 121
Purkinje cells, development of, 367 Reflexive saccadic eye movements, 329-330, 330f- Rule utilization, 121
Putamen 331f
development of, 367-368 Regulation, of affect/emotions, 611 S
in habit formation, 550-552, 551f 552t attention and, 353-354, 358-360, 638-639 Scaling, synaptic, 192-193
in obsessive compulsive disorder, magnetic cognitive prerequisites for, 638 Schizophrenia, 565
resonance imaging of, 140f early deprivation/orphanage rearing and, 623- age of onset, modal, 565
in Tourette's syndrome, 553, 554f 626 age-associated changes and, 585
PVE. See Pseudostratified ventricular epithelium fetal alcohol syndrome and, 506 animal models of, 586-587
P450X1XAI, 63 and temperament, 632-634, 636-640 associationism and, 569
Pycnotic nuclei, as indicator of cell death, in Regulators, hidden, in mother-infant interaction, attention in, 561, 565-567, 578-579, 587-588
hippocampus, 47-49, 51f 607-612, 609t brain development and, 577, 581-585
Pyramidal cell(s) Relational memory cardinal symptoms of, 577
cortical adult abilities in, emergence of, 370-371 childhood-onset, 565
development of, 80 development of, in monkeys and humans, 375t compromised neurocognitive systems implicated
plasticity of, 192 neural circuit mediating, maturation of, 371 in, 577-581
dopamine and, 525 Relational/configural/holistic encoding cortical developmental anomalies in, 519
event-related potentials from, 125-126 hypothesis, of face processing, 382 course of illness, 583-585
hippocampal Remediation, magnetic resonance imaging in, cross-sectional studies of, 584
development of, 52-54, 54f, 56-57, 369 139-140 diagnosis of, 565, 569
formation of, 45-46 functional, 141-143 dopamine in, 523, 525, 565-566, 586-587, 652
sex hormones and, 68 Representational zones episodic memory in, 577, 580-581
neurotransmitters to, maturation and plasticity of, 191-194 event-related potentials as markers for, 128
interactions of, 83-88 examples of, 195-196 executive functions in, 566, 579-580, 584
Pyramidal tract, myelination of, 35-36, 43 sensory, 192 genetic studies of, 152
somatosensory, 192, 193f ,195, 195f genetics of, 565
R Resolution, auditory, 207-208 and language learning, 570
Respiratory sinus arrhythmia, in autonomic longitudinal studies of, 583-584
Radial-arm maze
responsivity, 634 metachromatic leukodystrophy model of, 585-
for children, 115-117
Response, in spatial-relational learning, 115 586
apparatus and procedure in, 115-117, 116f
Response inhibition. See Inhibitory control morbid function in, 583
results with, 117, 120t
Response integration network, in object and motor skills, 570, 583
cued versus uncued testing in, 116-117 neurobiology of, 565-566, 570-571
forced-choice testing with, 116 processing model, 346f, 347
Restriction point, of neuronogenesis, 9f, 14-15 neuropsychology of, 577-588
free-choice testing with, 116 premorbid dysfunction with, 582-583
Reaching skills Reticular formation, myelination of, 42
cognitive studies of, 582-583
in adults, 254, 255f Retina(s)
neuromotor, 583
development of control in, 254-259 development of, 221-224, 239-240
prevalence of, 565
developmental changes in, 260-262 limitations in models of, 222
semantic memory in, 577, 581
in dopamine-dorsolateral prefrontal cortex function, PKU treatment and, 456-461, 457f- specificity of cognitive impairment to, 587-588
studies, 435-440 458f versus affective disorder, 587
dorsal pathway and, 340-341, 349, 401-404 Retinocortical pathway, 226-227, 228f versus Alzheimer's disease, 587
from early movement (wiggling) to, 255-256 Retrosplenial cortex, dorsolateral prefrontal versus attention-deficit/hyperactivity disorder,
linear synergy constraint on, 256-258, 257f cortex connection to, 435 587-588
muscle pattern change at onset of, 258-259, Reversible cooling, in dorsolateral prefrontal versus Huntington's disease, 587
258/-259f cortex studies, 436 versus obsessive compulsive disorder, 588
neural basis for, in first year, 262-264 Reward-seeking behavior, 647-659 versus temporal lobe epilepsy, 587
perseverative, 260-262 affect and, 648, 653, 658 specificity of neural development to, 585-587
development of, 261-262, 263f affective/emotional component of, 648 static lesion model of, 584-585
hand trajectories and, 261, 262f-263f consummatory attachment in, 655-656 windows of vulnerability and, 585-587
repetition and, 260-261 dopamine-opiate modulations of, 647-659 working memory in, 566-567, 578-580, 582
posture and, 258-259 interaction of cognitive seeking and Search task, in working memory assessment, 372-
selecting stable pattern for, 255-256 consummatory attachment in, 656-657 373
INDEX 681
Second messenger systems, sex hormones and, Sexual behavior processing of
64-65 dopamine and progesterone receptors in, 65 brain injury and, 406-410
Seeking system, 648, 657t sexual dimorphism in medial preoptic area development of, 406-410
cognitive extensions of, 650-651 and, 66 global level versus local level, 406-407, 408f
in human, 648-649 Sexual dimorphism/differences Spatial attention, development of, 404-405, 410
neurobiology of, 649 in bird song system, 66-67 Spatial cognition, 399
Segregation analysis, 152, 152t in brain, 59, 61, 65-71 Spatial cognitive development, 399-411
Seizures, and language impairment in children examples of, 66-71 associated with dorsal visual stream, 401-406
with brain injury, 296, 298 in cerebral cortical areas, 69-71, 70f associated with ventral visual pathway, 406-410
Selectionist hypothesis, of synaptogenesis, 29 in cognition plasticity in, 410-411
Selective attention, 323-324, 561-562, 565 higher functions, 69-71, 70f Spatial discrimination task, 449 t
feature integration theory of, 562 inferences about, from animal tests applied to Spatial incompatibility, 462
in modification of blink reflex, 327-329, 328f children, 112-115 Spatial location skills, development of, 401-404
in schizophrenia, 566-567 Spatial skills
in hippocampus, 67-69
Selective stabilization hypothesis, of hippocampal morphological development and,
in learning and memory, 67-69, 112-115
synaptogenesis, 29 56-57
in midbrain dopaminergic systems, 67
Selenium, and brain development, 491, 495t, 500 sex differences in, 69, 102-103
in recovery from early brain injury, 186-187,
Self-ordered pointing tasks, preservation of Spatial vision, development of, 237-248
187f dorsal pathway and, 339-349
performance on, in children treated for PKU, Shame, effortful control and, 361
453-454, 455t Spatial working memory, 647
Short stature, psychological, 622-623 development of, 653-655
Self-ordered working memory task, 372-373 Short-wavelength sensitive cones, development of,
Self-organizing learning algorithm, 162 dopamine and, 650-652
221 Spatial-relational learning, 115-120
Self-regulation, 611 Sibling pair studies, 155
attention and, 353-354, 358-360, 638-639 in animals, neural basis of, 115
Simon effect, 462 in children
cognitive prerequisites for, 638 Simultanagnosia, 427 Morris search task in testing of, 117-118
early deprivation/orphanage rearing and, 623- Single photon emission computed tomography, in neural mechanism underlying, 118-119
626 autism, 538 radial-arm maze in testing of, 115-117, 116f
fetal alcohol syndrome and, 506
Single unit recording, in dorsolateral prefrontal neural circuit mediating, maturation of, 371
and temperament, 632-634, 636-640
cortex studies, 436 Spatial-relational memory
Semantic memory, in schizophrenia, 577, 581
Sink, in event-related potentials, 126, 127f adult abilities in, emergence of, 370-371
Sensitive periods development of, in monkeys and humans, 375<
Six boxes task, 450t
for language development, 197-198 neural circuit mediating, maturation of, 371
performed by children treated for PKU, 453-
for plasticity, 192, 199 Specialization
454
for visual development, 211-212, 244-245 hemisphere
Sketch of one proposal, in number representation,
Sensorimotor cortex, in habit formation, 550-552, in emotions, 635
426
551f, 552t in face processing, 382, 388-389, 393
Skill learning, 111. See also Procedural learning
Sensory fibers, myelination of, 35-36, 42 in language development, 281-282, 304-305
Skilled movement, acquisition of
Sensory hypothesis, of face processing, 383 sex differences in, 70
neural basis for, in first year, 262-264
Sensory memory, aging and, 475-476 in speech processing, 270-271
stability and flexibility in, 253-265
Sensory system, development of, 192-193 in visual-motor integration, 555-556
SLI. See Specific language impairment
Separation responses, 607-609 neural, in auditory system, 210-216
SMA. See Supplementary motor area
Septal nuclei Specific language impairment (SLI), 310
Smooth pursuit eye movements, and attention,
androgen receptors in, 60 auditory/speech development and, 198-199,
329-330, 330f-331f 309-311
estrogen receptors in, 59
Sequence learning, auditory, 210 Social hypothesis, of face processing, 383 diagnosis of, 310
Serotonin Social skills temporal processing deficits in, 198-199, 309-
as neurotransmitter and executive function, 543-545 311
interactions and convergence of, with fetal alcohol syndrome and, 511 treatment of, 309
dopamine, 79, 88-89 Socially reticent children, 641-642 experience-driven plasticity in, 314-316
maturation of, 79, 85-86 Socio-emotional development, early deprivation/ Speech disorders
in schizophrenia, 565-566 orphanage rearing and, 623-626 auditory development and, 198-199, 208
Sex hormone (s) Sodium, and brain function, 491 in schizophrenia, 581
and behavior over lifespan, 71-72 Sokolov, Y. N., 353 Speech, infant-directed, 197-199, 209, 271
and brain development, 59-72, 138 Solitary tract Speech perception, 271-272, 282
androgen receptors in, 60 training
general principles of, 72
estrogen receptors in, 59 for aphasia, 316
and concurrent discrimination, 113-114
Somatosensory cortex, monoamines and, 524-525 experience-dependent plasticity after, 313
and gene expression, 64
Somatosensory system, plasticity of, 192-193, 193f for specific language impairment and
and higher cognitive functions, 69-71
195, 195f dyslexia, 314-316
and learning and memory, 66-69, 113-114
Soothing behavior, attention and, 358-359 and word learning, 274-276
mechanisms of, in modulation of neuronal
Sorting tasks Speech processing
activity, 64-65 experience and, 276-277
metabolism of, in mediating effects on brain, aging and, 473-474, 478-479, 483
hemispheric specialization in, 270-271
62-64, 62f, 63f, 65f, in attention-deficit/hyperactivity disorder, 568
in infancy, neurocognitive approach to, 269-277
and neuronogenesis, 65-69, 93, 97, 102-103 computer facilitation in, 544
versus processing of nonspeech, 269-271, 276
nongenomic actions of, 64 in dorsolateral prefrontal cortex studies, 440-
and specific language impairment or dyslexia,
and object reversal discrimination, 113-114 441 208, 309-316
organizational versus activational effects of, frontal lesions and, 473 temporal theory of, 208, 309-310
60-62 measurement imprecision in, 540-541 Spike potential, in ERP studies of eye movement,
receptors, in brain, 59-60 in schizophrenia, 580, 587-588 342-343, 343f-344f
and recovery from early brain injury, 175, 186- Source, in event-related potentials, 126, 127f Stability
188, 187f Source memory, aging and, 473-474, 483-484 in acquisition of skilled movement, 253-265
and second messenger systems, 64-65 Span of apprehension, in schizophrenia, 566 as constraint on reaching skills, 256-258, 257/
and sexual dimorphism, 65-71 Spatial analysis, 406 versus flexibility, 255, 262
682 INDEX
Stages, developmental, neural network model of, Superior colliculus, 211, 213-214 regulation of, 23
164-166, 169 and spatial attention, 404-405 selectionist versus constructivist hypothesis
Startle paradigm, 632 in visual development, 226-229, 228f, 356 of, 29
Starvation, and brain development, 492 Superior frontal gyrus, development of, 80t sex hormones and, 65
Static lesion model, of schizophrenia, 584-585 Superior medullary lamina, myelination of, 82 Systems neuroscience
Stem cells Superior temporal cortex, dorsolateral prefrontal in closing gap between biology and psychology,
development of, in normal brain, 175-176, 176/ cortex connection to, 435 417
early brain injury and, 181-183 Superior temporal gyrus, development of, 79, 80t integrating cognitive development with, 427-
Stereopsis, development of, 226, 226f 231 Superior temporale, sexual dimorphism in, 70-71, 429
Stereoscopic vision, 192 70f
Sternberg's memory search task, in older adults, Supervisory attention, 562-564
476-478 Supplementary motor area (SMA), 434, 435f T
Steroid hormone(s) in attention, 354-355
metabolism of, and effect of sex hormones on in self-regulation, 638 Tl, in magnetic resonance imaging
brain, 62-64, 62f-63f 65f in Tourette's syndrome, 551f, 552t, 553 Tl relaxation time, 137
and neuronogenesis, 65-69, 93, 97, 102-103 Supramarginal gyrus, development of, 79, 80t Tl-weighted images, of myelination, 36-41, 38f-
sex. See Sex hormone (s) Supraoptic nucleus, estrogen receptors in, 60 40f
Stimulation, framing questions about, 619 Supratentorial brain, myelination of, 42 T2, in magnetic resonance imaging
Stimulus onset asychrony, 404-405 T2 relaxation time, 137-138
magnetic resonance imaging of, 40
Stress T2-weighted images, 140
Surface features, perception of, ventral visual
of myelination, 36-41, 38f-40f
and learning, 102 pathway and, 341, 344-349
Tapping task, 449t
and neuronogenesis, 93, 97-98, 99f 102 Sustained attention, 324-325, 327-329, 437, 561,
in dorsolateral prefrontal cortex studies, 442-
Stress responses 564-565
444, 443f 448, 451f
early deprivation/orphanage rearing and, 623 SVZ. See Subventricular zone
PKU treatment and, 448, 451f
mother-infant interactions and, 609-610 Syllable discrimination, 270, 314-315
Tectal visual system, 399
Stress-hyporesponsive period, 609 Synapse (s)
Telencephalon
Stria terminalis, bed nucleus of density of
androgen receptors in, 60
androgen receptors in, 60 days after conception and, 26f
myelination of, 35-36, 42-43
aromatase activity in, 63 decline in, with age, 24-25, 28-29, 82
Temperament
estrogen receptors in, 59-60 and visual development, 225-226, 226f-227f
affective bias in, 635-636
in fear system, 632, 641 230, 240
and attention, 353-354, 360-361
sex hormones and, 59-61, 63 formation of. See Synaptogenesis
behavioral inhibition in, 631-632, 640-642
sexual dimorphism in, 66 mismatch of pre- and postsvnaptic activity of,
biology of, 631-642
Striatum visual deprivation and, 247 integrative approach to, 632-633
in attention-deficit/hyperactivity disorder, 567 number of, genetics and, 157 bottom-up regulation of, 633
dopamine in, 522, 525 plasticity of, regulation of, 192-193 components of, 634-640
in habit formation, 550-552, 551f, 552t as weights in neural network models, 160, 161f emotion-based approach to, 631-632
in memory, 366, 374 Synaptic pruning, 82, 373, 393 motivation-based view of, 631
sexual dimorphism in, 66-67 Synaptic scaling, 192-193 reactivity in, 632-635
Stroke, cortical reorganization after, functional Synaptic space, early brain injury and, 179-180 Rothbart's theory of, 631
magnetic resonance imaging of, 142 Synaptogenesis self-regulation in, 632-634, 636-640
Stroop tasks cortical, 82, 519-521, 520f strength of nervous system and, 631
aging and, 481-483 distinct waves of, 25-26, 25f, 32 top-down regulation of, 633
control condition, 450t early brain injury and, 179-180 Temporal dynamics, of plasticity, 194
in dopamine-dorsolateral prefrontal cortex environment and, 24f 27-29, 31-32, 150 Temporal lobe
studies, 462-463, 462t, 463f-464f epigenesis and, 23, 31-32 in autism, 538-539
in dorsolateral prefrontal cortex studies, 442 experience-dependent, 24f 27-28, 177-178 in declarative learning/memory, 111-112, 365,
in executive control, 359 experience-expectant, 24f 27 368-371
Structural analogy, 415-417 experience-independent, 24f, 27 dorsolateral prefrontal cortex connection to,
from number representation, 426-429 genetics of, new perspectives on, 31-32 435
Structural parallels, 415-417 heterochronic epigenesis hypothesis of, 32 in schizophrenia, 565, 580-581
Subcortical mechanisms, in face processing, 383 in hippocampus, 52, 54-57, 55f, 68, 371 volume of, 138
Subcortical plasticity, vertical, 228-229 and individuation, 31-32 Temporal lobe epilepsy, cognitive deficits of
Subcortical visual development, 221, 226-232 kinetics of, 24-25, 24f-2Sf 31 schizophrenia versus, 587
Subcortical white matter, myelination of, magnetic levels of, 23 Temporal order memory tasks, 450t
resonance imaging of, 41 neocortical preservation of performance on, in children
Subicular complex in newborn, 23-32 treated for PKU, 453-454
cell migration in, 49-52, 56 and structural variation, 18 in schizophrenia, 580
neuronal cell formation in, 45-47, 48f-49f 56 in normal brain, 176-177 Temporal pole, myelination of, 35-36
Substance P-containing neurons, testosterone nutrition and, 492 Temporal processing
and, 61 phase 1 of, 24-25, 24f 27 deficits, in speech and language disorders, 208,
Substantia nigra phase 2 of, 24-25, 24f 27 309-311
dopamine in, 522 phase 3 of, 24, 24f 31-32 training
in habit formation, 550-552, 551f 552t critical periods in, 27-28 for aphasia, 316
in Tourette's syndrome, 553, 554f days after conception and, 26f, 29-30 for specific language impairment and
Subthalamic nuclei, in habit formation, 551f, 551- in diverse cortical areas, 26 dyslexia, 314-316
552 duration of, 30-31, 31t Temporal resolution, auditory, 207-208
Subthalamic region, myelination of, 42 evolution of, in neocortex, 29-31, 30f 3H Temporal theory, of speech processing, 208, 309-
Subventricular zone (SVZ), cell formation and genetic hypothesis of, 30 310
migration in, 49, 49f 521 modifiability of, 26-27 Temporal-directed stream, visual, 227. See also
Sulcus, 434 morphofunctional hypothesis of, 30-31 Parvocellular pathway
Superior cerebellar peduncle, myelination of, 42 phase 4 (plateau) of, 24-25, 24f 28 Temporospatial domain, of neuronogenetic
magnetic resonance imaging of, 37t, 39 phase 5 of, 24-25, 24f 28 operation, 11-14, 13f
INDEX 683
10/20 system, of electrodes, for recording event- Training-induced plasticity, 311-312 Ventromedial hypodialamus
related potentials, 126 after speech perception training, 313 sex hormones and, 61-62
Terminal zones, myelination of, magnetic for aphasia treatment, 316 sexual dimorphism in, 66
resonance imaging of, 41 characteristics of training necessary to induce, Ventromedial nucleus
Testosterone 312-313,315 androgen receptors in, 60
aromatization of, to estrogen, 62-63, 62f-63f in treatment of specific language impairment progesterone receptors in, 60
66-67 and dyslexia, 314-316 sexual dimorphism in, 66
and brain development, 59-64, 66-69 Trajectory projection module, in object processing Verbal instructions, for animal cognition tests
5a-reductase pathway for, 62f, 63-64, 65f model, 346f, 347 applied to children, 110, 119-120
in learning and memory, 69, 113-114 Transformation-variant representation, 341 Verbal skills, sexual dimorphism and, 66, 70-71,
metabolism of, in brain tissue, 62-64, 62f -63f Transmission disequilibrium test, 153t, 156 70f
65f Transverse neuronogenetic gradient (TNG), 10, Vestibulocochlear system, myelination of, 42
organizational and activational effects of, 60-61 8f, 12-14, 13f, 16-18 View-centered representation, 341
and recovery from early brain injury, 186-187, Transverse patterning task, in relational memory Violation of expectancy studies, of number
187f assessment, 370-371, 375t representation, 419
and sexual dimorphism, 66 Transverse temporal gyrus, development of, 80t Visual acuity
Thalamocortical radiations, myelination of, 42-43 Trapezius muscles, in reaching skills, 258, 258f development of, 223-224, 229-230, 237-248
Thalamocorticopyramidal fiber system, Tricep muscles, in reaching skills, 258, 258f mechanisms underlying, 239-240
myelination of, 36 Trigeminal nerve, myelination of, 42 visual input and, 240-248, 349
Thalamus Trisynaptic circuit, hippocampal, 52, 55-57 effects of congenital cataracts on, 241-248, 349
in attention, 321-322, 322f 360, 561, 563-564 Tryptophan hydroxylase, activity, developmental after treatment, 241-244, 241f
in auditory development, 213-214 changes in, 85 competitive interaction between eyes with,
in habit formation, 550-552, 551f 552t Turned On After Division 64 kD, 94f, 96 243-244, 247-248
in isolation call of infants, 608 Twin studies, 151-154, 152t immediately after end of visual deprivation,
medial dorsal, in declarative learning, 111 DF method of, 153-154 240-241
myelination of, 42—43 2-back task mismatch of pre- and postsynaptic activity
magnetic resonance imaging of, 40 in older adults, 478 with, 247
in schizophrenia, 565-566, 570-571 in schizophrenia, 579 in nondeprived eye, 244
Theory of mind Two-step paradigm, of reaching skills, 254 permanent deficits from, mechanisms
in dorsolateral prefrontal cortex studies, 441- underlying, 245-248
Tyrosine
442 during sensitive period, 244-245, 245f
and dopamine production, 446-447, 447f 460,
early deprivation/orphanage rearing and, 624 grating, development of, 237-248, 238f
522, 523f
Thorny excrescences, of hippocampal cells, long-chain polyunsaturated fatty acids in diet
phenylketonuria and, 444-447, 447f
development of, 52-54, 53f 56 and, 496
ratio with phenylalanine, 446-447, 460
Thought disorders, in schizophrenia, 581 Visual cortex
Tyrosine hydroxylase, 85, 368, 522, 524
Three boxes task, 449t development of, 221, 224-232, 240
performed by children treated for PKU, 453- anatomical data on, 230
454 u neuronogenesis in, 230
3-back task, in schizophrenia, 579 Unilateral brain injury. See Brain injury, unilateral neurotransmitters in, glutamate, maturation
Three pegs task, 449t Units, in neural network models, 160, 161f of, 83
in dorsolateral prefrontal cortex studies, 442, Use-dependent plasticity, 193-195 primary, synaptogenesis in, 24-31
443f 444, 448, 45 If Use-it-or-lose-it hypothesis, 474-475 visual deprivation and, 246-248
PKU treatment and, 448, 451f Visual discrimination tasks, in procedural memory
Thymidine labeling, 94-95 V assessment, 366-368, 375t
Thyroid hormone synthesis, iodine and, 492, 500 Visual field (s)
Tics, 549-558 Vagal tone, as index of autonomic responsivity, 634 divided studies, in face processing, 382, 388
cortical-striatal-thalamo-cortical circuits in, 553, Vagus nerve, 325 receptive, enhancement of, and attention, 323-
554f Ventral tegmental area 324
and disregulation of neural circuits underlying and consummatory attachment, 655-656 Visual input, and visual development, 240-248,
habit formation, 552-554 dopamine/dopamine receptors in, 436, 446- 349
dopaminergic mechanisms in, 553-555 447, 522, 525, 649-652, 655-656 Visual paired-comparison test, 449t
as fragments of normal behavior, 552 in opiate reward pathway, 656 of recognition memory, 330-334, 368-369, 375t
neuroanatomic circuits in, 550-552, 551f 552t and spatial working memory, 651-652 Visual pathways
premonitory urges before, 552-553, 553f Ventral visual pathway, 339-349 dorsal, 339-349, 399, 400f
self-similarity in temporal patterns of, 556 anatomy of, 399-401, 400f anatomy of, 399-401, 400f
timing of, fractal nature of, 556, 556f and body-centered frames of reference, 344,
computational model of object processing in,
Time mode, of cell cycle, 9f 10 345f 348-349
346f 346-348
TNG. See Transverse neuronogenetic gradient computational model of object processing in,
development of, 341-348, 367-368
Tone contour, 209-210 346f, 346-348
dissociation of, with dorsal pathway, behavioral
Tone sequences, 209-210 development of, 341-348
Tourette's syndrome, 549-558 evidence of, 344-346, 349 dissociation of, with ventral pathway,
versus autism, 541-543 in face processing, 342, 381 behavioral evidence of, 344-346, 349
cortical-striatal-thalamo-cortical circuits in, 553, function of, 399 and eye movement control, 339-340, 342f-
554f high-density ERP studies of, 342-344 343f 342-344, 349, 401
dopaminergic mechanisms in, 553-555 neurocomputational properties of, 339-341 function of, 399
natural history of, 556-557, 557/ in procedural memory, 367-368 high-density ERP studies of, 342-344, 343f-
neuroanatomic circuits in, 550-552, 551f 552t spatial processes associated with, development 344f
premonitory urges in, 552-553, 553f of, 406-410 neurocomputational properties of, 339-341
timing of tics in, fractal nature of, 556, 556f in visual processing, 339-341 plasticity of, 348-349
visual-motor integration in, 554-556 as what or perception pathway, 340, 399-401 processing in, 340-341
Trace elements, and brain development, 491-492 Ventricular zone (VZ), cell formation and spatial processes associated with, development
Tracking, visual migration in, 49, 49f of, 401-406
and attention, 329-330, 330/-331f Ventrolateral nucleus, progesterone receptors in, in visual processing, 339-341
dorsal pathway and, 340 60 as where or action pathway, 340, 399
684 INDEX
genicostriate, 399 and memory, 367-370 Wisconsin General Testing Apparatus, for
magnocellular, 226-227, 228f pathway interaction and overlapping in, 230- children, 109-115, 110f 120
development of, 221, 229-232 231 caveats of, 115
visual deprivation and, 246 prenatal alcohol exposure and, 507, 512 inferences about declarative and procedural
major, 226-227, 228f retinal, 221-224, 239-240 learning from, 111
parvocellular, 226-227, 228f retinal model versus cortical model of, 224- inferences about ontogenesis sex-specific
development of, 221, 229-232 225 cognition from, 112-115
visual deprivation and, 246 sensitive period for, 211-212, 244-245 nonverbal procedures in, 110
tectal, 399 and spatial cognition, 399-411 reinforcement and lack of food restriction in,
ventral, 339-349 subcortical, 221, 226-232 110
anatomy of, 399-401, 400f synaptic density and, 225-226, 226/-227f 230,
repeated days of testing in, 110-111
computational model of object processing in, 240
versus test for monkeys, 109-110
346f 346-348 temporal lags between emergence of function
Within-family association methods, 153t, 156
development of, 341-348, 367-368 and behavioral evidence in, 231
Word learning, 274-277
dissociation of, with dorsal pathway, visual input and, 240-248, 349
behavioral evidence of, 344-346, 349 Working memory, 365-366, 473, 541, 579
myelination of, 35-36, 40-43, 230
and face processing, 342, 381 Visual tracking adult level of proficiency in, emergence of, 372-
function of, 399 and attention, 329-330, 330f-331f 373
high-density ERP studies of, 342-344 dorsal pathway and, 340 aging and, 473-484
neurocomputational properties of, 339-341 Visually guided behavior. See Eye movement effects of working memory load and, 476-478
and procedural memory, 367-368 control; Reaching skills frontal cortex activation patterns in, 479, 480f
spatial processes associated with, development Visual-motor integration, in Tourette's syndrome, frontal lobe function in, 473-484
of, 406-410 554-556 future directions for study of, 484
in visual processing, 339-341 Vitamin A, and brain development, 492 number of operations to be performed and,
as what or perception pathway, 340, 399-401 Vitamin B6 and brain development, 492 478-479
Visual processing Vitamins, and brain development, 492 type of material to be processed and, 479
dorsal and ventral routes of, 339-341 Vna, in autonomic responsivity, 634 attention and, 563-564
dual route paradigm of, 339 Vocabulary burst, 302 in attention-deficit/hyperactivity disorder, 568
Visual recognition memory, 119-120 Vocal communication, origins of, 607-608 in autism, 541-543
Visual system Voluntary saccadic eye movements, 329-330, 330f- development of, 371-374
and attachment, 602 331f in monkeys and humans, 375t
and auditory development, 211-216 Vomeronasal system, sexual dimorphism in, 66 dorsolateral prefrontal cortex in, 366, 371-374,
development, dorsal and ventral pathways in, Vowel discrimination, 270, 272
473
339-349 Vowel triangle, 197-198, 198f
development of, 225f dopamine and, 436, 440-444, 448-454, 459-
Vulnerable periods. See Critical periods
absolute thresholds in, 222-223 463
VZ. See Ventricular zone
acuity in, 223-224, 229-230, 237-248, 238f maturation and, 373-374
and attention, 323-324, 329-330, 356-359 w PKU treatment and, 448-454, 449t-450f, 451f
459-463
botdeneck developmental theories of, 231-
232 Weber's law, 421-422 in schizophrenia, 566-567, 578-580, 582
brain-behavior relationships in, 221-232 Weights, in neural network models, 160, 161f spatial, 647
contrast sensitivity in, 223-224, 229-231, Wernicke's area, sexual dimorphism in, 70-71 development of, 653-655
237-248, 238f White matter dopamine and, 650-652
cortical, 221, 224-232, 240 myelination of, magnetic resonance imaging of, Wynn's ease of learning argument, against
cortical parcellation hypothesis of, 231 37-43, 37t, 38f-40f continuity hypothesis of number
critical period for, 211-212 volume of, developmental changes in, 138 representation, 419-421
differential maturation in, 226-232 Wiggling, of infants, and development of reaching
early, 221-232 skills, 255-256 Y
experience and, 192, 240-248, 619 Wisconsin Card Sorting Test
and eye movement control systems, 329-330, aging and, 473-474, 478-479, 483 Yale Global Tic Severity Scale, 557
330/-331f in attention-deficit/hyperactivity disorder, 568
Hansen and Fulton model of, 222-223 in autism, 540-541, 544 z
linking propositions in, 221 computer facilitation in, 544
long-chain polyunsaturated fatty acids in diet frontal lesions and, 473 Zebra finches, sexual dimorphism in song system
and, 496 measurement imprecision in, 540-541 of, 66-67
mechanisms underlying, 239-240 in schizophrenia, 580, 587-588 Zinc, and brain development, 491, 494t, 499-500
INDEX 685