REVIEW ■

The Control of Voluntary Eye Movements:

New Perspectives
RICHARD J. KRAUZLIS
Systems Neurobiology Laboratory
Salk Institute for Biological Studies

Primates use two types of voluntary eye movements to track objects of interest: pursuit and saccades.
Traditionally, these two eye movements have been viewed as distinct systems that are driven automatical-
ly by low-level visual inputs. However, two sets of findings argue for a new perspective on the control of
voluntary eye movements. First, recent experiments have shown that pursuit and saccades are not con-
trolled by entirely different neural pathways but are controlled by similar networks of cortical and subcorti-
cal regions and, in some cases, by the same neurons. Second, pursuit and saccades are not automatic
responses to retinal inputs but are regulated by a process of target selection that involves a basic form of
decision making. The selection process itself is guided by a variety of complex processes, including atten-
tion, perception, memory, and expectation. Together, these findings indicate that pursuit and saccades
share a similar functional architecture. These points of similarity may hold the key for understanding how
neural circuits negotiate the links between the many higher order functions that can influence behavior and
the singular and coordinated motor actions that follow. NEUROSCIENTIST 11(2):124–137, 2005. DOI:
10.1177/1073858404271196
KEY WORDS Pursuit, Saccade, Eye movement, Attention, Perception

Primates make two kinds of voluntary eye movements to
place the retinal images of objects of interest onto the
fovea and to keep them there: saccades and pursuit.
Saccades are discrete ballistic movements that direct the
eyes quickly toward a visual target, thereby translating
the image of the target from an eccentric retinal location
to the fovea within tens of milliseconds. Pursuit is a con-
tinuous movement that rotates the eyes smoothly and
slowly to compensate for any motion of the visual target
and thus minimizes the drift of the target’s image across
the retina that might otherwise blur the image and com-
promise visual acuity.
Much of what we have learned about voluntary eye
movements over the past 40 years has involved treating
these movements as visuomotor reflexes that act to min-
imize visual “error” signals. Indeed, many species can
generate smooth optokinetic eye movements, which help
stabilize the eyes during head and body movements by
minimizing the motion of the entire visual surround.
However, when we move about in most natural environ-
ments, it is impossible to eliminate the slip of images
across the retina. Instead, choices need to be made about
which visual inputs have top priority. Primates appear to
be unmatched in their ability to identify individual
objects within a complex, dynamic visual scene and to
track selected objects with their eyes. Voluntary eye
movements in primates are therefore not just a motor
phenomenon but depend on the sophisticated sensory
Address correspondence to: Richard J. Krauzlis, Salk Institute for
Biological Studies, 10010 North Torrey Pines Road, La Jolla, CA
92037 (e-mail: rich@salk.edu).
and cognitive processing capabilities of the primate cen-
tral nervous system. The importance of these higher
order processes, and the complexity of the underlying
mechanisms, pose both challenges and opportunities for
using voluntary eye movements as a model for under-
standing the neural circuits involved in visuomotor con-
trol. This review highlights some recent findings that
provide new perspectives on the functional organization
of these voluntary motor systems.
The Neural Pathways for
Pursuit and Saccades
Although the pursuit and saccadic systems have tradi-
tionally been viewed as anatomically distinct, more
recent evidence indicates that there is considerable over-
lap in the neural pathways for pursuit and saccades. Both
systems involve a similar set of areas in the cerebral cor-
tex (Fig. 1). For saccades, these cortical areas evaluate
and update the locations of potential targets and provide
motor commands for saccades and include the lateral
intraparietal area (LIP), the frontal eye fields (FEFs),
and the supplementary eye fields (SEFs). For pursuit,
cortical areas are involved in processing the visual
motion and other control signals necessary for pursuit
and include the middle temporal (MT) and medial supe-
rior temporal (MST) areas and subregions of areas LIP,
FEF, and SEF. Thus, many of the same cortical areas are
involved in the control of both pursuit and saccades, but
each area contains separate subregions for the two types
of movements, and the corresponding subregions are
interconnected to form a closely matched pair of cortical

124 THE NEUROSCIENTIST Voluntary Eye Movements

Copyright © 2005 Sage Publications
ISSN 1073-8584

Fig. 1. Outline of the pathways for pursuit and saccadic eye
movements. Schematic diagram of the descending pathways
are depicted on a lateral view of the monkey brain. Shaded
regions indicate specific areas within the cerebral cortex, basal
ganglia, cerebellum, and brain stem, and arrows indicate the
anatomical connections between these areas. Regions demar-
cated with dashed lines indicate structures normally covered by
the cerebral cortex. For clarity, not all relevant areas are depict-
ed (e.g., ascending pathways are omitted), and arrows do not
always correspond to direct anatomical connections. CN = cau-
date nucleus (basal ganglia); FEF = frontal eye field; LIP = later-
al intraparietal area; MT = middle temporal area; MST = medial
superior temporal area; PMN = brain stem premotor nuclei
(PPRF, riMLF, cMRF); PON = precerebellar pontine nuclei; SC =
superior colliculus (intermediate and deep layers); SEF = sup-
plementary eye field; SNr = substantia nigra pars reticulate;
Verm = oculomotor vermis (cerebellum, lobules VI and VII); VN =
vestibular nuclei; VPF = ventral paraflocculus (cerebellum).
networks (Tian and Lynch 1996a, 1996b). Functional
imaging studies in humans also support the idea of par-
allel but distinct cortical pathways for pursuit and sac-
cades (Petit and Haxby 1999; Rosano and others 2002).
These multiple cortical areas influence eye motor con-
trol through several descending pathways. First, there are
direct projections to eye-movement-related structures in
the brain stem such as the superior colliculus (SC) and
premotor nuclei in the reticular formation (PMN). These
pathways, which have figured prominently in the control
of saccades, have been recently demonstrated to exist for
pursuit cortical areas as well (Yan and others 2001).
There are also several less direct routes. One pathway
passes through the pontine nuclei to eye movement
regions of the cerebellum (oculomotor vermis, ventral
paraflocculus [VPF]), which access the output motor
nuclei for eye movements by projections to the vestibu-
lar nucleus and other brain stem motor nuclei (PMN).
For pursuit, this cortico-ponto-cerebellar route has been
traditionally considered the primary control pathway,
whereas for saccades, it has been viewed primarily as a
regulatory side loop. There are also descending path-
ways involving nuclei of the basal ganglia, such as the
caudate nucleus and the substantia nigra pars reticulata,
which exert their influence on eye movements through
Volume 11, Number 2, 2005
the SC. As with the direct projections to the brain stem,
the pathways through the basal ganglia are well estab-
lished for saccades but have only recently been demon-
strated for pursuit (Cui and others 2003).
In summary, although pursuit and saccades have his-
torically been viewed as anatomically distinct systems,
new data argue that they have a similar functional archi-
tecture and involve many of the same brain regions,
including the brain stem, cerebellum, superior collicu-
lus, and the cerebral cortex. In this admittedly selective
review of the recent literature, we will start at the circuits
that form and regulate the motor commands and wind
our way up through the areas that evaluate and extract
the signals needed to trigger and guide the movements.
Brain Stem
It has been known for some decades that the motor com-
mands for saccades are constructed primarily by a circuit
in the brain stem that generates the burst of neural activ-
ity necessary to cause the rapid changes in muscle
force that propel saccades. The elements of this
circuit are spread across several nuclei in the pons and
mesencephalon—the paramedian pontine reticular for-
mation (PPRF), the rostral interstitial nucleus of the
medial longitudinal fasciculus (riMLF), and the nucleus
raphe interpositus (nRIP)—and contain several classes
of saccade-related neurons (Luschei and Fuchs 1972;
Keller 1974; Sparks and Sides 1974; Henn and Cohen
1976; Raybourn and Keller 1977; Van Gisbergen and
others 1981; Henn and others 1984). Short-lead burst
neurons emit a burst of spikes whose precise timing
determines the amplitude of the saccade. Long-lead
burst neurons exhibit a prelude of activity before emit-
ting a saccade-related burst. Pause neurons in the nRIP
discharge steadily but stop firing during some or all sac-
cades (omnipause neurons [OPNs]). Several models
(e.g., Scudder 1988) have suggested how these neurons
might participate in saccade generation. A trigger signal,
probably from the SC, causes OPNs to pause their firing
momentarily, which then disinhibits burst neurons. This
disinhibition evokes a burst whose duration corresponds
to the amplitude of the saccade; the burst duration is
controlled by a negative feedback circuit and is adap-
tively regulated in conjunction with the cerebellum.
New evidence indicates that parts of this brain stem
circuit for saccades are also involved in the control of
pursuit. The primary brain stem nuclei for controlling
horizontal and vertical gaze (the PPRF, riMLF, and
cMRF) all receive direct inputs from the pursuit subre-
gion of the FEF as well as from the saccade-related sub-
region (Yan and others 2001). Recent recording studies
have shown that subsets of the neurons in these nuclei
have pursuit-related as well as saccade-related activity.
For example, some burst neurons in the PPRF are active
only during saccades, but a second category of burst
neurons is active during both saccades and pursuit
(Missal and Keller 2001). Similarly, in the riMLF of the
cat, some burst neurons fire in relationship to eye veloc-
ity not only during saccades but also during pursuit
THE NEUROSCIENTIST 125
(Missal and Keller 2001). Perhaps most surprising are
the recent findings suggesting that OPNs play a role in
pursuit. About half of the OPNs show significant
decreases in activity during the onset of pursuit as well
as pauses for saccades; they do not completely stop fir-
ing as for saccades but reduce their activity by about
one-third (Missal and Keller 2002). Microstimulation in
the region of the OPNs has long been known to halt sac-
cades, but recent experiments show that such microstim-
ulation also strongly decelerates pursuit (Fig. 2),
although it does not completely stop pursuit (Missal and
Keller 2002).
These studies indicate that the construction of the
motor commands for pursuit and saccades involves
shared circuitry in the brain stem, and Figure 3 shows
one candidate scheme for how the motor circuits for the
two movements might be related. Analogous to the way
that OPNs are believed to gate the occurrence of sac-
cades through inhibitory effects on excitatory burst neu-
rons, OPNs could regulate the gain of pursuit through
their inhibitory effect on pursuit neurons in the nucleus
prepositus hypoglossi (NPH) and the medial vestibular
nuclei (MVN). Another novel class of pursuit-related
neurons, the burst neurons in the PPRF and riMLF,
might acquire their smooth-eye-velocity modulation
through excitatory inputs from the PNs. By inhibiting
the OPNs and completing a loop with the pursuit neu-
rons in the NPH/MVN, these neurons might act to latch
the pursuit system in an “on” state.
Many important details about this putative gating
mechanism remain unknown, but a circuit with these
features could account for several properties of pursuit
and saccades. If the gating of pursuit and saccades
involved shared circuitry in the brain stem, this would
provide a straightforward way to coordinate and regulate
the triggering of pursuit and saccades, consistent with
behavioral evidence that there is a shared inhibitory
mechanism for pursuit and saccades (Kornylo and others
2003). On the other hand, the difference in the level of
disinhibition associated with the two movements could
provide flexibility in determining what is required to
trigger the two types of movements, consistent with the
observations that pursuit generally has a shorter latency
than saccades and that pursuit and saccades usually but
not always agree in their choice of a target (Krauzlis and
others 1999; Liston and Krauzlis 2003). The graded inhi-
bition of the OPNs during pursuit would also be predict-
ed to produce a smoothly graded disinhibition of the
NPH/MVN neurons, consistent with the suggestion from
behavioral experiments that there is a variable gain con-
troller in the pathways for pursuit eye movements
(Grasse and Lisberger 1992; Krauzlis and Lisberger
1994a; Keating and Pierre 1996; Krauzlis and Miles
1996c).
These recent findings seemingly contradict clinical
observations that damage to the brain stem reticular for-
mation causes selective palsy for saccades (Hanson and
others 1986). However, the discrepancy is resolved when
one compares lesions of different sizes. Smaller brain
stem lesions in humans and monkeys can result in
126 THE NEUROSCIENTIST
Fig. 2. Microstimulation in the region of the omnipause neurons
(OPNs) decelerates pursuit eye velocity. Average eye velocity
on trials with microstimulation (thick solid line, n = 9) is com-
pared to average eye velocity on trials without microstimulation
(thin solid line, n = 8). During the period of microstimulation
(indicated by the orange bar), eye velocity is reduced com-
pared to the trials with no stimulation. Thinner lines and dashed
lines indicate 95% confidence intervals of the mean eye veloc-
ity. The vertical arrow indicates the onset of the 40 deg/s right-
ward target motion. Adapted from Missal and Keller (2002,
p 1889). Used with permission from the American Physiological
Society.
deficits of large saccades, with relative sparing of both
pursuit and small saccades (Henn and others 1984;
Hanson and others 1986), but larger lesions of the retic-
ular formation result in a conjugate gaze palsy that
affects both saccades and pursuit (Bogousslavsky and
Meienberg 1987). Thus, depending on the size of the
lesion, brain stem damage appears to limit the amplitude
range of the eye movements that can be generated, rather
than the type of eye movements.
Cerebellum
The cerebellar cortex and deep cerebellar nuclei play a
crucial role in supporting the accuracy and adaptation of
voluntary eye movements. Although several regions
have been implicated in the control of eye movements,
two areas are especially well understood: the VPF and
the midline oculomotor vermis. Output neurons in the
VPF project directly to oculomotor nuclei in the brain
stem, whereas in the vermis, output neurons exert their
effect via projections to the fastigial oculomotor region
(FOR), a deep cerebellar nucleus.
Damage to the cerebellum does not eliminate eye
movements but renders them highly variable and inaccu-
rate. Ablation of the VPF and adjacent flocculus causes
large and lasting deficits in smooth eye movements and
the ability to maintain fixation (Zee and others 1981;
Rambold and others 2002). These dramatic effects may
reflect the close association of the VPF with the circuit
in the brain stem that integrates eye position signals
(Cannon and Robinson 1987). As illustrated by the
examples in Figure 4, lesions of the vermis or FOR dis-
Voluntary Eye Movements

Fig. 3. Possible diagram of how oculomotor nuclei in the brain
stem contribute to both pursuit and saccades. Excitatory
synapses are shown with small white circles; inhibitory synaps-
es are shown with small black circles. Note that there are two
distinct types of inputs to the circuit: gating signals that are
shared by pursuit and saccades and separate drive signals
conveying location and motion information. The omnipause
neurons (OPNs) play a crucial role in this circuit by regulating
when the descending drive signals are allowed to access the
final motor pathways. EBN = excitatory burst neuron; NPH,
MVN = pursuit-related neurons in the nucleus prepositus
hypoglossi and the medial vestibular nuclei; OMN = ocular
motor neurons; trig = interneuron that inhibits OPNs, thereby
triggering a saccade and possibly pursuit; latch = interneuron
that putatively keeps OPNs inhibited during the saccade and
pursuit movements.
rupt the timing, accuracy, and dynamics of saccades and
also the ability to adapt saccades (Robinson and others
1993; Takagi and others 1998; Barash and others 1999).
After damage to the vermis and FOR, saccades also
exhibit a dysmetria that depends on eye position, sug-
gesting that the cerebellar signals normally act to coun-
terbalance the changing mechanical forces encountered
by the eye at different positions in the orbit.
The activity of neurons in the cerebellum provides
insight into how the motor commands for eye move-
ments are shaped into their final forms. During pursuit
and saccades, neurons in the FOR emit an early burst of
spikes for contraversive movements and a later burst of
spikes for ipsiversive movements (Ohtsuka and Noda
1991; Fuchs and others 1993, 1994; Helmchen and oth-
Volume 11, Number 2, 2005
Fig. 4. Disruption of the timing, accuracy, and adaptation of
saccades after lesions of the cerebellar oculomotor vermis.
Traces show horizontal eye position as a function of time
(aligned with respect to saccade onset) during saccades before
and after lesions of the oculomotor vermis. In this experiment,
saccades were adapted by presenting a 10-degree forward
step of the target, followed by a 3-degree backward step.
Prelesion, the animal showed a decrease in saccade amplitude
late in adaptation (black arrow) as compared to early in adap-
tation. Postlesion, there was a marked increase in the variabili-
ty of saccade amplitudes (blue arrow) and an increase in laten-
cy for corrective saccades (orange arrow). These effects
persisted through the late phases of adaptation. From Takagi
and others (1998, p 1925). Used with permission from the
American Physiological Society.
ers 1994). These bursts reflect a push-pull arrangement
in which the same neurons that provide an accelerative
command for movements in one direction also provide a
braking signal for movements in the other direction.
Similarly, neurons in the VPF exhibit overshoots in fir-
ing rate when pursuit eye velocity increases or decreas-
es. These transient overshoots in the VPF also operate in
a push-pull fashion and appear to reflect a calculated attempt
to compensate for the sluggish mechanics of the eye mus-
cles and orbital tissues (Krauzlis and Lisberger 1994b;
Krauzlis 2000). The timing and size of these bursts change
after adaptation of eye movements and for eye move-
ments made from different orbital positions (Kleine and
others 2003; Scudder and McGee 2003), consistent with
the idea that the cerebellar output acts to maintain the
accuracy of eye movements under a variety of condi-
tions. Although the discharge of individual cerebellar
neurons is variable, the population response can provide
a motor command that is very precise; changes in the
contributions of individual neurons could therefore pro-
vide a mechanism for adjusting the size and timing of
eye movements (Krauzlis 2000; Thier and others 2000).
Superior Colliculus
The SC has been traditionally described as a motor map
of saccade end points, but several lines of evidence argue
that the SC comprises a map of motor goals rather than
the specific movement required to achieve that goal.
THE NEUROSCIENTIST 127
First, the locus of activity in the SC does not uniquely
determine the amplitude of the eye movement that is
made. Neurons in the SC fire differently for saccades
made to moving targets compared to saccades made to
stationary targets, arguing that the SC neurons specify
the initial retinotopic location and that additional circuits
are responsible for getting the saccade to land accurate-
ly on the target (Keller and others 1996). SC neurons
also fire differently for saccades made to remembered
targets compared to saccades made directly to visual tar-
gets, again indicating that activity in the SC does not
determine the exact metrics of the saccade (Stanford and
Sparks 1994).
SC activity also does not determine whether saccades
will be accomplished with the eye alone or with a com-
bination of the eye and head. When the head is immobi-
lized, activity in the SC is associated with eye saccades
with a specific direction and amplitude (Robinson 1972;
Schiller and Stryker 1972). However, when the head is
free to move, SC neurons exhibit activity that is closely
related to the amplitude and direction of combined eye-
head movements rather than to either the eye or head
component alone (Freedman and Sparks 1997). In these
unrestrained conditions, SC stimulation produces coor-
dinated movements of both the eyes and head (Freedman
and others 1996). The amplitudes of these combined
movements are larger compared to those evoked with the
head fixed because the evoked eye movements are the
same whether or not the head is free to move.
Consequently, the standard depiction of the SC motor
map obtained with the head restrained is distorted
because it systematically underestimates the amplitudes
of encoded gaze movements.
The SC also plays a role in the control of pursuit eye
movements. Activation and inactivation of the rostral
SC, which represents the central visual field, modifies
the metrics of pursuit, demonstrating a causal link
between SC activity and pursuit (Basso and others
2000). Many neurons in the rostral SC modulate their
firing rates during pursuit eye movements as well as dur-
ing small saccades (Krauzlis and others 1997, 2000).
This activity is not simply a visual response because it
persists in the absence of a visual target (Krauzlis 2001).
This activity also does not convey motion signals for
pursuit because although SC neurons respond to motion
stimuli, they are not selective for the direction of motion
(Krauzlis 2004). On the other hand, the complicated pat-
tern of activity exhibited by these neurons during
pursuit—and also fixation—can be explained by consid-
ering the location of the tracked target within the neu-
ron’s retinotopically organized response field (Krauzlis
and others 1997, 2000). The distribution of activity
across the SC motor map therefore appears to provide a
real-time estimate of the retinal location of the eye motor
goal for pursuit and fixation, as well as for saccades.
Recent experiments in cats have underscored this idea
that SC activity represents the motor goal and does not
necessarily specify the saccade end point. These experi-
ments exploit the fact that cats tend to accomplish large
128 THE NEUROSCIENTIST
orienting movements with a series of smaller saccades in
rapid succession rather than with a single large saccade.
During these multistep movements, activity in the SC is
initially at the site corresponding to the retinal location
of the eccentric target and then progresses toward more
central sites in a single sweep, even though the move-
ment itself is achieved with multiple saccades. As a
result, the locus of activity in the SC does not match the
amplitudes of the individual saccades used to acquire the
target but instead indicates the remaining distance to the
target (Bergeron and others 2003). The complementary
pattern holds for neurons in the rostral SC, which repre-
sent the central visual field and tend to be active during
fixation (Munoz and others 1991; Munoz and Wurtz
1993). These rostral SC neurons remain inactive during
the multistep movement, even though the movement
pauses between each small saccade of the sequence,
resuming their tonic activity only as the sequence draws
to a close and the target is acquired (Bergeron and
Guitton 2002). Activity in the SC motor map therefore
does not appear to be exclusively involved with control-
ling saccade end points but serves a more general func-
tion associated with specifying the goal for orienting
movements.
One possibility that has gained support is the idea that
the SC plays a role in representing and selecting the tar-
gets for orienting movements. For example, decreasing
the probability that a visual stimulus will be the target,
by adding a variable number of irrelevant stimuli to the
display, decreases the visually-evoked and tonic activity
of many SC neurons (Basso and Wurtz 1997, 1998).
These changes are correlated with the latencies of the
saccades that follow but are not related to the amplitude
or peak velocity of the saccade. Similar effects are found
with a single visual stimulus by varying the probability,
between blocks of trials, that the target will appear in the
neuron’s response field (Dorris and Munoz 1998). These
effects of stimulus probability are especially evident
either before or soon after the visual stimuli are present-
ed, indicating that prior information may be especially
influential during the period of uncertainty that prevails
before unambiguous stimulus information is available to
guide the eye movement choice.
During the latent period after the candidate targets
have been presented but before the movement is initiat-
ed, SC neurons display a preference for the stimulus that
will become the eye movement target. In a color-oddity
search task using saccades, some SC neurons discrimi-
nate the target from the distractor with a delay that is
time locked to stimulus onset, rather than saccade laten-
cy, suggesting that they play a role in target selection in
addition to saccade preparation (McPeek and Keller
2002). In contrast, other neurons discriminate the target
with timing that is well correlated with saccade latency,
suggesting that they are more directly involved with trig-
gering saccades (McPeek and Keller 2002). In a match-
to-sample task using pursuit and saccades, many SC
neurons again exhibit selectivity for target stimuli, and
this selectivity can predict the timing of pursuit as well
Voluntary Eye Movements
as saccade choices (Krauzlis and Dill 2002). The signal
indicating the correct choice emerges over time, forming
a trade-off between speed and accuracy. The observed
pursuit and saccade performances fall on different parts
of the speed-accuracy curve predicted by neuronal activ-
ity, supporting the idea that pursuit and saccades are
guided by shared selection signals but involve different
trade-offs between speed and accuracy (Krauzlis and
others 1999; Liston and Krauzlis 2003).
Manipulation of the fixated visual stimulus can also
modify target-related activity in the SC. Many neurons
in the SC increase their firing rate after the fixation stim-
ulus is extinguished, even if a visual target has not yet
appeared in their response field, and these changes are
correlated with the latencies of both pursuit (Krauzlis
2003) and saccades (Dorris and others 1997; Sparks and
others 2000; Krauzlis 2003). Conversely, neurons in the
rostral SC that are typically active during fixation
decrease their firing after the offset of the fixation spot
(Dorris and Munoz 1995; Dorris and others 1997).
These changes in activity indicate a shift in the distribu-
tion of activity across the SC in favor of those neurons
that are likely to represent the impending target. By
changing the baseline activity, the subsequent volley of
activity evoked by the appearance of the target can more
readily trigger an eye movement, providing a neural cor-
relate for the shared effects on pursuit and saccade laten-
cies observed in this paradigm (Krauzlis and Miles
1996a, 1996b; Krauzlis 2003). From these results, it has
been suggested that the same signals in the rostral SC
that are involved in the covert preparation of saccades
might also control the gating of inputs for pursuit
(Krauzlis 2003). This type of shared control could
explain the linkage that has been observed in the selec-
tion of targets for pursuit and saccades (Gardner and
Lisberger 2001, 2002). Although the mechanism has not
yet been identified, one possibility is that this shared
control is exerted by a projection from the SC to brain
stem OPNs, the gatekeepers for saccades that have also
been recently implicated in the inhibitory control of pur-
suit (Missal and Keller 2002).
The idea that the SC is involved in target selection has
now been directly tested in a pair of studies (Fig. 5). One
study used a visual search task in which the target was
defined as the “oddball” element in an array of visual
stimuli (McPeek and Keller 2004). When the region of
the SC representing the target was focally inactivated,
saccades were often misdirected to distractors appearing
in unaffected areas of the visual field (Fig. 5A).
Importantly, the amplitude of this deficit was larger
when the task of identifying the target was harder, argu-
ing for an effect at the stage of target selection beyond
any effect on saccade motor execution. The other study
used a luminance discrimination task and showed that
weak activation of the SC (i.e., microstimulation that is
subthreshold for evoking saccades) biased the selection
of targets toward the stimulated location not just for sac-
cades but for pursuit as well (Carello and Krauzlis
2004). Using the classic “step-ramp” paradigm, the
stimuli for pursuit appeared in one hemifield before
Volume 11, Number 2, 2005
Fig. 5. Activation and inactivation of the superior colliculus (SC)
affects target selection. A, Effects of SC inactivation on sac-
cades during a visual search task. Under normal conditions,
monkeys were able to identify the target based on its unique
color and make saccades directly to it (left). The pattern of sac-
cades changed after a local area of the SC was inactivated by
injection of muscimol, corresponding to the portion of the visu-
al field in which the target was located (blue ellipse). After local
inactivation, the monkeys made many inappropriate saccades
to the distractor stimuli (right). Reproduced with permission
from McPeek and Keller (2004, p 758). B, Effects of SC activa-
tion on pursuit during a discrimination task. Under normal con-
ditions, monkeys were able to correctly identify the target
based on its luminance and generate a pursuit movement (hor-
izontal eye speed) to follow it (right). Performance was changed
after a local area of the SC was activated with microstimulation,
corresponding to the portion of the visual field in which the dis-
tractor was located (orange ellipse). With local activation, the
monkeys generated many more inappropriate smooth eye
movements to follow the distractor (left). Note that in the case
of this pursuit experiment, the affected site in the SC corre-
sponds to the location of the selected stimulus (the distractor is
on the right), even though this requires an eye movement in the
opposite direction (the eye moves smoothly to the left). From
Carello and Krauzlis (2004, p 577). Copyright 2004 by Cell
Press.
moving toward and into the opposite hemifield
(Rashbass 1961), making it possible to distinguish
between the initial location of the target and the direction
of the eye movement, a distinction that is not possible
with saccades. Critically, the effect of SC activation was
based on the target location, not the eye movement direc-
tion. For example, as illustrated in Figure 5B, when the
stimulated region of the SC matched the distractor loca-
tion (right), pursuit was more likely to follow the dis-
tractor, even though this required an eye movement in
the opposite direction (leftward). These results argue that
the SC plays a role in target choice per se, distinct from
its traditional role in motor preparation.
One important issue left unresolved by these studies is
whether the SC participates in target selection by biasing
the selection of the response goal or by shifting the allo-
THE NEUROSCIENTIST 129
cation of visual attention. The visual responses of SC
neurons show enhancement consistent with an effect of
attention (Goldberg and Wurtz 1972; Kustov and
Robinson 1996), and many SC neurons are active during
covert shifts of attention evoked by spatially precise cues
but not by nonspatial symbolic cues (Ignashchenkova
and others 2004). These findings support the idea that
there is a common network for controlling attention and
saccades, consistent with the premotor theory of atten-
tion (Rizzolatti and others 1987; Sheliga and others
1995). Together with the activation and inactivation
results, these studies raise the intriguing possibility that
the SC not only receives the selection signal and applies
it toward implementing the motor choice but also helps
regulate the sensory-motor processing that leads to that
selection.
Another unresolved issue is how target-related activi-
ty in the SC is read out to trigger the appropriate eye
movement choice (Krauzlis and others 2004). One help-
ful approach to this problem introduces the assumption
that firing rates are proportional to the likelihood that
the target is present, and the decision is affirmed when
activity reaches a particular significance level
(Carpenter and Williams 1995; Gold and Shadlen 2001).
As shown schematically in Figure 6, the relevant deci-
sion signal for target selection might be based not only
on the firing rate associated with possible new targets
(represented by activity at caudal sites in the SC) but
also on the firing rate associated with the currently Fig. 6. Hypothetical decision mechanism explaining how activ-
ity in the superior colliculus (SC) might be read out to accom-
foveated stimulus (represented by activity at the rostral plish target selection. In the top panel, the monkey is initially
SC). If these firing rates are proportional to target likeli- fixating the central stimulus (blue square) and is considering
hood, then this comparison between caudal and rostral whether to make an eye movement to a possible new target in
sites in the SC could amount to a likelihood ratio test the periphery (orange square). The middle panel shows
(Gold and Shadlen 2001; Krauzlis and others 2004). In schematically how activity corresponding to the two stimuli is
distributed across the SC, with activity related to the fixated
general, likelihood ratio tests are useful for testing stimulus at the rostral end and activity for the new stimulus at
whether a more complex model (in this case, that the tar- a more caudal location. Available data suggest, but have not
get is at an eccentric location) provides a better descrip- yet proved, that the firing rate (FR) of SC neurons is propor-
tion of the data than the simpler model (that the target is tional to the likelihood that the target is in the response field of
the SC neuron (Krauzlis and others 2004). If so, then compari-
already foveated) because it gives values that are related son of activity across the SC amounts to a comparison of alter-
to common test statistics such as the F test and the χ2. In native hypotheses, and the difference in activity would indicate
this case, every decision by the SC to select an eccentric the relative likelihood of one hypothesis over the other. The
target would amount to a rejection of the null hypothe- decision of whether to select the new target (“GO”) or remain
sis. This type of decision framework also has the advan- fixating (“STAY”) could then be determined by comparing the
difference in firing rate between caudal and rostral neurons (the
tage of being very flexible because the source of the putative decision signal) to a threshold value.
information does not really matter; what matters is how
the information improves the estimate of target likeli-
hood. For example, target selection should be based not
just on visual evidence but also on information about
prior probability and expected rewards (Platt and
Glimcher 1999; Ikeda and Hikosaka 2003). If these dif- mild effects (Albano and others 1982). However, com-
ferent processes gave their answers in the same units bined damage to the SC and areas of the cerebral cortex
(e.g., something proportional to likelihood), it would be can eliminate voluntary saccades (Schiller and others
possible to combine and exchange these different 1980), indicating the importance of the cerebral cortex
sources of information on an equal footing and then read in providing signals that trigger and guide voluntary eye
the answer out from the SC in meaningful way. movements.

Cerebral Cortex Frontal Eye Fields

The SC plays a pivotal role in the control of voluntary The functional importance of the FEF is especially evi-
eye movements, but ablation of the SC has surprisingly dent when the outcome involves some degree of choice

130 THE NEUROSCIENTIST Voluntary Eye Movements

or self-control. For saccades, lesions of the FEF produce
only mild and temporary deficits in saccades when per-
formance is tested with solitary visual targets (Dias and
others 1995; Sommer and Tehovnik 1997; Dias and
Segraves 1999). However, the deficits after FEF lesions
are much more severe when the target stimulus is
accompanied by other irrelevant distracter stimuli
(Schiller and Chou 1998, 2000) or when the saccade is
directed to a remembered location (Dias and others
1995; Sommer and Tehovnik 1997; Dias and Segraves
1999). Similarly, disruption of FEF activity in humans
using magnetic stimulation disrupts performance in
visual search tasks (Muggleton and others 2003). For
pursuit, the effects are somewhat more dramatic (Fig. 7).
Inactivation of the smooth eye movement subfield of the
FEF (FEFsem) scales down the pursuit motor of visual
targets to about 25% of its normal value (Shi and others
1998), and lesions of the FEFsem eliminate the predic-
tive component of pursuit eye movements (Keating
1991; MacAvoy and others 1991).
Neurons in the FEF exhibit properties consistent with
determining when voluntary eye movements are initiat-
ed. For pursuit, neurons in the FEFsem exhibit direc-
tionally selective responses appropriate for guiding pur-
suit, and, in addition, many of them discriminate the
direction of motion before the onset of pursuit (Tanaka
and Lisberger 2002b). For saccades, the trial-to-trial
variability in reaction times is related to the variability in
when the firing rates of FEF neurons reach a relatively
constant threshold value (Hanes and Schall 1996); when
an impending saccade is canceled, the firing rates drop
(Hanes and others 1998), suggesting that FEF activity
can regulate when and if a saccade will be triggered. As
in the SC (McPeek and Keller 2002), the FEF appears to
contain at least two classes of saccade-related neurons:
One type is time locked to the stimulus and therefore
appears to be associated with the process of target selec-
tion, whereas a second type is time locked to the move-
ment onset and therefore appears to be involved with
triggering the movement (Sato and Schall 2003). Indeed,
some FEF neurons discriminate visual targets even in the
absence of saccades or saccades directed elsewhere, sug-
gesting that their activity corresponds to the allocation
of attention rather than the motor preparation of sac-
cades (Thompson and others 1997; Murthy and others
2001).
The interplay between eye motor planning and visual
functions such as selection and attention has been high-
lighted in several experiments. Stimulation of the
FEFsem evokes smooth eye movements and is the only
cortical region in which pursuit can be evoked when
the eyes are fixating, but in addition to introducing a
direction-specific signal into the velocity command for
pursuit, stimulation also changes the gain of the pursuit
response to new visual motion inputs (Tanaka and
Lisberger 2001, 2002a). If saccades are evoked by FEF
stimulation as monkeys perform a motion discrimination
task, the movement end points are shifted toward the
direction corresponding to the nascent perceptual judg-
ment (Gold and Shadlen 2000). Importantly, the ampli-
Volume 11, Number 2, 2005
Fig. 7. Deficits in pursuit eye velocity after inactivation of the
smooth eye movement portion of the frontal eye fields
(FEFsem) by injection of muscimol. Top, Single trial of step-
ramp tracking just before injection of muscimol into the right
FEFsem (dark solid line) superimposed on the first trial after
injection (dashed line). At time 0 ms, the target stepped 4
degrees to the left and moved at 40 deg/s to the right. In the
preinjection trial, the eye trajectory approximately matched that
of the target, whereas in the postinjection trial, the tracking was
accomplished mostly by the saccades. Bottom, Eye velocity
profiles superimposed for several trials preinjection (solid lines)
and postinjection (dashed lines). For all six preinjection trials,
the peak velocity reached or exceeded that of the target,
whereas for all six postinjection trials, the peak velocity was far
below that of the target. Rapid upward and downward deflec-
tions of the velocity traces correspond to saccades. From Shi
and others (1998, p 460). Used with permission from the
American Physiological Society.
tude of the shift depends on the strength of the visual
signal; this result argues that the perceptual evaluation of
the stimulus and the motor preparation of the saccade are
not serial stages of processing but instead occur togeth-
er and perhaps involve a common level of neural organ-
ization. Conversely, the allocation of attention itself
appears to be altered by stimulation of the FEF.
Stimulation within the FEF with currents too weak to
evoke saccades can nonetheless enhance visual respons-
es in extrastriate area V4 (Moore and Armstrong 2003)
and improve performance on a visual discrimination task
(Moore and Fallah 2004).
Lateral Intraparietal Area
The LIP also plays a key role in the process of visual
selection. Inactivation of the LIP does not produce
deficits in the latency or accuracy of saccades to single
targets but dramatically reduces the frequency of sac-
cades to the affected visual field when competing stim-
uli are present (Fig. 8) and increases the time required to
find the target during visual search (Wardak and others
2002). The emergence of these deficits when there are
multiple choices indicates a competitive interaction
between the candidate targets and indicates how animal
models may be useful for addressing the visual neglect
THE NEUROSCIENTIST 131
and extinction syndromes that occur in humans (Payne
and Rushmore 2003).
The activity of LIP neurons is strongly affected by
information relevant for visual selection. For example,
as in the SC and FEF, neurons in LIP respond more
strongly when the stimulus in their response field is a
target or behaviorally relevant than when it is a distrac-
tor or irrelevant (Platt and Glimcher 1997; Gottlieb and
others 1998). When monkeys are asked to discriminate
the direction of motion in a random-dot visual display
and subsequently report their answer with a saccade, LIP
activity changes during the viewing of motion in a way
that predicts the monkey’s upcoming perceptual decision
(Shadlen and Newsome 2001). Like the FEF, changes in
LIP activity related to attention and selection can be dis-
tinguished from motor preparation. For example, LIP
activity is lower for a visual cue prompting a saccade
than for a visual cue indicating that a saccade should not
be made; this difference does not match the change in
motor plans but is compatible with the idea that such
changes garner increased attention (Bisley and Goldberg
2003).
The activity of LIP neurons is also modulated by
reward. When the size of reward is varied across blocks
of trials, LIP neurons are more active when the expected
reward is higher (Platt and Glimcher 1999).
Interestingly, using a different experimental design, neu-
rons in the FEF did not show a reward-related modula-
tion (Leon and Shadlen 1999), raising the possibility that
the presence or absence of reward-related information is
a point of distinction between the two cortical areas.
The parietal cortex also plays some role in pursuit, but
this has been less studied. Stimulation of the LIP can
evoke smooth eye movements as well as saccades
(Kurylo and Skavenski 1991), and about half of the neu-
rons in the LIP and the ventral intraparietal area exhibit
direction-specific activity during pursuit (Bremmer and
others 1997; Schlack and others 2003). The pursuit-
related activity of many LIP neurons is also modulated
by eye position and other extraretinal signals (Bremmer
and others 1997; Schlack and others 2003), consistent
with the idea that the parietal cortex represents the goals
for movements in coordinate frames appropriate for
effector organs such as the eyes, head, and hands
(Andersen and others 1997; Calton and others 2002).
Supplementary Eye Field
The SEF plays a less direct role in the control of sac-
cades and pursuit than the FEF does, but it appears to be
especially important for movements that are guided by
internal factors, rather than driven by external events.
During a saccade task in which monkeys are free to
choose either of two identical stimuli to receive their
rewards, neurons in the SEF, FEF, and LIP exhibit activ-
ity that anticipates the upcoming choice, but this activi-
ty is largest and occurs earliest in the SEF (Coe and oth-
ers 2002). Neurons in the SEF are also strongly modu-
lated during tasks in which the goal is defined by
abstract instructions, such as saccades directed to a part
132 THE NEUROSCIENTIST
Fig. 8. Inactivation of the lateral intraparietal area (LIP) disrupts
saccades to the affected visual field during a search task.
Single-trial examples of visual search patterns after injection of
muscimol into the right LIP. The small dots show eye position
sampled every 4 ms, large dots represent the search stimuli,
and the open circle represents the location of the target. When
the target was in the ipsilateral visual field, unaffected by the
lesion (right), the monkey typically found the target within a
small number of saccades. When the target was in the con-
tralateral visual field, matching the site of the lesion (left), the
number of saccades and overall search time dramatically
increased. Adapted from Wardak and others (2002, p 9882).
Copyright 2002 by the Society for Neuroscience.
of an object rather than a spatial location (Olson and
Gettner 1995; Tremblay and others 2002), saccades
directed to the location opposite the visual stimulus
(“antisaccades”; Schlag-Rey and others 1997), and sac-
cades that occur within learned combinations or
sequences of saccades. During pursuit, SEF neurons
exhibit the largest changes in activity when the target
motion changes, especially when the timing of those
changes is predictable (Heinen and Liu 1997).
Accordingly, as shown in Figure 9, stimulation of the
SEF can facilitate smooth pursuit eye movements, and
this effect is largest if the stimulation is applied just as a
period of fixation is predictably drawing to a close and
the signal to initiate pursuit is about to be given (Missal
and Heinen 2001, 2004).
MT and MST Areas
The MT and MST areas are the major sources of visual
motion information that is critical for guiding pursuit
and for adjusting the amplitudes of saccades to moving
targets (Newsome and others 1985; Dürsteler and Wurtz
1988). Recent studies have clarified how visual process-
ing in these areas changes over time and is related to
processes such as attention and perception.
Most of the directional information that can be
extracted from MT neurons is conveyed within the first
100 milliseconds of the neuronal response (Osborne and
others 2004). However, the precision of the directional
information conveyed by MT neurons is relatively poor,
indicating that responses are probably pooled across the
population to match the direction discrimination of pur-
suit. One possibility is that the pursuit system relies on
Voluntary Eye Movements

Fig. 9. Activation of the supplementary eye fields (SEFs) can
facilitate anticipatory pursuit eye movements. On each trial,
after a 500-ms fixation period (horizontal dashed lines), the tar-
get was extinguished for 200 ms (gap in the dashed lines).
When the target reappeared, it stepped to an eccentric position
and moved at a constant speed in the opposite direction. Top,
Positions of the eye and target as a function of time from single
trials with and without stimulation. Bottom, The effect of
microstimulation was more evident in the traces of horizontal
eye velocity and occurred even before the target was visible
(orange arrow). The orange bar at the bottom indicates the peri-
od of stimulation. From Missal and Heinen (2004, p 1258). Used
with permission from the American Physiological Society.
the center of mass of the population of MT neurons, and
recent studies have illustrated that an estimate of target
speed can be obtained by taking a weighted average of
the responses across the population of neurons
(Churchland and Lisberger 2001; Priebe and Lisberger
2004).
When the onset of a target stimulus is accompanied by
a distractor, the initial activity of MT and MST neurons
exhibits very little selectivity for the target, and accord-
ingly, the initial pursuit eye velocity mainly follows the
average of the two motion signals (Ferrera and Lisberger
1997; Recanzone and Wurtz 2000). The subsequent
activity of MT and MST neurons exhibits greater selec-
tivity, and the eye movements elicited at these longer
latencies selectively follow one or the other stimulus,
reflecting a winner-take-all mechanism (Recanzone and
Wurtz 2000). However, the changes in activity are rela-
tively small and occur in only a minority of neurons, so
it is not clear that these changes alone are sufficient to
account for the selectivity of pursuit.
Solving the problem of computing motion signals for
tracking appears to take some time. When multiple mov-
Volume 11, Number 2, 2005
ing stimuli are presented that can be perceptually
grouped as a single moving object, some MT neurons
initially respond to the local motion of the stimulus com-
ponents, but over the course of a few hundred millisec-
onds, they begin to respond to the global motion of the
object as a whole. The changes in the directional tuning
of the neural activity following a time course are similar
to the changes in the direction of pursuit eye velocity
(Pack and Born 2001). In behavioral experiments, sub-
jects can readily perceive and track the veridical motion
of partially occluded objects, despite the ambiguous and
often misleading local motions of the component edges
(Stone and others 2000). The perceived and pursued
directions are initially more closely related to the aver-
age direction of the local edge motions, but they con-
verge to the veridical object motion direction after ~100
milliseconds (Masson and Stone 2002). These findings
also indicate that, over time, pursuit is guided by a sig-
nal related to the perceived motion of the object, rather
than the physical motion of the stimulus on the retina.
This idea is supported by recent studies showing that the
motion signals conveyed by some neurons in MST do
not depend on retinal inputs (Ilg and Thier 2003) and
that they encode target motion in world-centered, rather
than retina-centered, coordinates (Ilg and others 2004).
Conclusion and Outlook
Recent studies at a variety of levels have shown that the
functional organization of the pursuit and saccadic eye
movement systems are much more similar than previ-
ously recognized. Rather than composing two distinct
systems that operate as visuomotor reflexes, pursuit and
saccades are mediated by similar and sometimes over-
lapping pathways and are guided by a variety of higher
order processes as well as by more direct sensory inputs
(Fig. 10). The picture that emerges from these studies is
quite different from that found in most textbooks, and
each point of departure from the traditional view raises
its own set of questions and challenges.
The overlap in the brain stem pathways argues that the
gating of pursuit and saccades involves shared circuitry
that has been previously viewed as strictly part of the
saccadic system. Working out the brain stem wiring for
saccades alone has been difficult and is still not com-
pletely resolved (Scudder and others 2002); it is unclear
whether extending it to pursuit will make it easier or
harder to understand the functional states and transitions
accomplished by this circuit.
Consistent with its role in other motor systems, ocu-
lomotor regions of the cerebellum (VPF, vermis) appear
to expertly tweak the commands for pursuit and saccades
to compensate for mechanical constraints and to adapt
the movements to changing circumstances. In addition to
understanding how the cerebellar circuits accomplish
this function, there is also the conundrum that most of
the descending signals that would appear relevant for the
visual control of pursuit and saccades go to the dorsal
paraflocculus (Glickstein and others 1994) and not to the
VPF and vermis.
THE NEUROSCIENTIST 133

Fig. 10. A hypothetical model of the functional organization of
voluntary eye movements. Rather than linking signals obtained
from early visual processing steps directly to the motor outputs
for pursuit and saccades (inset), the control of voluntary eye
movements involves a cascade of steps with several check-
points that provide flexibility in how the movements are guided,
selected, and implemented. Evaluation of the sensory inputs
can be rapid but can also be influenced by higher order
processes such as perception, memory, and expected rewards
(sensory evaluation). The results from this evaluation proceed
along two tracks to influence the motor outputs. One track is
responsible for selecting the target and gating the motor
response (gating, target selection) and involves structures
including the SC. The other track is responsible for providing
the drive signals that determine the metrics of the movements
(drive) and involves structures such as the cerebellum. In this
model, the choice of whether to generate a pursuit movement
or a saccade movement, or some combination of the two, is
not solely determined by the descending signals as in the tra-
ditional view (inset) but instead depends on a comparison
between the descending signals and the current motor state
(motor execution).
In contrast to earlier descriptions of the pursuit and
saccade systems, the SC appears to be part of a shared
mechanism for selecting targets and perhaps triggering
the two types of voluntary eye movements. It remains to
be clarified how this target-related activity is read out to
trigger the appropriate motor commands, especially
because the SC does not mediate visual motion signals
for pursuit (Krauzlis 2004). It is also unclear whether the
SC simply applies a selection signal that is established
elsewhere—for example, the cerebral cortex—or
whether it also plays a crucial role in the selection
process itself, perhaps by regulating shifts of visual
attention.
Many of the signals that guide pursuit and saccades
come from a network of cortical areas and involve a vari-
ety of processes important for the selectivity and guid-
ance of voluntary eye movements: motor preparation,
attention, perception, and expected reward. Under most
134 THE NEUROSCIENTIST
circumstances, these processes tend to give the same
answer—attention and motor preparation are typically
directed toward the most rewarding target—making it
difficult to tease them apart or to localize functions to
particular areas or classes of neurons. The broader chal-
lenge is to move beyond identifying neural correlates of
processes we expect to find and instead to begin enu-
merating the unique factors that operate in each region
and to explain how these factors interact across the net-
work of neurons and brain regions.
References
Albano JE, Mishkin M, Westbrook LE, Wurtz RH. 1982. Visuomotor
deficits following ablation of monkey superior colliculus. J
Neurophysiol 48:338–51.
Andersen RA, Snyder LH, Bradley DC, Xing J. 1997. Multimodal rep-
resentation of space in the posterior parietal cortex and its use in
planning movements. Annu Rev Neurosci 20:303–30.
Barash S, Melikyan A, Sivakov A, Zhang M, Glickstein M, Thier P.
1999. Saccadic dysmetria and adaptation after lesions of the cere-
bellar cortex. J Neurosci 19:10931–9.
Basso MA, Krauzlis RJ, Wurtz RH. 2000. Activation and inactivation
of rostral superior colliculus neurons during smooth-pursuit eye
movements in monkeys. J Neurophysiol 84:892–908.
Basso M, Wurtz R. 1997. Modulation of neuronal activity by target
uncertainty. Nature 389:66–9.
Basso MA, Wurtz RH. 1998. Modulation of neuronal activity in supe-
rior colliculus by changes in target probability. J Neurosci
18:7519–34.
Bergeron A, Guitton D. 2002. In multiple-step gaze shifts: omnipause
(OPNs) and collicular fixation neurons encode gaze position error;
OPNs gate saccades. J Neurophysiol 88:1726–42.
Bergeron A, Matsuo S, Guitton D. 2003. Superior colliculus encodes
distance to target, not saccade amplitude, in multi-step gaze shifts.
Nat Neurosci 6:404–13.
Bisley JW, Goldberg ME. 2003. Neuronal activity in the lateral intra-
parietal area and spatial attention. Science 299:81–6.
Bogousslavsky J, Meienberg O. 1987. Eye-movement disorders in
brain-stem and cerebellar stroke. Arch Neurol 44:141–8.
Bremmer F, Distler C, Hoffmann KP. 1997. Eye position effects in
monkey cortex: II. Pursuit- and fixation- related activity in poste-
rior parietal areas LIP and 7A. J Neurophysiol 77:962–77.
Calton JL, Dickinson AR, Snyder LH. 2002. Non-spatial, motor-
specific activation in posterior parietal cortex. Nat Neurosci
5:580–8.
Cannon SC, Robinson DA. 1987. Loss of the neural integrator of the
oculomotor system from brain stem lesions in monkey. J
Neurophysiol 57:1383–409.
Carello CD, Krauzlis RJ. 2004. Manipulating intent: evidence for a
causal role of the superior colliculus in target selection. Neuron
43:575–83.
Carpenter RH, Williams ML. 1995. Neural computation of log likeli-
hood in control of saccadic eye movements. Nature 377:59–62.
Churchland MM, Lisberger SG. 2001. Shifts in the population
response in the middle temporal visual area parallel perceptual and
motor illusions produced by apparent motion. J Neurosci
21:9387–402.
Coe B, Tomihara K, Matsuzawa M, Hikosaka O. 2002. Visual and
anticipatory bias in three cortical eye fields of the monkey during
an adaptive decision-making task. J Neurosci 22:5081–90.
Cui DM, Yan YJ, Lynch JC. 2003. Pursuit subregion of the frontal eye
field projects to the caudate nucleus in monkeys. J Neurophysiol
89:2678–84.
Dias EC, Kiesau M, Segraves MA. 1995. Acute activation and inacti-
vation of macaque frontal eye field with GABA-related drugs. J
Neurophysiol 74:2744–8.
Dias EC, Segraves MA. 1999. Muscimol-induced inactivation of mon-
key frontal eye field: effects on visually and memory-guided sac-
cades. J Neurophysiol 81:2191–214.
Voluntary Eye Movements
Dorris MC, Munoz DP. 1995. A neural correlate for the gap effect on
saccadic reaction times in monkey. J Neurophysiol 73:2558–62.
Dorris MC, Munoz DP. 1998. Saccadic probability influences motor
preparation signals and time to saccadic initiation. J Neurosci
18:7015–26.
Dorris MC, Pare M, Munoz DP. 1997. Neuronal activity in monkey
superior colliculus related to the initiation of saccadic eye move-
ments. J Neurosci 17:8566–79.
Dürsteler MR, Wurtz RH. 1988. Pursuit and optokinetic deficits fol-
lowing chemical lesions of cortical areas MT and MST. J
Neurophysiol 60:940–65.
Ferrera VP, Lisberger SG. 1997. Neuronal responses in visual areas
MT and MST during smooth pursuit target selection. J
Neurophysiol 78:1433–46.
Freedman EG, Sparks DL. 1997. Eye-head coordination during head-
unrestrained gaze shifts in rhesus monkeys. J Neurophysiol
77:2328–48.
Freedman EG, Stanford TR, Sparks DL. 1996. Combined eye-head
gaze shifts produced by electrical stimulation of the superior col-
liculus in Rhesus monkeys. J Neurophysiol 76:927–52.
Fuchs AF, Robinson FR, Straube A. 1993. Role of the caudal fastigial
nucleus in saccade generation I: neuronal discharge patterns. J
Neurophysiol 70:1723–40.
Fuchs AF, Robinson FR, Straube A. 1994. Participation of the caudal
fastigial nucleus in smooth-pursuit eye movements: I. Neuronal
activity. J Neurophysiol 72:2714–28.
Gardner JL, Lisberger SG. 2001. Linked target selection for saccadic
and smooth pursuit eye movements. J Neurosci 21:2075–84.
Gardner JL, Lisberger SG. 2002. Serial linkage of target selection for
orienting and tracking eye movements. Nat Neurosci 29:29.
Glickstein M, Gerrits N, Kralj-Hans I, Mercier B, Stein J, Voogd J.
1994. Visual pontocerebellar projections in the macaque. J Comp
Neurol 349:51–72.
Gold JI, Shadlen MN. 2000. Representation of a perceptual decision in
developing oculomotor commands. Nature 404:390–4.
Gold JI, Shadlen MN. 2001. Neural computations that underlie deci-
sions about sensory stimuli. Trends Cogn Sci 5:10–6.
Goldberg ME, Wurtz RH. 1972. Activity of superior colliculus in
behaving monkeys: II. Effect of attention on neuronal responses. J
Neurophysiol 35:560–74.
Gottlieb JP, Kusunoki M, Goldberg ME. 1998. The representation of
visual salience in monkey parietal cortex. Nature 391:481–4.
Grasse KL, Lisberger SG. 1992. Analysis of a naturally occurring
asymmetry in vertical smooth pursuit eye movements in a monkey.
J Neurophysiol 67:164–79.
Hanes DP, Patterson WF II, Schall JD. 1998. Role of frontal eye fields
in countermanding saccades: visual, movement, and fixation activ-
ity. J Neurophysiol 79:817–34.
Hanes DP, Schall JD. 1996. Neural control of voluntary movement ini-
tiation. Science 274:427–430.
Hanson MR, Hamid MA, Tomsak RL, Chou SS, Leigh RJ. 1986.
Selective saccadic palsy caused by pontine lesions: clinical, physi-
ological, and pathological correlations. Ann Neurol 20:209–17.
Heinen SJ, Liu M. 1997. Single-neuron activity in the dorsomedial
frontal cortex during smooth-pursuit eye movements to predictable
target motion. Vis Neurosci 14:853–65.
Helmchen C, Straube A, Buttner U. 1994. Saccade-related activity in
the fastigial oculomotor region of the macaque monkey during
spontaneous eye movements in light and darkness. Exp Brain Res
98:474–82.
Henn V, Cohen B. 1976. Coding of information about rapid eye move-
ments in the pontine reticular formation of alert monkeys. Brain
Res 108:307–25.
Henn V, Lang W, Hepp K, Resine H. 1984. Experimental gaze palsies
in monkeys and their relation to human pathology. Brain
107:619–36.
Ignashchenkova A, Dicke PW, Haarmeier T, Thier P. 2004. Neuron-
specific contribution of the superior colliculus to overt and covert
shifts of attention. Nat Neurosci 7:56–64.
Ikeda T, Hikosaka O. 2003. Reward-dependent gain and bias of visual
responses in primate superior colliculus. Neuron 39:693–700.
Volume 11, Number 2, 2005
Ilg UJ, Schumann S, Thier P. 2004. Posterior parietal cortex neurons
encode target motion in world-centered coordinates. Neuron
43:145–51.
Ilg UJ, Thier P. 2003. Visual tracking neurons in primate area MST are
activated by smooth pursuit eye movements of an “imaginary” tar-
get. J Neurophysiol 90:1489–502.
Keating EG. 1991. Frontal eye field lesions impair predictive and
visually-guided pursuit eye movements. Exp Brain Res 86:311–23.
Keating EG, Pierre A. 1996. Architecture of a gain controller in the
pursuit system. Behav Brain Res 81:173–81.
Keller EL. 1974. Participation of medial pontine reticular formation in
eye movement generation in monkey. J Neurophysiol 37:316–32.
Keller EL, Gandhi NJ, Weir PT. 1996. Discharge of superior collicular
neurons during saccades made to moving targets. J Neurophysiol
76:3573–7.
Kleine JF, Guan Y, Buttner U. 2003. Saccade-related neurons in the pri-
mate fastigial nucleus: what do they encode? J Neurophysiol
90:3137–54.
Kornylo K, Dill N, Saenz M, Krauzlis RJ. 2003. Cancelling of pursuit
and saccadic eye movements in humans and monkeys. J
Neurophysiol 89:2984–99.
Krauzlis RJ. 2000. Population coding of movement dynamics by cere-
bellar Purkinje cells. Neuroreport 11:1045–50.
Krauzlis RJ. 2001. Extraretinal inputs to neurons in the rostral superi-
or colliculus of the monkey during smooth-pursuit eye movements.
J Neurophysiol 86:2629–33.
Krauzlis RJ. 2003. Neuronal activity in the rostral superior colliculus
related to the initiation of pursuit and saccadic eye movements. J
Neurosci 23:4333–44.
Krauzlis RJ. 2004. Activity of rostral superior colliculus neurons dur-
ing passive and active viewing of motion. J Neurophysiol
92:949–58.
Krauzlis RJ, Basso MA, Wurtz RH. 1997. Shared motor error for mul-
tiple eye movements. Science 276:1693–5.
Krauzlis RJ, Basso MA, Wurtz RH. 2000. Discharge properties of neu-
rons in the rostral superior colliculus of the monkey during
smooth-pursuit eye movements. J Neurophysiol 84:876–91.
Krauzlis RJ, Dill N. 2002. Neural correlates of target choice for pursuit
and saccades in the primate superior colliculus. Neuron 35:355–63.
Krauzlis RJ, Lisberger SG. 1994a. A model of visually-guided smooth
pursuit eye movements based on behavioral observations. J Comp
Neurosci 1:265–83.
Krauzlis RJ, Lisberger SG. 1994b. Simple spike responses of gaze
velocity Purkinje cells in the floccular lobe of the monkey during
the onset and offset of pursuit eye movements. J Neurophysiol
72:2045–50.
Krauzlis RJ, Liston D, Carello CD. 2004. Target selection and the supe-
rior colliculus: goals, choices and hypotheses. Vis Res 44:1445–51.
Krauzlis RJ, Miles FA. 1996a. Decreases in the latency of smooth pur-
suit and saccadic eye movements produced by the “gap paradigm”
in the monkey. Vis Res 36:1973–85.
Krauzlis RJ, Miles FA. 1996b. Release of fixation for pursuit and sac-
cades in humans: evidence for shared inputs acting on different
neural substrates. J Neurophysiol 76:2822–33.
Krauzlis RJ, Miles FA. 1996c. Transitions between pursuit eye move-
ments and fixation in the monkey: dependence on context. J
Neurophysiol 76:1622–38.
Krauzlis RJ, Zivotofsky AZ, Miles FA. 1999. Target selection for pur-
suit and saccadic eye movements in humans. J Cogn Neurosci
11:641–9.
Kurylo DD, Skavenski AA. 1991. Eye movements elicited by electrical
stimulation of area PG in the monkey. J Neurophysiol 65:1243–53.
Kustov AA, Robinson DL. 1996. Shared neural control of attentional
shifts and eye movements. Nature 384:74–7.
Leon MI, Shadlen MN. 1999. Effect of expected reward magnitude on
the response of neurons in the dorsolateral prefrontal cortex of the
macaque. Neuron 24:415–25.
Liston D, Krauzlis RJ. 2003. Shared response preparation for pursuit
and saccadic eye movements. J Neurosci 23:11305–14.
Luschei ES, Fuchs AF. 1972. Activity of brain stem neurons during eye
movements of alert monkeys. J Neurophysiol 35:445–61.
THE NEUROSCIENTIST 135
MacAvoy MG, Gottlieb JP, Bruce CJ. 1991. Smooth-pursuit eye move-
ment representation in the primate frontal eye field. Cereb Cortex
1:95–102.
Masson GS, Stone LS. 2002. From following edges to pursuing
objects. J Neurophysiol 88:2869–73.
McPeek RM, Keller EL. 2002. Saccade target selection in the superior
colliculus during a visual search task. J Neurophysiol 88:2019–34.
McPeek RM, Keller EL. 2004. Deficits in saccade target selection after
inactivation of superior colliculus. Nat Neurosci 7:757–63.
Missal M, Heinen SJ. 2001. Facilitation of smooth pursuit initiation by
electrical stimulation in the supplementary eye fields. J
Neurophysiol 86:2413–25.
Missal M, Heinen SJ. 2004. Supplementary eye fields stimulation
facilitates anticipatory pursuit. J Neurophysiol 92:1257–62.
Missal M, Keller EL. 2001. Neurons active during both saccades and
smooth pursuit suggest a convergence of oculomotor systems in
the pontine reticular formation. Soc Neurosci Abstr 27:208.
Missal M, Keller EL. 2002. Common inhibitory mechanism for sac-
cades and smooth-pursuit eye movements. J Neurophysiol
88:1880–92.
Moore T, Armstrong KM. 2003. Selective gating of visual signals by
microstimulation of frontal cortex. Nature 421:370–3.
Moore T, Fallah M. 2004. Microstimulation of the frontal eye field and
its effects on covert spatial attention. J Neurophysiol 91:152–62.
Muggleton NG, Juan CH, Cowey A, Walsh V. 2003. Human frontal eye
fields and visual search. J Neurophysiol 89:3340–3.
Munoz DP, Pélisson D, Guitton D. 1991. Movement of neural activity
on the superior colliculus motor map during gaze shifts. Science
251:1358–60.
Munoz DP, Wurtz RH. 1993. Fixation cells in monkey superior col-
liculus I. Characteristics of cell discharge. J Neurophysiol
70:559–75.
Murthy A, Thompson KG, Schall JD. 2001. Dynamic dissociation of
visual selection from saccade programming in frontal eye field. J
Neurophysiol 86:2634–7.
Newsome WT, Wurtz RH, Dürsteler MR, Mikami A. 1985. Deficits in
visual motion processing following ibotenic acid lesions of the
middle temporal visual area of the macaque monkey. J Neurosci
5:825–40.
Ohtsuka K, Noda H. 1991. Saccadic burst neurons in the oculomotor
region of the fastigial nucleus of macaque monkeys. J
Neurophysiol 65:1422–34.
Olson CR, Gettner SN. 1995. Object-centered direction selectivity in
the macaque supplementary eye field. Science 269:985–8.
Osborne LC, Bialek W, Lisberger SG. 2004. Time course of informa-
tion about motion direction in visual area MT of macaque mon-
keys. J Neurosci 24:3210–22.
Pack CC, Born RT. 2001. Temporal dynamics of a neural solution to
the aperture problem in visual area MT of macaque brain. Nature
409:1040–2.
Payne BR, Rushmore RJ. 2003. Animal models of cerebral neglect and
its cancellation. Neuroscientist 9:446–54.
Petit L, Haxby JV. 1999. Functional anatomy of pursuit eye movements
in humans as revealed by fMRI. J Neurophysiol 82:463–71.
Platt ML, Glimcher PW. 1997. Responses of intraparietal neurons to
saccadic targets and visual distractors. J Neurophysiol 78:1574–89.
Platt ML, Glimcher PW. 1999. Neural correlates of decision variables
in parietal cortex. Nature 400:233–8.
Priebe NJ, Lisberger SG. 2004. Estimating target speed from the pop-
ulation response in visual area MT. J Neurosci 24:1907–16.
Rambold H, Churchland A, Selig Y, Jasmin L, Lisberger SG. 2002.
Partial ablations of the flocculus and ventral paraflocculus in mon-
keys cause linked deficits in smooth pursuit eye movements and
adaptive modification of the VOR. J Neurophysiol 87:912–24.
Rashbass C. 1961. The relationship between saccadic and smooth
tracking eye movements. J Physiol (Lond) 159:326–38.
Raybourn MS, Keller EL. 1977. Colliculo-reticular organization in pri-
mate oculomotor system. J Neurophysiol 40:861–78.
Recanzone GH, Wurtz RH. 2000. Effects of attention on MT and MST
neuronal activity during pursuit initiation. J Neurophysiol
83:777–90.
136 THE NEUROSCIENTIST
Rizzolatti G, Riggio L, Dascola I, Umilta C. 1987. Reorienting atten-
tion across the horizontal and vertical meridians: evidence in favor
of a premotor theory of attention. Neuropsychologia 25:31–40.
Robinson DA. 1972. Eye movements evoked by collicular stimulation
in the alert monkey. Vis Res 12:1795–808.
Robinson FR, Straube A, Fuchs AF. 1993. Role of the caudal fastigial
nucleus in saccade generation: II. Effects of muscimol inactivation.
J Neurophysiol 70:1741–58.
Rosano C, Krisky CM, Welling JS, Eddy WF, Luna B, Thulborn KR,
and others. 2002. Pursuit and saccadic eye movement subregions in
human frontal eye field: a high-resolution fMRI investigation.
Cereb Cortex 12:107–15.
Sato TR, Schall JD. 2003. Effects of stimulus-response compatibility
on neural selection in frontal eye field. Neuron 38:637–48.
Schiller PH, Chou I. 2000. The effects of anterior arcuate and dorso-
medial frontal cortex lesions on visually guided eye movements: 2.
Paired and multiple targets. Vis Res 40:1627–38.
Schiller PH, Chou IH. 1998. The effects of frontal eye field and dor-
somedial frontal cortex lesions on visually guided eye movements.
Nat Neurosci 1:248–53.
Schiller PH, Stryker M. 1972. Single-unit recording and stimulation in
superior colliculus of the alert rhesus monkey. J Neurophysiol
35:915–24.
Schiller PH, True SD, Conway JL. 1980. Deficits in eye movements
following frontal eye field and superior colliculus ablations. J
Neurophysiol 44:1175–89.
Schlack A, Hoffmann KP, Bremmer F. 2003. Selectivity of macaque
area VIP for smooth pursuit eye movements. J Physiol 551(pt
2):551–61.
Schlag-Rey M, Amador N, Sanchez H, Schlag J. 1997. Antisaccade
performance predicted by neuronal activity in the supplementary
eye field. Nature 390:398–401.
Scudder CA. 1988. A new local feedback model of the saccadic burst
generator. J Neurophysiol 59:1455–75.
Scudder CA, Kaneko CS, Fuchs AF. 2002. The brainstem burst gener-
ator for saccadic eye movements: a modern synthesis. Exp Brain
Res 142:439–62.
Scudder CA, McGee DM. 2003. Adaptive modification of saccade size
produces correlated changes in the discharges of fastigial nucleus
neurons. J Neurophysiol 90:1011–26.
Shadlen MN, Newsome WT. 2001. Neural basis of a perceptual deci-
sion in the parietal cortex (area LIP) of the rhesus monkey. J
Neurophysiol 86:1916–36.
Sheliga BM, Riggio L, Rizzolatti G. 1995. Spatial attention and eye
movements. Exp Brain Res 105:261–75.
Shi D, Friedman HR, Bruce CJ. 1998. Deficits in smooth-pursuit eye
movements after muscimol inactivation within the primate’s frontal
eye field. J Neurophysiol 80:458–64.
Sommer MA, Tehovnik EJ. 1997. Reversible inactivation of macaque
frontal eye field. Exp Brain Res 116:229–49.
Sparks D, Rohrer WH, Zhang Y. 2000. The role of the superior col-
liculus in saccade initiation: a study of express saccades and the
gap effect. Vis Res 40:2763–77.
Sparks DL, Sides JP. 1974. Brain stem unit activity related to horizon-
tal eye movements occurring during visual tracking. Brain Res
77:320–5.
Stanford TR, Sparks DL. 1994. Systematic errors for saccades to
remembered targets: evidence for a dissociation between saccade
metrics and activity in the superior colliculus. Vis Res 34:93–106.
Stone LS, Beutter BR, Lorenceau J. 2000. Visual motion integration
for perception and pursuit. Perception 29:771–87.
Takagi M, Zee DS, Tamargo RJ. 1998. Effects of lesions of the oculo-
motor vermis on eye movements in primate: saccades. J
Neurophysiol 80:1911–31.
Tanaka M, Lisberger SG. 2001. Regulation of the gain of visually
guided smooth-pursuit eye movements by frontal cortex. Nature
409:191–4.
Tanaka M, Lisberger SG. 2002a. Enhancement of multiple compo-
nents of pursuit eye movement by microstimulation in the arcuate
frontal pursuit area in monkeys. J Neurophysiol 87:802–18.
Voluntary Eye Movements
Tanaka M, Lisberger SG. 2002b. Role of arcuate frontal cortex of mon-
keys in smooth pursuit eye movements: I. Basic response proper-
ties to retinal image motion and position. J Neurophysiol
87:2684–99.
Thier P, Dicke PW, Haas R, Barash S. 2000. Encoding of movement
time by populations of cerebellar Purkinje cells. Nature 405:72–6.
Thompson KG, Bichot NP, Schall JD. 1997. Dissociation of visual dis-
crimination from saccade programming in macaque frontal eye
field. J Neurophysiol 77:1046–50.
Tian JR, Lynch JC. 1996a. Corticocortical input to the smooth and sac-
cadic eye movement subregions of the frontal eye field in Cebus
monkeys. J Neurophysiol 76:2754–71.
Tian JR, Lynch JC. 1996b. Functionally defined smooth and saccadic
eye movement subregions in the frontal eye field of Cebus mon-
keys. J Neurophysiol 76:2740–53.
Volume 11, Number 2, 2005
Tremblay L, Gettner SN, Olson CR. 2002. Neurons with object-
centered spatial selectivity in macaque SEF: do they represent
locations or rules? J Neurophysiol 87:333–50.
Van Gisbergen JAM, Robinson DA, Gielen S. 1981. A quantitative
analysis of generation of saccadic eye movements for burst neu-
rons. J Neurophysiol 45:417–42.
Wardak C, Olivier E, Duhamel JR. 2002. Saccadic target selection
deficits after lateral intraparietal area inactivation in monkeys. J
Neurosci 22:9877–84.
Yan YJ, Cui DM, Lynch JC. 2001. Overlap of saccadic and pursuit eye
movement systems in the brain stem reticular formation. J
Neurophysiol 86:3056–60.
Zee DS, Yamazaki A, Butler PH, Güçer G. 1981. Effects of ablation of
flocculus and paraflocculus on eye movements in primate. J
Neurophysiol 46:878–99.
THE NEUROSCIENTIST 137