27, 168171 (1998)

PM970271

PREVENTIVE MEDICINE
ARTICLE NO.

How a Child Builds Its Brain: Some Lessons from Animal Studies of
Neural Plasticity
James E. Black, M.D., Ph.D.1
Department of Psychiatry, Neural Science Program, and Beckman Institute, University of Illinois at UrbanaChampaign,
Urbana, Illinois 61801

Although the potential vulnerability of childrens

brain development is generally recognized, relatively
little is known about the timing, resiliency, or mechanisms involved. While animal research should be applied only cautiously to human policy, some findings
do have important clinical implications. This paper
briefly reviews animal studies demonstrating the effects of experience on brain structure. Contemporary
theories emphasize the self-organizing potential of
brain structure, particularly regions that seem to have
evolved for the purpose of storing information. We emphasize three major findings: (1) many regions of the
brain are responsive to experience, but they differ in
the types of information stored and in their developmental timing. (2) One type of plasticity is typically
embedded in a developmental program, and it requires
appropriate timing and quality of the information
stored for the animals development to be normal. (3)
Another category of plasticity stores information that
is idiosyncratic and unpredictable, but is often useful
for species such as humans that learn throughout their
life span. We therefore expect that some aspects of human brain development use the first type of plasticity
and that abnormal experience or deprivation may
cause lasting harm to brain and behavior. However,
because the other type of plasticity lasts a lifetime,
efforts such as psychotherapy or social interventions
may help heal a wounded brain. q1998 American Health Foundation and Academic Press

Key Words: neural plasticity; synaptogenesis; systems

theory; critical periods; early intervention; cognitive
development; developmental psychopathology.

The idea of early experience causing enduring effects

on human development has long been at the core of
psychiatry. Not only is experience often used to explain
the nature of a given psychopathology, but experience
1
To whom reprint requests should be addressed at Beckman Institute, University of Illinois at Urbana-Champaign, 405 N. Mathews,
Urbana, IL 61801. E-mail: jblack@s.psych.uiuc.edu.

is used correctively, e.g., using psychotherapy to help

heal the wounds of early experience. A central issue,
however, has been whether the remarkable neural plasticity observed in childhood persists into maturity or
whether some types of early trauma or deprivation are
irreversible in their effects on the brain.
Although it has commonly been assumed that the
effects of early experience must persist via enduring
biological changes in the brain, the mechanisms of that
change have been largely unknown until recently. We
will argue here that experience alters brain structure
to form persisting memories, not in a monolithic or rigid
fashion, but rather utilizing multiple, flexible brain systems that can encode different types of experience and
often on different developmental schedules [1].
It is clear that much of brain development is directed
by information from the genome, and some aspects of
experience merely trigger developmental processes
without storing any detailed information. Waddingtons
concept of embryonic canalization is applicable to
many of these developmental processes, as typically
they are relatively resistant to experience. The adaptive
value of such canalized, genetically directed brain development is obvious, as these types of processes will
result in uniform, species-specific brain anatomy, protected from the vagaries of the environment.
Clearly, however, not all information about brain
structure can be stored in the genome. In some species,
specialized brain structures have evolved to store information in patterned synaptic connections, information
that originates from external experience, not from
genes. We have previously argued [1] that two important types of neural plasticity can be observed in
mammalian brain: The type termed experience-expectant is tied closely to the brains developmental timetable. The other type of plasticity, termed experiencedependent, allows individual members of a species to
flexibly incorporate useful but idiosyncratic information into the brain.

168

0091-7435/98 $25.00
Copyright q 1998 by American Health Foundation and Academic Press
All rights of reproduction in any form reserved.

HOW A CHILD BUILDS ITS BRAIN

EXPERIENCE-EXPECTANT NEURAL PLASTICITY

During development, some aspects of experience can

provide useful information for fine-tuning brain anatomy. For example, when kittens open their eyes, they
reliably experience an abundant presence of contrast
and contour in their visual fields, and this information
can then be used to refine neural connections. Correspondingly, if early visual experience is interfered with,
then the neural systems depending on that information
will develop an abnormal structure and potentially abnormal behavior. The most well-known examples of
such disruptions in development are the experiments
of Hubel and Wiesel [2], in which monocular deprivation
affected visual cortex development in kittens. Many
diverse species, such as honeybees or songbirds, have
evolved such specialized information-storage mechanisms. Their anatomy can be quite different, but they
share a common developmental task: at a certain point
in development, information will be typically be provided to the young animal, which will store it away as
modified patterns of neural connections.
Developmentalists [e.g., Ref. 3] recently have been
advocating for a self-organizing, systems-theory perspective of brain development. While guided and controlled to an extent by genetic information, a great deal
of brain structuring or neural patterning occurs as an
interaction of the child with the environment. A common feature of cortical development in mammals, including humans, is the early overproduction of synaptic
connections followed by a substantial decrease in their
number, allowing information to be stored by selecting
useful connections and pruning back surplus connections. In addition, as different regions of the brain become relatively mature, they can serve to organize and
stabilize information for other brain regions to incorporate [1,3,4]. In an older but still elegant series of experiments [5,6], disrupting the integrated sequence of visual, proprioceptive, and other experiences in kittens
caused pathological behavioral development. In general
terms, some species have apparently evolved self-organizing programs for brain development, components
of which sometimes incorporate expected forms of experience, with other more canalized brain regions developing independently of experience, some brain regions
serving as scaffolding to stabilize or organize information for other components, and still others using experience to fine-tune their anatomy for optimal function.
Note that if a species relies heavily on the quality and
timing of experience for organizing its brain, it also
becomes quite vulnerable to disruption of it.
Since many aspects of neural plasticity are preserved
across species, we can expect humans to have maintained elements of them [e.g., Ref. 7]. Some features of
the human brain are certainly genetically determined
and canalized (as described above, it is advantageous

169

for a species to have some neural components rendered

uniform and predictable), but children learn a phenomenal amount of information during their first years of
life. Consider the massive scale of such tasks, for example, of mastering vocabulary or social interactions. The
comparatively overdeveloped human cerebral cortex
apparently has the role of an enormous information
storage device, a neural organ that facilitates the infants rapid acquisition of large amounts of information
[1]. The newly organized cerebral cortex is then ready
to incorporate other types of information that can be
important but are relatively unpredictable and idiosyncratic.
EXPERIENCE-DEPENDENT NEURAL PLASTICITY

In contrast to expected experience, other types of

information may be useful only for the survival of individual animals, not being developmentally predictable
in either timing or quality. These experience-dependent
processes need to be flexible, such that at any time
neural connections may need to be altered to reflect the
unique experiences of an animal. Because they often
store information that has adaptive value but is unpredictable in timing or quality, experience-dependent processes cannot be programmed into a developmental
schedule and must remain inherently flexible. In this
sense experience-dependent processes are more like
what is commonly meant by learning and memory,
and experience-expectant processes are more like critical periods. Some examples of valuable but experiencedependent types of information include a squirrels
memory of where it buried its nuts, a childs acquisition
of Spanish vocabulary rather than English, or a medical
student memorizing gross anatomy. From our animal
studies, we believe that this type of unpredictable information is stored in the formation of new neural connections, rather than the pruning back of surplus connections.
Rats, for example, experience substantial brain
changes when exposed to a comparatively enriched
environment in which animals are housed in groups
with ample opportunities for exploration of new objects
and for social interactions. The effects of this experience
include increased brain weight and cortical thickness
[8], production of new synapses [9], more complex dendritic branching [10], and new blood vessels [11]. Many
of these effects on behavior and brain structure can
still be demonstrated in late adulthood or even old age.
Animals raised in complex environments are superior
at many different types of learning tasks [reviewed in
Ref. 1]. These abilities are generalized across a wide
range of learning tests, however, suggesting that the
rats do not simply rely on specific information from
their rearing environment, but rather the enriched rat
appears to have learned to learn better.

170

JAMES E. BLACK

Changes in synaptic connections have also been observed in animals learning different tasks. For example, training in complex mazes requiring visuospatial
memory has been found to result in increased dendritic
branching in the visual cortex of adult rats [12]. When
the cerebral hemispheres of rats are surgically disconnected from each other and vision is occluded in one
eye, the visual cortex receiving information about maze
training from the nonoccluded eye shows greater
growth in dendritic branching than the other side of the
visual cortex, which received essentially no information
from the occluded eye [13]. Training animals on motor
learning tasks has also been found to result in sitespecific synaptic changes. Rats trained to use one forelimb to reach for cookies show dendritic growth within
the region of the cortex found to control that forelimbs
function [14].
When middle-aged rats were trained in difficult acrobatic tasks [15], they clearly demonstrated substantial
improvement in behavioral performance, and they had
increased the number of synapses in the cerebellum by
about 25%. Rats in an exercise control group had much
less learning to do, but they exercised much more than
the acrobatic rats. Another set of rats also served as an
inactive control, with very little opportunity for either
learning or exercise. Because the cerebellum helps control movement, the exercised rats used the preexisting
cerebellar synapses much more than the acrobatic or
inactive rats did, but this increased repetitious activity
in the exercised animals apparently produced no new
synaptic connections. Vigorous exercise, however, did
produce new blood vessels, presumably to support the
increased metabolic demand. These results suggest
that learning new skills, rather than mere repetitive
exercise, produced synaptogenesis in the cerebellar
cortex.

abuse or deprivation), particularly during periods of

rapid creation or modification of synaptic connections.
While there are some substantial issues in leaping from
animal research to human policy, our reading of this
literature suggests that interventions should be early
and substantial, as the damage of pathological experience may be long-lasting and quite difficult to undo.
Unfortunately, pathological experience may become
something of a vicious cycle, as the pathology induced
in brain structure may distort the childs experience,
with consequent alterations in cognition or social interactions, causing additional pathological experience and
additional brain pathology. Consider, for example, a
child with early deprivation or abuse, who becomes depressed and withdrawn and consequently does not extract the optimal kind of information from his or her
environment. Scaffolding of one component assisting
anothers information retrieval [1,3] could fall apart
with deprivation or abuse, for example, if components
for emotional regulation or attention were damaged
and consequently failed to stabilize other structures.
However, the animal literature does offer some hope
for breaking such cycles and healing the damage. Clinicians speak of corrective experience, and we would
argue that many interventions, such as Head Start,
psychotropic medications, and psychotherapy, alter the
childs experience in a positive way, and, like the training our animals received, such positive experience may
help repair or strengthen neural componenets previously damaged. If such corrective experience accumulates, it may help restore the developmental path of
brain development and benefit the child.
ACKNOWLEDGMENTS
Preparation of this paper was supported by a grant from the National Association for Research on Schizophrenia and Depression. In
addition, many of the ideas expressed here are the results of long
collaboration with Bill Greenough and his colleagues.

CONCLUSIONS AND CLINICAL IMPLICATIONS

REFERENCES

The animal studies described here demonstrate considerable plasticity in multiple areas of the mammalian
brain and demonstrate that neural plasticity is likely
to be even more substantial in humans. For this reason,
we suspect that positive, enriching experience will
likely produce more synaptic connections in human
children. Indeed, we suspect that the incorporation of
massive amounts of information into cerebral cortex is
a normal part of the self-organizing program of child
brain development. On the other hand, aberrant experience or deprivation will probably affect a young childs
brain anatomy as well, and here the issue of sensitive
periods arises again from the animal literature. While
adults certainly retain some neural plasticity and can
be traumatized by experience, children are likely to be
far more vulnerable to pathological experience (either

1. Black JE, Greenough WT. Induction of pattern in neural structure by experience: implications for cognitive development. In:
Lamb ME, Brown AL, Rogoff B, editors. Advances in developmental psychology, Vol. 4. Hillsdale, NJ: Earlbaum, 1986:150.
2. Hubel DH, Wiesel TN. The period of susceptibility to the physiological effects of unilateral eye closure in kittens. J Physiol
1970;206:419436.
3. Thelen E, Smith LB. A dynamic systems approach to the development of cognition and action. Cambridge, MA: MIT Press, 1994.
4. Turkewitz G, Kenny PA. Limitations on input as a basis for
neural organization and perceptual development: a preliminary
theoretical statement. Dev Psychol 1982;15:357368.
5. Held R, Hein A. Development and segmentation of visually controlled movement by selective exposure during rearing. J Comp
Physiol Psychol 1963;73:181187.
6. Hein A, Diamond R. Contributions of eye movements to the representation of space. In: Hein A, Jeannerod M, editors. Spatially
oriented behavior. New York: Springer, 1971:119134.

HOW A CHILD BUILDS ITS BRAIN

7. Karni A, Meyer G, Jezzard P, Adams M, Turner R, Ungerleider
LG. Functional MRI evidence for adult motor cortex plasticity
during motor skill learning. Nature 1995;377:155158.
8. Bennett EL, Diamond MC, Krech D, Rosenzweig MR. Chemical
and anatomical plasticity of brain. Science 1964;146:610619.
9. Turner AM, WT. Greenough. Differential rearing effects on rat
visual cortex synapses. I. Synaptic and neuronal density and
synapses per neuron. Brain Res 1985;329:195203.
10. Greenough WT, Volkmar FR. Pattern of dendritic branching in
occipital cortex of rats reared in complex environments, Exp Neurol 1973;40:491504.
11. Black JE, Sirevaag AM, Greenough WT. Complex experience promotes capillary proliferation in young rat visual cortex. Neurosci
Lett 1987;83:351355.

171

12. Greenough WT, Juraska JM, Volkmar FR. Maze training effects
on dendritic branching in occipital cortex of adult rats. Behav
Neural Biol 1979;26:287297.
13. Chang F-LF, Greenough WT. Lateralized effects of monocular
training on dendritic branching in adult split-brain rats. Brain
Res 1982;232:283292.
14. Greenough WT, Larson JR, Withers GS. Effects of unilateral and
bilateral training in a reaching task on dendritic branching of
neurons in the rat motor-sensory forelimb cortex. Behav Neur
Biol 1985;44:301314.
15. Black JE, Isaacs KR, Anderson BJ, Alcantara A, Greenough WT.
Learning causes synaptogenesis, while motor activity causes angiogenesis, in cerebellar cortex of adult rats. Proc Nat Acad Sci
USA 1990;87:55685572.