U.S. copyright law
(title 17 of U.S. code)
governs the reproduction
and redistribution of
copyrighted material.
Analysis of Visual Behavior
tl
edited by
David J. Ingle
Melvyn A. Goodale
Richard J. W. Mansfield
1982 by
The Massachusetts Institute of Technology
All rights reserved. No part of this book may be reproduced in any form or
by any means, electronic or mechanical, including photocopying, recording, or
by any information storage and retrieval system, without permission in writing
from the publishers.
The MIT Press
Cambridge, Massachusetts
London, England
18 Two Cortical Visual Systems
Leslie G. Ungerleider
Mortimer Mishkin
549
Damage to the primary visual cortex (striate cortex or area 17), either
by disease, by acute assault, or through surgical intervention for
therapeutic purposes, results in a scotoma or blind region in the visual
field. Moreover, because of the precise topographic mapping of the visual
field onto striate cortex, the region of the scotoma corresponds to
the part of striate cortex damaged. In contrast to this primary sensory
impairment, ”higher-order” visual dysfunctions ensue when damage occurs
more rostrally in the brain, in either the temporal or parietal lobes;
however, the effects that follow lesions of these two cortical areas are
markedly different (Newcombe and Russell 1969).Whereas damage to
temporal cortex produces an impairment in visual recognition (see, for
example, Milner 1958, 1968; Kimura 1963; Lansdelll968; Benson et al.
1974;Meadows 1974),damageto parietal cortexproduces a constellation
of visual spatial impairments (see, for example, McFie et al. 1950;
Semmes et al. 1963; De Renzi and Faglioni 1967; Butters et al. 1972;
Ratcliff and Davies-Jones 1972; Ratcliff and Newcombe 1973).
Although it is clear that visual information must reach the temporal
and parietal association areas to enable their participation in visual recognition
aqd visual spatial perception, respectively, the complex circuitry
through which this information is transmitted has yet to be unraveled.
It is known, however, that two major fiber bundles emerge
from occipital cortex and project rostrally in the brain (Flechsig 1896,
1920).One, the superior longitudinal fasciculus, follows a dorsal path,
traversing the posterior parietal region in its course to the frontal lobe;
the other, the inferior longitudinal fasciculus, follows a ventral route
into the temporal lobe. It has been our working hypothesis (Mishkin
1972; Pohl 1973) that the ventral or occipitotemporal pathway is
specialized for object perception (identifyingwhat an object is) whereas
the dorsal or occipitoparietalpathway is specialized for spatial perception
(locating where an object is). This distinction between the two
types of visual perception is not new (see, for example, Ingle 1967; Held
1968). In the past, however, the neural mechanisms underlying object
and spatial vision were seen as localized in geniculostriate and
tectofugal systems, respectively (Schneider 1967; Trevarthen 1968).
Two Cortical Visual Systems
550
By contrast, in the present formulation, corticocortical connections
originating in the striate area are viewed as mediating both types of
vision, with two diverging cortical systems replacing the geniculostriate-tectofugal
dichotomy. The reasons for stressing corticocortical
mechanisms for both types of visual perception in primates
are developed below, but this emphasis is not meant to deny that the
tectofugal system (including the tectofugal pathway to cortex) contributes
to spatial vision, particularly to its visuomotor aspects; it is simply
that, with regard to the perceptual aspects of spatial vision, the tectofugal
system in the primate appears to play a subsidiary role. In our
investigations of the two cortical visual systems, we have used the
rhesus monkey (Macaca mulatta) as our subject and have employed a
combination of behavioral, electrophysiological, and anatomical
techniques.
Occipitotemporal Mechanisms in Object Perception
The results of numerous behavioral studies have demonstrated that
bilateral removal of inferior temporal cortex in monkeys produces a severe
impairment in visual-discrimination performance (for reviews see
Gross 1973; Dean 1976; Wilson 1978).In brief, inferior temporal lesions
produce a deficit that is exclusively visual, affecting both the retention
of discriminations acquired prior to surgery and the postoperative acquisition
of new discriminations. Among the deficits that have been
reported are those involving hue, brightness, two-dimensional patterns,
and three-dimensional shapes. More recent work has shown that
damage to the posterior part of the inferior temporal cortex (area TEO)
interferes mainly with discriminative ability, whereas damage to the
anterior part (area TE) affects primarily visual memory (Iwai and Mishkin
1968; Cowey and Gross 1970).
The pathway through which inferior temporal cortex receives visual
information was first suggested on the basis of early neuronographic
data (vonBonin et al. 1942).It had been shown in both the monkey and
chimpanzee that if strychnine is applied to the striate cortex spike discharges
can be recorded from a prestriate cortical belt, whereas if
strychnine is applied to any part of this prestriate region spikes can be
recorded in the inferior part of the temporal lobe. These neuronographic
findings were later confirmed in a neuroanatomical study
(Kuypers et al. 1965)that employed the Nauta-Gygax (1954) technique
for tracing projections by the silver-staining of degeneratingaxons after
removal of their cell bodies. As anticipated, it was found that striate
cortex projects to a prestriate corticalbelt, which, in turn, projectsto the
inferior temporal area. It was also confirmed that each prestriate area
projects across the splenium of the corpus callosum to reach the prestriate
area of the opposite hemisphere.
These neuronographic and neuroanatomical findings are scheUngerleider,
Mishkin
551
Figure 18.1 Striate-prestriate-temporal pathways. Abbreviations refer to cytoarchitectonic
divisions of von Bonin and Bailey (1947). Lower diagram indicates pathways remaining
after crossed striate and inferior temporallesions (inblack). Pathwaysare shown
as two-way, in accord with the neuroanatomicalevidence that the connectionsare reciprocal
(Kuypers et al. 1965; Rockland and Pandya 1979). Adapted from Mishkin 1966.
matized in figure 18.1. As indicated in the upper diagram, each striate
area transmits visual information, relayed through the prestriate
cortex, to the ipsilateral inferior temporal area; but, in addition, because
of the reciprocal prestriate connections across the corpus
callosum, each striate area also transmits visual information to the contralateral
inferior temporal area. The behavioral significance of these
pathways was first demonstrated in a series of crossed-lesion disconnection
experiments (Mishkin 1966). Since such disconnection studies
are not commonly employed to investigate brain-behavior relationships,
a brief explanation of the underlying logic may be in order.
The usual method of identifying the function of a cortical area is to
remove that same area from both hemispheres and then determine what
function has been lost. This method does not reveal whether the function
was lost because it depended on the area (or station) that was
removed or because it depended on a later station along the same
pathway that could no longer operate after the pathway leading to it
was destroyed. Crossed-lesion disconnection studies make it possible
to demonstrate that a pathway with many cortical stations exists. The
method is to remove one of the stations in the postulated pathway on
one side of the brain, and to remove a different one (later in the pathway)
on the other side (see lower diagram of figure 18.1).This leaves
Two Cortical Visual Systems
552
one of each pair in the series intact, and it can often be shown that up to
this point little or no loss in function occurs. However, if the connections
between the two hemispheres are then cut, the functions served
by the stations beyond the cut may be lost. This demonstrates that the
function in question depends not only on the stations themselves, but
also on the connections between them. The same demonstration theoretically
could have been achieved by cutting the fiber connections
between the stations within each hemisphere, but the complexity of
these connections is so great, and they form such a compact network,
that the method is technically difficult if not impossible.
In the disconnection study referred to above it was found that if an
inferotemporal lesion in one hemisphere was combined serially with
total striate removal in the other, animals continued to perform a
pattern-discrimination task; however, when these asymmetrical or
' crossed lesions were followed by transection of the corpus callosum, the
performance of the animals fell to chance and they failed to relearn the
task. Presumably, the single crossed pathway from the intact striate
cortex on one side across the corpus callosum to the intact inferior
temporal cortex on the other side is sufficient to mediate patterndiscrimination
habits, but if this pathway is cut a severe deficit results.
This study thus provided the first behavioral indication that inferior
temporal cortex is a late station along a corticalvisual pathway running
from striate through prestriate cortex.
Electrophysiological experiments (Gross et al. 1972)subsequently revealed
that single neurons in7inferiortemporal cortex, like those in
striate (Hubel and Wiesel 1968jand prestriate (Hubel and Wiesel 1970)
cortex, have visual receptive fields. The optimal trigger features for inferior
temporal neurons, however, are considerably more complex.That
visual input to these neurons originates in striate cortex was shown in a
combined recording and ablation study by Rocha-Mirandaet al. (1975).
The method entailed measuring the defect in the visual receptive fields
of inferior temporal neurons after selective cerebral lesions or commissural
transections. As predicted from earlier ablation studies (Mishkin
1972), the tectofugal pathway from the superior colliculus through the
pulvinar to cortex turned out to be unimportant for the receptive field
properties of inferior temporal neurons. By contrast, the cortical pathway
from striate through prestriate to inferior temporal cortex proved to
be essential, since interruption of this pathway by a striate removal
or by transection of the forebrain commissures eliminated the corresponding
visual input to inferior temporal cells. The results, shown in
figure 18.2, indicate that normally over 60% of inferior temporal
neurons have bilateral receptive fields. However, after bilateral striatecortex
removal these neurons are totally unresponsive to visual stimulation,
after unilateral striate-cortex removal they respond to visual
stimulation only in the hemifield opposite the intact striate cortex, and
after commissurotomy they respond to stimulation only in the conUngerleider,
Mishkin
Normal
BilateralFields
mContralateral Fields
aIpsilateral Fields
f i d e recordedfrom
,--.~-Inferior Temporal Cortex
-..:,_-..-* -PrestriateCortex
B C I Striate Cortex
" Q
Bilat. Striate Unilat. Striate Unilat. Striate
V
B C I B C I B C I
Split
B C I
Figure 18.2 Proportion of inferior temporal neurons that had bilateral, contralateral,or ipsilateral receptive
fields in (A) normal monkeys, (B) monkeys with bilateral removal of striate cortex, (C, D) monkeys with
unilateral removal of striate cortex, and (E) monkeys with transection of the forebrain commissures. The
brain diagrams show how information from the right (R) and left (L) visual hemifields could reach inferior
temporal cortex along a corticocortical route and how each lesion (in black) interferes with this pathway.
Adapted from Rocha-Miranda et al. 1975.
tralateral hemifield. Thus, the dependence of inferior temporal cortex
on corticocortical connections arising in striate cortex has now been
made evident at the single-cell level.
Most recently, this entire visual pathway was functionally mapped
using 14C-labeled2-deoxyglucose (2-DG)via the method developed by
Sokoloff and his colleagues (Kennedy et al. 1975). In this technique
2-DG is used as a marker of local cerebralglucose utilization and hence
indicates regions that are metabolically active during the experimental
procedure. The visual-mapping studies (Jarvis et al. 1978; Kennedy et
al. 1978)were carried out in awakemonkeys previously prepared with a
unilateral optic-tractsection combined with transection of the forebrain
commissures (a procedure that visually deafferented one hemisphere
while leaving the other intact). On the day of the experiment, the monkeys
were presented with visual patterns in a rotating drum or in a
discrimination apparatus. In both situations, reduced glucose utilization
in the blind as compared with the seeing hemisphere was deen
553 Two Cortical Visual Systems
554
cortically, not only in the geniculostriate projection but throughout the
entire expanse of prestriate and inferior temporal cortex as far forward
as the temporal pole. With one exception, no other cortical area in the
deafferented hemisphere showed reduced activity.
The evidence that has been cited favors a sequential-activation model
for object vision in which information that reaches the striate cortex is
transmitted for further processing to the prestriate cortex, and from
there to the inferior temporal area. This system appears to be important
for the analysis and coding of the physical dimensions of visual stimuli
needed for their identification and.recognition. It is unlikely, however,
that any part of this system up to and including the inferior temporal
area is involved in the still higher-order process of associating visual
stimuli with other events, such as motivational and emotional ones.
Recordings from single cells in monkeys performing visual discrimination
and reversal tasks have shown that although inferior temporal
neurons are sensitive to the physical properties of stimuli, they are relatively
insensitive to changes in the reward value of the stimuli (Jawis
and Mishkin 1977).Similar results have been reported by others (Rolls
et al. 1977; but see Ridley and Ettlinger 1973).Presumably, the process
of attaching reward value to a stimulus depends on stations beyond the
occipitotemporal pathway (Jonesand Mishkin 1972; Sunshine and Mishkin
1975; Spiegler and Mishkin 1978; Rolls et al. 1979b; Sanghera et al.
1979).However, for object vision, the inferiqr temporal cortex may well
be the final station. It is significant in this regard that, by virtue of the
extremely large receptive fields of inferior temporal neurons (Gross et
al. 1972), this area provides the neural mechanism of stimulus equivalence
across retinal translation (Gross and Mishkin 1977; Seacord et al.
1979)-thatis, the ability to recognize a stimulus as the same regardless
of its position in the visual field and, by extrapolation, regardless of its
spatial location. Indeed, a necessary consequence of this equivalence
mechanism is that within the occipitotemporalpathway there is a loss
of information about the spatial locations of objects.
Occipitoparietal Mechanisms in Spatial Perception
The neural mechanism for the analysis of the spatial locations of objects
also entails the transmission of visual information from the striate
through the prestriate area; however, the rest of the pathway for spatial
vision appears to be quite separate from the ventral pathway into the
temporal cortex. Evidence in support of this dichotomy of corticalvisual
systems comes from recent studies in our laboratory on the parietal
lobe.
In the initial study of this series, Pohl(1973) demonstrated a dissociation
of visual deficits after inferior temporal and posterior parietal
lesions. Whereas the temporal but not the parietal lesion produced a severe
impairment on an object discrimination task, just the reverse was
Ungerleider, Mishkin
555
found on tests in which the animal was required to choose a response
location on the basis of its proximity to a visual ”landmark” (seepart A
of figure 18.5). The results suggested that I’.. . the inferior temporal
cortex participates mainly in the acts of noticing and remembering an
object’s qualities, not its position in space,” and that ”conversely, the
posterior parietal cortex seems to be concerned with the perception of
spatial relations among objects, and not their intrinsic qualities” (Mishkin
1972).
Accumulating evidence from other laboratories supports this view of
posterior parietal function. Not only has the impairment of landmark
tasks after posterior parietal lesions been corroborated (Milner et al.
1977; Ungerleider and Brody 1977; Brody and Pribram 1978; however,
see Ridley and Ettlinger 1975),but impairments after such lesions have
also been found on other visual spatial tasks, including a stylus maze
(Milneret al. 1977),patterned-string tests (Ungerleiderand Brody 1977),
cage finding (Sugishita et al. 1978), and route following (Petrides and
Iversen 1979).Visual spatial disorientation, however, is not the only
deficit produced by lesions of this region. Indeed, the classical symptoms
of posterior parietal dysfunction are misreaching in the dark as
well as in the light (Ettlinger and Wegener 1958; Bates and Ettlinger
1960; Ettlinger and Kalsbeck 1962; Hartje and Ettlinger 1973; LaMotte
and Acuña 1978; Faugier-Grimaud et al. 1978; Stein 1978),contralateral
neglect of auditory and tactile as well as visual stimuli (Denny-Brown
and Chambers 1958; Heilman et al. 1970), and impairments of tactile
discrimination (Blum 1951; Pribram and Barry 1956; Wilson 1957, 1975;
Ettlinger and Wegener 1958; Pasik et al. 1958; Bates and Ettlinger 1960;
Wilson et al. 1960; Ettlinger and Kalsbeck 1962; Ettlinger et al. 1966;
Moffett et al. 1967; Moffett and Ettlinger 1970; Ridley and Ettlinger
1975).
The posterior parietal lesions in these studies have nearly always included
two or even more cytoarchitectonicareas, and so it is natural to
assume that the heterogeneity of effects is a consequence of the heterogeneity
of the tissue included in the removals. However, the few
studies that have directly examined this question (Moffett et al. 1967;
Ridley and Ettlinger 1975) failed to uncover any clear-cut instance of
functional dissociation. From the evidence at hand, the most parsimonious
interpretation is one that has been offered by Semmes
(1967), namely, that the multiple deficits are but different reflections
of a single, supramodal spatial disorder (see also Ratcliff et al. 1977).
According to this view, and in the light of more recent anatomical information
(Pandya and Kuypers 1969; Jones and Powell 1970); a supramodal
spatial framework could be constructed out of converging’
inputs to posterior parietal cortex from all exteroceptive sensory modalities,
with a significant contribution from vision. In the studies in our
laboratory that followed the ”landmark” experiment by Pohl(1973),we
addressed the questions raised above regarding localization of spatial
Two Cortical Visual Systems
556
function within the posterior parietal region and the dependence of this
function on visual inputs.
As in most other posterior parietal ablation studies, the lesions in
Pohl’s study were composites that included not only inferior parietal
but also dorsal prestriate tissue. To test for localization of function
within this region, we partitioned it into three approximately equal
sectors-inferior parietal, lateral preoccipital, and medial parietalpreoccipital-as
shown in figure 18.3. Monkeys were then prepared
with lesions of either one, two, or all three of these sectors and tested on
the landmark, or distance discrimination, task (Mishkin, Lewis, and
Ungerleider, in prep.). Like the earlierattempts, this onefailed to reveal
any evidence of functional specialization within the larger area. Instead,
there was an especially clear relationship between severity of effect
and extent of lesion, completely independent of locus. This finding
is illustrated in figure 18.4.
The fact that performance on the landmark task depends as much on
dorsal prestriate tissue as on inferior parietal tissue pointed to the possibility
that the visual information on which posterior parietal cortex
operates arises in striate cortex. The alternative possibility, that the
criticalvisual input arises in the superior colliculus,found no clear support
in a study of the effects of tectal lesions on landmark discrimination
(Snyderand Mishkin, unpublished data); even complete destruction of
the superior colliculus failed to produce a reliable loss in retention.
Therefore, to test for the functional dependence of posterior parietal
cortex on striate output, a crossed-lesion disconnection experiment
analogous to the striate-temporal disconnection experiment described
earlier (Mishkin 1966) was undertaken (Ungerleider and Mishkin
1978a).Monkeys were preoperativelytrained on the landmark task, and
then received a three-stage operation intended to serially disconnect
posterior parietal from striate cortex. At each stage, the monkeys were
given a preoperative retention test followed by surgery, allowed 2
weeks to recover, and then retested.
As shown in part B of figure 18.5, the first removal was a unilateral
posterior parietal ablation, which included all three sectors investigated
in the partial-lesion study to ensure a complete effect. On the hypothesis
that each striate areahas corticocorticalconnectionswith both posterior
parietal areas, this first lesion may be viewed as a test of the effects
of destroying two of these connections(theremaining two are indicated
by the black arrows in the figure). A comparison of preoperative with
postoperative retention scores on the task shows that this amount of
damage to the system had only minimal effect.
The second-stageremoval was a contralateral striate-cortexablation,
illustrated in part C of figure 18.5. It included all of the striate cortex,
both laterally and medially. The results of the retest given after this
operation revealed a severe impairment; the monkeysrequired an average
of 880 trials to relearn, despite the remaining interhemispheric corUngerleider,
Mishkin
ce
ci
t
trn
+3 -3 -9 -15
Sector I: Inferior Parietal mSector L: Lateral Preoccipital Sector M: Medial Parietal-Preoccipital
Figure 18.3 Partition of posterior parietal cortex into three sectors: inferior parietal (I), lateral preoccipital
(L), and medial parietal-preoccipital(M). Monkeys received bilaterally symmetrical lesions intended to remove
one (groups I, L and M), two (groups IL, LM, and IM), or all three (group PP) of these sectors. The
sectors are shown on lateral and medial surface views and on representative cross-sectionsof a standard
rhesus monkey brain. Numerals indicate approximate A-P stereotaxic levels. AC: anterior commissure. ai:
inferior arcuate sulcus. as: superior arcuate sulcus. Ca: calcarine fissure. CC: corpus callosum. ce: central
sulcus. ci: cingulate sulcus. CO: collateral sulcus. ec: ectocalcarine sulcus. h: hippocampal fissure. ip: intraparietal
sulcus. 1:lunate sulcus. la: lateral fissure. MI: massa intermedia. oi: inferior occipital sulcus. or:
orbital sulcus. ot: occipitotemporalsulcus. P: principal sulcus. PO: parieto-occipitalincisure. pom: medial
parieto-occipitalfissure. rh: rhinal fissure. ro: rostral sulcus. sp: subparietal sulcus. tn-ta: anterior middle
temporal sulcus. tmp: posterior middle temporal sulcus. ts: superior temporal sulcus.
557 Two Cortical Visual Systems
n
a
z
I
2-Sector Lesion
3-Sector Lesion
600
400
MEAN TRIALS
TO RELEARN
200
O
I L M IL LM IM PP
Figure 18.4 Postoperative retention on the landmark task after bilateral,lesions of one,
two, or three posterior parietal sectors, illustrated in Figure 18.3.Bars indicatemean trials
to criterion; circles within bars indicate mean errors. N in this and following figures
equals number of monkey in group. Mean number of trials to relearn task before operation
(dashed line) is based on scores on a two-week retention test given just prior to
surgery and represents the data from all 23 monkeys. The results showed no significant
differences among the groups given a one-sector lesion, or among the groups given a
two-sector lesion. However, as the number of sectors included in the posterior parietal
lesion increased, so did the severityof the impairment. (Source:Mishkin et al., in prep.).
tical connections between the intact striate and parietal areas. This
suggested that the interaction between the striate and parietal cortexvia
uncrossed connections had been of considerable importance for performance.
However, whether such interaction was mediated by a corticocorticalpathway
could only be determined by examining the effects
of the third stage of operation, transection of the corpus callosum.
In this operation, the posterior half of the callosum was transected,
thereby cortically disconnectingthe intact striate and posterior parietal
areas. As shown by the postoperative retention score in part D of figure
18.5, the effect of the transection was to produce a partial reinstatement
of the impairment. We can infer from these results that performance on
the landmark task had indeed been mediated via corticocortical pathways
between the striate and parietal areas.
However, the disruption of landmark discrimination that followed
this lesion was not complete; all monkeys successfully relearned the
task. At this point, the question was whether recovery of the discrimination
was mediated by the interaction of the intact posterior parietal
cortex with subcortical visual structures or by the interaction of the intact
striate cortex with other cortical areas. In subsequent work on this
preparation we found that removal of the remaining posterior parietal
cortex was completely without effect, which indicated that tectofugal
558 Ungerleider, Mishkin
A B
C D
Figure 18.5 Design and results of the striate-parietal crossed-lesion disconnection study. (A) Landmark
task. Monkeys were rewarded for choosing the covered foodwell located closer to a striped cylinder (the
"landmark"), which was positioned on the leftor the right randomlyfromtrial to trial, but always5 cm from
one foodwell and 20 cm from the other. Training was given for 30 trials per day to a criterion of 90 correct
responses in 100consecutive trials. (B) Discriminationretention before and after first-stagelesion (unilateral
posterior parietal; N = 3); 10 preoperative trials and 130postoperative trials. (C) Discrimination retention
before and after second-stage lesion (contralateralstriate; N = 3); 70 preoperative and 880 postoperative
trials. (D)Discriminationretention before and after third-stage lesion (corpuscallosum;N = 3); 30 preoperative
and 400 postoperative trials. At each stage the lesion is shown in black and the lesions of prior stages
are shaded. Arrows denote hypothetical connections left intact by lesions. Adapted from Ungerleider and
Miskin 1978a.
559 Two Cortical Visual Systems
inputs to this cortex, which appear to be unnecessary when corticocortical
connections are intact, also play no significant role in the recovery
after corticocortical disconnection. By contrast, removal of prestriate
cortex ipsilateral to the intact striate cortex produced a marked impairment
on the task. Taken together, the results of this serial-lesion experiment
demonstrate that the visual spatial functions of posterior
parietal cortex depend on inputs fromstriate cortex and that such inputs
are transmitted across corticocorticalpathways. In this regard, it is significant
that the only cortical area apart from those along the occipitotemporal
pathway to show differential glucose utilization in the
2-DG experiments described earlier (Jarvis et al. 1978; Kennedy et al.
1978)was located within the posterior parietal region. The affected area
extended forward from the dorsal prestriate cortex to include the posterior
half of the inferior parietal gyrus, both on its crown and on the
ventral bank of the intraparietal sulcus.
The Two Cortical Visual Systems Compared
These results on striate-parietal interaction parallel the earlier finding
that corticocortical inputs from striate cortex are crucial for the visual
recognition functions of inferior temporal cortex. But, though the results
of these two studies are analogous, one important difference
emerges (seefigure 18.6).On the pattern-discriminationtask of the earlier
study, only a moderate effect followed the striate lesion, and a severe
deficit did not result until the callosal transection. By contrast, on
the landmark task, the relative effectsof the striate and callosal lesions
were reversed; that is, the striate-cortex lesion produced the more severe
effect. These data suggest that the posterior parietal cortex, like the
inferior temporal, depends heavily on corticocorticalinputs from striate
cortex for its visual function, but that, unlike the inferior temporal, the
posterior parietal cortex in a given hemisphere does not seem to receive
a heavy input via the corpus callosum from striate-cortex neurons representing
the ipsilateral visual field. Thus, each posterior parietal area
may be organized largely as a substrate for contralateral spatial function,
and this could account in part for the symptom of contralateral
neglect that has so often been reported after unilateral injury to the
parietal lobe in man (see, for example, Denny-Brown and Chambers
1958).
A second important difference in the organization of visual inputs to
the two systems was uncovered in an experiment comparing the behavioral
effects of selective striate-cortexremovals. In this experiment,
monkeys received bilateral lesions of the striate area representing either
central vision (lateral striate) or peripheral vision (medial striate). The
lateral lesion included the entire lateral surface of striate cortex, and the
medial lesion included the calcarine fissure as well as both banks of its
ascending and descending limbs (figure 18.7).The monkeys were tested
560 Ungerleider, Mishkin
A
Figure 18.6 Comparison of two crossed-lesion disconnection experiments: (A) pattern discrimination after
striate-temporaldisconnection(Mishkin1966)and (B) landmark discrimination after striate-parietaldisconnection
(Ungerleiderand Mishkin 1978a).In each case, brain diagrams illustrate the three lesions (in black)
involved in the disconnection and bar graphs show the mean trials to relearn after each lesion. F indicates
failure to relearn within the limits of training.
Figure 18.7 Locus of two different bilateral striate-cortex removals (in black): (A) lateral striate, the area
representing centralvision (N = 5); (B) medial striate, the area representing peripheral vision (N = 5). For
details of visuotopic organization see figure 18.13; for abbreviations see figure 18.3.
561 Two Cortical Visual Systems
B
1PP
LS
1MS
1000
800
600
400
200
O
aControl (N = 6)
W Posterior Parietal (N = 3)
f9Inferior Temporal (N = 3)
Bzls Lateral Striate (N =5)
IMedial Striate (N =5)
Figure 18.8 Comparisonof the effectsof lateral striatelesions with those of medial striate
lesions on (A)postoperativeacquisitionof a pattern discriminationand (B) postoperative
retention of the landmark task. Adapted from Ungerleider and Mishkin 1978a.
before operation both on a pattern-discrimination task, to assess residual
inferior temporal function, and on the landmark task, to assess
residual posterior parietal function. For purposes of comparison, their
scores are plotted in figure 18.8 together with those of monkeys with
either bilateral inferior temporal or bilateral posterior parietal lesions
The data indicate that on the pattern-discrimination task severe impairment
was produced by inferior temporal but not by posterior
parietal lesions, whereas on the landmark task severe impairment followed
posterior parietal but not inferior temporal lesions. This, then, is
another instance of the dissociation of visual recognition and visual
spatial functions of temporal and parietal cortex, respectively. Of more
direct concern for the question of differences in anatomical organization,
however, were the effects of the striate lesions. Here the data indicate
that on the pattern-discrimination task impairment was produced
only by lateral striate-cortex lesions, whereas on the landmark task
equally severe impairment followed lateral and medial striate-cortexlesions.
Apparently, inputs from central vision are especially important
for the visual recognition functions of the inferior temporal cortex, but
inputs from central and peripheral vision are equally important for the
visual spatial functions of posterior parietal cortex.
Thus, although interactions with striate cortex are critical for the
parietal just as for the temporal area, the inputs of the striate cortex to
these two regions appear to be organized differently; relative to inferior
temporal cortex, posterior parietal cortex receives a smallercontribution
from contralateral striate-cortex inputs (representing the ipsilateral visual
field) but a greater contribution from medial inputs (representing
562 Ungerleider, Mishkin
A
+t B
+t
t- t - t
Figure 18.9 Receptive-field properties of neurons in (A)inferior temporal and (B) posterior
parietal cortex. Axes represent horizontal and vertical meridians. Plus sign indicates
upper or right visual field; minus sign indicates lower or left. Scaleis in degrees of visual
angle. A slow-moving stimulus typically effective in activating the given population of
neurons is shown at the left of each diagram. The enclosed geometricshape within each
set of axes indicates the border of the receptive field for a typical neuron in that population.
Note that of the inferior temporal neurons over 60% are bilateral, 100%include the
fovea, and there is selectivityfor stimulus features; of the posterior parietal neurons over
60% are contralateral only, over 60% do not include the fovea, and there is no selectivity
for stimulus features. (PartA adapted from Gross et al. 1972; part B adapted from Robinson
et al. 1978.)
the peripheral visual field). Interestingly, these differences in the organization
of striate-cortexinputs inferred from the ablation studies are
clearly reflected in the receptive-field properties of neurons within the
inferior temporal (Grosset al. 1972;Jarvis and Mishkin 1975;Rolls et al.
1977) and posterior parietal areas (Hyvärinen and Poranen 1974;
Mountcastle et al. 1975; Robinson et al. 1978; Rolls et al. 1979a). The
relevant properties are shown diagrammatically in figure 18.9.
First, there is a smaller representation of the ipsilateral visual field in
the receptive fields of posterior parietal neurons than in those of inferior
temporal neurons: Over 60% of inferior temporal neurons have
bilateral receptive fields, while over 60% of posterior parietal neurons
have contralateralfields only. Second, there is a relatively greater representation
of the peripheral visual field in the receptive fields of posterior
parietal neurons than in those of inferior temporal neurons: 100%
of inferior temporal neurons have receptive fields that include the
fovea, while over 60% of posterior parietal neurons have receptive
fields outside the fovea. Finally, and perhaps most important, not only
are the striate-cortexinputs to the temporal and parietal areas organized
differently, but these inputs must alsobe carrying different information
to the two areas, as is indicated by the behavioral evidence for functional
dissociation. This too is clearly seen at the level of single neurons,
for, in contrast to the specificand often complex trigger features needed
for maximal activation of inferior temporal neurons (see, for example,
Gross et al. 1972), most posterior parietal neurons can be maximally
driven by simple spots of light (seeRobinson et al. 1978).These several
differences in the receptive-field properties of inferior temporal and
posterior parietal neurons, each of which has a behavioral correlatethat
was revealed in the ablation studies, presumably reflect differences in
the cortical processing required for object versus spatial vision.
563 Two Cortical Visual Systems
Corticocortical Pathways Through Prestriate Cortex
564
The evidence presented thus far strongly supports the view that neural
mechanisms underlying object and spatial perception depend on the
relay of different kinds of visual information from striate cortex through
prestriate cortex to targets in inferior temporal and posterior parietal
areas, respectively. Despite the abundance of positive evidence, however,
the support has not been unanimous. The results of prestriatecortex
ablation studies have repeatedly raised serious problems for this
conception of corticocortical pathways. If prestriate cortex were an essential
relay in either a striate-temporal or a striate-parietal pathway,
then damage to this relay should yield efffects at least as severe as damage
to its target areas. Yet, monkeys sustaining extensive bilateral prestriate
lesions commonly exhibit only mild visual effects. In a few instances
in which severe visual deficits did follow prestriate removals
(Keatingand Hore11972; Keating 1975),the removals included the posterior
part of inferior temporal cortex (area TEO), damage to which, by
itself, produces severe visual-discrimination impairment (Iwai and
Mishkin 1968). Extensive prestriate-cortex removals that spare this
posterior temporal region, however, have repeatedly failed to yield appreciable
effects in either visual discrimination or in spatial orientation
(Lashley 1948;Meyer et al. 1951;Riopelle et al. 1951; Chow 1952; Evarts
1952; Pribram et al. 1969; Ungerleider and Pribram 1977).
Is there a resolution of this paradox? It was suggested by Mishkin
(1966) that failures to obtain severe impairments after extensive prestriate-cortex
removals simply indicated that spared prestriate remnants
continued to serve as effectiverelays between the striate area and
its rostral targets-that is, that the prestriate cortex was invested with
a high degree of equipotentiality. According to this proposal, only a
complete prestriate lesion would be expected to yield a corticocortical
disconnection. The results of a study by Iwai and Mishkin (Mishkin
1972) prdvided support for this proposal with regard to occipitotemporal
transmission. When putatively completeprestriate-cortex lesions
were made in a large group of monkeys, none was able to relearn a
visual pattern discrimination. By contrast, partial prestriate ablations,
irrespective of their location, were nearly without effect.These findings
therefore pointed to the existence of multiple pathways through prestriate
cortex, any of which can be utilized to convey visual information
out of striate cortex. It seemed that only if all these equivalent pathways
were destroyed would the transmission be completely disrupted and
the anticipated deficit ensue.
Although it has since been found that even the massive removals
of prestriate cortex referred to above did not totally disconnect the
striate-temporal pathway, the conclusion reached remains valid nonetheless.
Indeed, it has now been demonstrated that any sparing, no
matter how minimal, of the visual functions of inferior temporal corUngerleider,
Mishkin
565
tex after a prestriate ablation is directly attributable to the continued
relay of information across a viable prestriate pathway. This demonstration
emerged from a series of interrelated behavioral, electrophysiological,
and anatomical experiments (Ungerleider, Iwai, Gross,
'Bender, Snyder, and Mishkin, in prep.) that began in the following
way: In an attempt to verify that the inferior temporal cortex had
been functionally disconnected in the Iwai-Mishkin study, two of the
monkeys that had failed to show relearning after the prestriate lesions
were given prolonged pattern-discrimination training by a method of
approximation, in preparation for a second-stage lesion. The method
entailed presentation of a series of stimulus pairs that differed first in
brightness, then in size, then in contour, and finally in pattern; the
successive pattern pairs ,approximated the originals more and more
closely until the original pattern discrimination had been relearned.
Over several months both monkeys were successfully retrained by this
method, and both were then given bilateral inferior temporal ablations
in a second-stage operation. The supposition was that if the initial prestriate
removal had produced a total disconnection of inferior temporal
from striate cortex, then the slow pattern-discrimination relearning by
approximation should have been achieved without the participation of
inferior temporal cortex, and, consequently, removal of this cortex
should not disrupt the relearned habit. As it turned out, however, the
inferior temporal ablation produced a complete reinstatement of the
deficit in both monkeys. Thus, unless inferior temporal cortex can participate
in visual discrimination learning in the absence of all corticocortical
visual input, the only conclusion to be drawn from this
unexpected result is that the initial prestriate-cortex removal had not
produced a total striate-temporal disconnection after all.
In order to examine directly this question of preserved neural transmission
despite massive prestriate removals, an electrophysiological
study was undertaken in animals prepared with removals identical to
those in the Iwai-Mishkin study except that they were limited to one
hemisphere. The distribution of the receptive fields of inferior temporal
neurons in the operated animals is summarized in figure 18.10. According
to the proposed route of visual information flow, indicated by the
arrows in the brain diagrams, these neurons should have responded to
visual stimulation only in the hemifield opposite the intact prestriate
cortex. That is, inferior temporal neurons in the intact hemisphere
should have had strictly contralateralvisual fields, while inferior temporal
neurons in the hemisphere with the prestriate ablation should
have had strictly ipsilateral fields. In short, the results should have duplicated
those obtained after unilateral striate removals, illustrated in
parts C and D of figure 18.2. In fact, however, the receptive fields of
approximately25% of the neurons sampled did not fit this prediction.
In the normal monkey, stimulation in a given hemifield will activate
about 80% of inferior temporal neurons. For example, as indicated in
Two Cortical Visual Systems
A B
loor r I
" B C I B C
Figure 18.10 Proportion of inferior temporal neurons that had bilateral (B), contralateral
(C),or ipsilateral(I)receptivefields after unilateral ablation of prestriate cortex (in black).
Data for part A were obtained from inferior temporal neurons contralateral to the lesion,
whereas data for part B were obtained from those ipsilateral to the lesion. Asterisks denote
the receptive fields deviating from the prediction; that is, deviating from those
obtained after unilateral striate-cortexremovals, shown in figure 18.2. Adapted from Ungerleider,
Snyder, and Mishkin, in prep.
part A of figure 18.2, stimulation in the right hemifield will activate
about 90% of the inferior temporal neurons in the left hemisphere (all
those with either bilateral or contralateral visual fields) and about 70%
of the inferior temporal neurons in the right hemisphere (allthose with
either bilateral or ipsilateral visual fields). By contrast, in the animals
with left prestriate removals, stimulation in the right hemifield activated
about 25% of inferior temporal neurons. That is, as indicated in
figure 18.10, right-hemifield stimulation activated about 15%of the inferior
temporal neurons in the left hemisphere and about 35% in the
right. Thus, the massive prestriate removals did greatly reduce the visual
input to inferior temporal neurons-in fact, by more than twothirds
(80% to 25%)-thereby accounting for the severe deficit that
such lesions produced in pattern discrimination relearning. On the
other hand, the removals did not completely eliminate striate input to
inferior temporal neurons (25%, as compared with 0% after striate removals),
and this fact accounted for the ultimately successfulretraining
by approximation that was achieved in the animals with such lesions.
Even with this electrophysiologicalevidence, the conclusion that be-
566 Ungerleider, Mishkin
567
havioral sparing resulted from sparing of remnants of the prestriate
relay was only inferential. In an attempt to demonstrate the preserved
pathway directly, a neuroanatomical experiment was undertaken on
two additional monkeyswith massive prestriate removalsthat had been
successfully retrained by approximation on the pattern discrimination.
More than a year after the prestriate-cortex removals, one of these animals
was given a second-stage operation in which the entire occipital
lobe posterior to the initial lesion was removed from one hemisphere,
and the brain was examined for anterograde degeneration by the
Fink-Heimer (1967) technique. The critical question was whether degeneration
would appear in the inferior temporal cortex ipsilateral to
the occipital lobectomy. Since striate cortex does not project directly to
the inferior temporal area, the presence of degeneration here would indicate
that the lobectomy had destroyed prestriate-relay tissue spared
by the initial lesion-a finding that would account for the functional
sparing that had been demonstrated both behaviorally and electrophysiologically.
In fact, terminal degeneration did result in the ipsilateral
inferior temporal cortex, specificallyin the ventral part of area
TEO. In addition, degenerating material was found in an unexpected
area, the floor of the caudal portion of the superior temporal sulcus,
which turned out to be a second spared route through which visual
input could reach inferior temporal cortex. This was demonstrated in
the other monkey of the anatomical experiment, which was given a
second-stagelesion in just this portion of the superior temporal sulcus
of one hemisphere, again more than a year after the initial prestriate
removal. Terminal degeneration was found in the ipsilateral inferior
temporal cortex of this case also, specifically in the dorsal part of area
TEO.
The foregoing series of experiments made it clear that, although the
massive prestriate-cortexremovals had severely disrupted the prestriate
relay, they had not totally abolished it. As illustrated in figure 18.11,
spared pathways that presumably continued to serve as effective relays
were directly demonstrated anatomically. At the same time, however,
the anatomical results raised some important new questions. For example,
what were the loci of the spared prestriate remnants whose removal
gaverise to the degeneration in the ventral part of areaTEO? Also, what
was the source of the unexpected degeneration in the depth of the
caudal portion of the superior temporal sulcus, an area from which a
projection had been traced to dorsalTEO? These questions highlighted
our lack of knowledgeabout the precise links through which visual information
originating in the striate cortex is transmitted to inferior
temporal cortex. Even greater ignorance surrounded the location and
arrangement of the links in the striate-parietal pathway. It became apparent
that, in order to obtain this information, a comprehensive
anatomical investigation was needed, beginning with an analysis of the
projections of the striate cortex itself.
Two Cortical Visual Systems
568
Figure 18.11 Schematicrepresentation of results from two anatomical experiments, demonstrating
that striate-prestriate-temporalpathways were preserved despite massiveprestriate
removals. In the first experiment, an occipital lobectomy (dots) was performed
more than a year after the initial prestriate removal (stripes). Degeneration from the
lobectomy was found in ventral area TE0 and in the floor of the caudal portion of the
superior temporal sulcus. The exact source of these two projections, however, was indeterminate
(indicated by question mark). In the second experiment, the floor of the
superior sulcus was removed, again more than a year after the initial prestriate lesion.
Degenerationin this casewas found in dorsal area TEO. For abbreviations seefigure 18.3.
Adapted from Ungerleider et al., in prep.
Clarification of Striate-Prestriate Connections
Early attempts to examine striate-prestriate connections (von Bonin et
al. 1942; Bailey et al. 1944) employed the method of strychnine
neuronography, described above. Although the map of striate projections
obtained with this method has been confirmed by more recent
work using degeneration (Kuypers et al. 1965; Cragg and Ainsworth
1969; Zeki 1969, 1971a; Jones and Powell 1970) and autoradiographic
(Zeki 1976)and horseradish-peroxidase (Lund et al. 1975)tracing techniques,
all these methods, both old and new, have been used to define
projectionsprimarily from lateral striate cortex, the part of striate cortex
representing central vision. Surprisingly, there has been almost no information
regarding the projections from posterior and medial striate
cortex, the parts representing peripheral and far-peripheral vision. Indeed,
it was undoubtedly the near absence of information about the
locationof the prestriate tissue servingnoncentralvision that accounted
for the repeated failure, recounted above, to completely disconnect the
higher-order visual areas from their striate input. We therefore undertook
a study of the cortical efferents ffom all parts of striate cortex, with
the aim of defining the locus, extent, and topographic organization of
the entire striate-prestriate projection system.
To delineate the entire map of striate projections to prestriate cortex,
one series of monkeys was prepared with partial striate lesions such
that, collectively, they included all of area 17with little or no invasion of
area 18.The brains were then processed by the Fink-Heimer (1967)procedure
for silver staining of degenerating axon terminals. Reconstructions
of the lesions in three of the five cases from this series are shown
’ Ungerleider, Mishkin
569
in figure 18.12. The lateral lesion involved the lateral surface of area 17
only, with no invasion of area 18; the posterior lesion included the lateral
and medial banks of the verticallimbs of the calcarine fissure, again
with no invasion of area 18; and the medial striate lesion involved the
tissue within the stem of the calcarine fissure, in this case with slight
damage to area 18around the lips of the fissure. According to the electrophysiologicalmap
of Daniel and Whitteridge (1961),these three sectors
of striate cortex correspond to the central 7" of the contralateral
visual field, to the field between 7" and 22" from fixation, and to eccentricities
greater than 22" from fixation, respectively (see figure 18.13).
Thus, it was anticipated that not only would the entire striate-prestriate
projection system emerge from the data on this series of monkeys, but
the generalvisuotopic organization of the system would be apparent as
well.
To verify the degeneration results, and also to investigate the details
of the visuotopic organization, a second series of monkeys were prepared
with injections of radioactivelylabeled amino acids into selected
sites in striate cortex, and the brains processed for autoradiography. A
summary of the injection sites is shown in figure 18.14. These loci correspond
to positions in the visual field ranging from less than %" from
fixation (site number 1on the lateral surface) to greater than 45" from
fixation (site number 10 in the stem of the calcarine fissure). The loci
injected included representations of both the upper and the lower visual
fields.
The results from both sets of experiments, degeneration and autoradiographic,
indicated that all parts of striate cortex project to three
separate visual areas within prestriate cortex (Ungerleider and Mishkin
1979a): The first is a circumstriate cortical belt surrounding area 17 at
the 17-18 border, the second is located along the caudal portion of the
superior temporal sulcus, and the third is buried deep within the caudal
part of the intraparietal sulcus. Each of these three projection areas will
be described in turn.
Within the first projection area-the circumstriate cortical belt-the
representations of the upper and- lower visual fields are entirely separate
(figure 18.15).The topographic organization of these separate representations
is illustrated in figure 18.16. Progression from central to
peripheral to far peripheral vision is represented in the lower field by a
progression into the posterior bank and depth of the lunate sulcus, medially
along the surface of the buried annectent gyrus into the parietooccipitalincisure,
and then rostrally along the upper lip of the calcarine
fissure; and in the upper field by a progression into the inferior occipital
sulcus, ventromedially into the occipitotemporaland collateral sulci,
and then rostrally along the lower lip of the calcarine fissure. The total
extent of this projection field corresponds remarkably closely to area OB
of von Bonin and Bailey (1947); indeed, the two may be equivalent.
Moreover, the autoradiographic evidence indicates that whereas the
Two Cortical Visual Systems
A
h
Figure 18.12 Location of corWl-da-mage in brains with removals of (A)lateral, (B) posterior, or (C)medial
striate cortex. The lesions are mapped on lateral and medial surface views and on representative crosssections
through a standard rhesus monkey brain. Cross-sectionsare at levelsindicated by vertical lines on
surfaceviews. Dashed lines indicate extentof striate cortexon cross-sectionsbut borders of striate cortex on
surfaceviews; anterior limit of striate cortexwithin the calcarinefissure is indicated on medialsurfaceviews
by dashed arrow. Removalof striate cortex is shown in crosshatch; black arrows on the medial surfaceviews
indicate removal of striate tissue from within the banks and depths of the calcarine fissure. Damage to
prestriate tissue is shown in stripes. For abbreviations see figure 18.3.
570 Ungerleider, Mishkin
Figure 18.13 Relationship between topography of striate cortex and representation of the contralateralvisualfield.
Accordingto the electrophysiologicalmap of Daniel and Whitteridge (1961),shown at bottom, the
centerof gaze (fixation)is represented in the far anterolateralpart of striate cortex, just below the ventral tip
of the lunate sulcus.Therepresentation of the vertical meridian passes through fixation(X) and continues all
along the 17-18 border (dashed line). The representation of the horizontal meridian is a horizontal line (not
illustrated)passing from fixation across the lateral surface, entering the depth of the calcarine fissure, and
continuing to the rostral limit of the medial striate cortex (dashed arrow). The upper visual field is represented
below the horizontal meridian and the lower visual field above the horizontal meridian. The lateral
surface of striatecortex (A),the lateral bank of the calcarinelimbs (B), the medial bank of the calcarinelimbs
(C,D), and the stem of the calcarinefissure (E, F) represent approximatelythe center 7",7"-13", 13"-22", and
eccentricities greater than 22" from fixation, respectively. Thus, a progression from lateral to posterior to
medial striate cortex corresponds to a progression from central to peripheral to far-peripheral vision.
vertical meridian is represented at the inner boundary of this projection
field (that is, at the striate-OBborder), the horizontal meridian is represented
at its outer boundary (at the OB-OA border).
The second projection field of striate cortex is located within cytoarchitectonic
area OA along the caudal portion of the superior temporal
sulcus (Ungerleider and Mishkin 197813,197913).As indicated in the lateral
view of the brain shown in figure 18.17, the ventral limit of this
region can be demarcated by an imaginary line connecting the ventral
tips of the lunate and intraparietal sulci; from this limit the region extends
dorsocaudally for about 1cm to the point at which the superior
temporal sulcus frequently bifurcates, sending one spur forward into
the inferior parietal lobule. Within this portion of the superior temporal
sulcus there is again an orderlymapping of the contralateralvisual field,
571 Two Cortical Visual Systems
1
-1o -12 -14
-18 -2o -22 -24 -26
Figure 18.14 Sites in striate cortex injected with tritiated amino acids. Injection sites
(shown in black) are numbered according to the eccentricityof their representation in the
visual field. These representations ranged from less than one-half degree from fixation
(site 1)to greater than 45" from fixation (site 10). Each injection was made in a separate
hemisphere. Dashed lines indicate extent of striate cortex on cross-sections,'but borders
of striate cortex on surfaceviews; anterior limit of striate cortex within the calcarine fissure
is indicated on medial surface view by a dashed arrow. Negative numerals indicate
approximate A-P stereotaxic levels. For abbreviations see figure 18.3.
572 Ungerleider, Mishkin
573
Figure 18.15 Photomicrographs exemplifying autoradiographic results after injections of
striate cortex.Thebright-fieldphotomicrograph on the left shows an injection(markedby
arrows) involving both the dorsal and ventral banks of the calcarine fissure (see figure
18.14,injectionsite 10).In the dark-field photomicrograph on the right, labelingis seen in
two prestriate loci, marked by arrows: dorsally in the medial parieto-occipitalfissure and
ventrallyin the ventral bank of the calcarinefissure. Thepatchypattern of label, especially
apparent in the ventral locus, is typical of striate projections to the circumstriate cortical
belt. Adapted from Ungerleider and Mishkin 1979a.
as shown on the three selected cross sections. Progression from central
to peripheral to far-peripheral vision is represented by a progression
down the posterior bank of the superior temporal sulcus and up along
the lower and then the upper part of the sulcal floor. Figure 18.18summarizes
this topographic arrangement. The finding that the floor of the
superior temporal sulcus receives a direct projection from posterior and
medial striate cortex provides an answer to one of the anatomical questions
posed in the preceding section (seefigure 18.11).As yet, however,
we do not have an answer to the second question raised there, except
the confirmation that the immediate source of the occipitalprojections
to ventral area TE0 is not the striate cortex.
The third area that receives direct striate-cortex projections is also
located within cytoarchitectonic area OA, in the depth of the most
caudal portion of the intraparietal sulcus. A cross-section through the
projection shows that it is buried in the foot-shaped part of the sulcus
beneath its lateral bank (figure 18.19).This projection field is by far the
smallest of the three; its total rostral-caudal length extends only about 2
mm. There is no discernible topographic representation of the visual
field within this area, for after a lesion or an injection involving any part
of striate cortex a projection is always seen in the same part of the sulcus.
Thus, although the dark-field photomicrograph in figure 18.19
shows degeneration in the foot of the intraparietal sulcus after a medial
striate lesion, the very same picture could as well represent the projections
of either posterior or lateral striate cortex. Indeed, the first evidence
of direct striate projections to this area came from studies that
Two CorticalVisual Systems
o1
-18
Figure 18.16 Location, extent, and topographic organization of striate projections (O: lateral; A: posterior;
U:medial) to the circumstriatecorticalbelt. The drawings are compositesbased on both degeneration and
autoradiographic material. Shaded area represent total projection field on surface views. As shown on the
surface views of the hemisphere, this projection zone surrounds the striate cortex at the 17-18 border and
corresponds remarkably closely to cytoarchitectonic area OB of von Bonin and Bailey (1947). Within the
projectionzone, there is a precisetopographicmap of the contralateralvisualfield (seecross-sections);note,
however, that the representations of the upper and lower fields are entirely separate. Dashed lines indicate
extent of striate cortex on cross sections, but borders of striate cortex on surface views. Numerals indicate
approximateA-P stereotaxiclevels. For abbreviations see figure 18.3. Adapted from Ungerleider and Mishkin
1979a.
had examined only lateralstriate connections (Kuyperset al. 1965;Jones
and Powell 1970),though the area had not been recognized as a separate
projection field at that time.
These three projection fields thus comprise all the possible first-stage
prestriate relays through which striate output can reach the inferior
temporal and posterior parietal cortex. Undoubtedly, second- and
third-stage prestriate relays also participate in both corticocortical
pathways (Zeki 1971b; Mesulam et al. 1977; Stanton et al. 1977; Desirnone
et al. 1980), although the exact location and topographic arrangement
of these further relays remain to be delineated. But even
from the evidence pertaining to the first stage alone, we can begin to
understand why attempts to produce striate-temporal disconnection by
removal of prestriate cortex have repeatedly failed. In all prestriate-
574 Ungerleider, Mishkin
-8
Figure 18.17 Location, extent, and topographic organization of striate projections (O: lateral; A: posterior;
B:medial) to cortex in the superior temporal sukus. The drawings are compositesbased on both degeneration
and autoradiographicmaterial. Shaded area represents locus of projection on lateral view. All areas of
striate cortex project to a restricted region inside the caudalportion of the superior temporal sulcus, its A-P
extent indicated on the lateral view of the hemisphere. As indicated on the cross-sections, the topographic
arrangement of the contralateral visual field is such that a progression from central to peripheral to far
peripheral vision is represented by a progression down the posterior bank of the sulcus and up along the
lower and then the lowerpart of the sulcalfloor. Althoughthis topography is clearestin the rostralportion of
the projection zone, it is also discernible caudally. Numerals refer to approximateA-P stereotaxiclevels. For
abbreviations see figure 18.3. Adapted from Ungerleider and Mishkin 1979b.
575
ablation studies, certain areas have consistently been spared that can
now be identified as receiving direct projections from posterior and
medial striate cortex. These prestriate areas are the medial portion of
area OB (figure 18.16), the floor of the caudal portion of the superior
temporal sulcus (figure18.17), and the depth of the most caudal portion
of the intraparietal sulcus,(figure 18.19). Each of these three areas has
escaped damage for a different reason: the first because of its relative
inaccessibility, the second because it was not even recognized to be a
part of prestriate cortex, and the third because its precise location had
not been described. In light of the nearly complete sparing of these
areas in most ablation studies, the finding of only mild visual impairment
is no longer surprising. Only when these areas were invaded substantially,
as in the massive prestriate removals of the Iwai-Mishkin
study, did severe visual impairment result. However, even in those leTwo
CorticalVisual Systems
Figure 18.18 Drawings of the lateral and medialviews of the left hemisphere and of a cross-section through
the superior temporal sulcus to illustrate the topographic arrangement of striate projections to this secondaryvisual
area. The orderlymapping within the sulcusof the entire representation of the contralateralvisual
field is shown. For abbreviations see figure 18.3. Source: Ungerleider and Mishkin 1979b.
sions, where there was almost total destruction of the prestriate areas
that receive direct projections from lateral.striate cortex (the part representing
central vision), there was no more than 50% destruction (Ungerleider
et al., in prep.) of the prestriate areas that receive direct projections
from posterior and medial striate cortex (theparts representing
peripheral and far peripheral vision). Clearly, it was the sparing of this
tissue that accounted for the continued (though reduced) transmission
of visual information to inferior temporal cortex that had been demonstrated
not only behaviorally but also electrophysiologically. Thus, the
present anatomicalfindings, many of which have now been confirmed
by others (Weller and Kaas 1978; Maunsell et al. 1979; Rockland 1979;
Van Essen et al. 1979;Weller et al. 1979),provide a coherent explanation
for a largebody of seeminglyparadoxical results regarding the effects of
prestriate lesions. That is, just as "peripheral"striatecortex can mediate
pattern discrimination in the absence of "central"striate cortex (Blakeet
576 Ungerleider, Mishkin
577
Figure 18.19 The striate projection zone within the intraparietal sulcus. This zone is located
in the depth of the most caudal portion of the sulcus. Its approximate A-P level is
indicated by the vertical line on the lateral view of the hemisphere, and its depth by the
black square on the cross-section. The dark-field photomicrograph shows the locus of
degeneration after a medial striate lesion. Projections are seen in this very same region
from all parts of striate cortex, which suggests the absence of a visuotopic organization
within this projection zone. Source: Ungerleider and Mishkin 1979a.
al. 1977), so can "peripheral"prestriate cortex substitute for "central"
prestriate cortex in this function. And if this is the case for the visual
recognition functions of inferior temporal cortex, where inputs from
peripheral vision are less important than those from centralvision, then
the same must surely be true for the visual spatial functions of posterior
parietal cortex, where inputs from peripheral vision are equal in importance
to those from central vision (see figure 18.8).
Now that the total system of striate efferents has been delineated, the
effects of completely disconnecting temporal or parietal from striate
cortex can finally be investigated. Such disconnection should be
achievableby removal of just the three striateprojection fields that have
been described, without inclusion of any other prestriate tissue. In addition,
with special testing methods it may now be possible to study the
behavioral effects of prestriate-cortex damage without producing total
Two CorticalVisual Systems
578
disconnection. In view of the visuotopic organization of the striate
projection zones in both area OB and the caudal part of the superior
temporal sulcus, the details of which have been corroborated electrophysiologically
(Gattass and Gross 1979; Gattass et al. 1979), even
limited lesions within these prestriate zones should yield severe visual
deficits if the animals are forced to use the part of the visual field corresponding
to the area damaged. Testing of visual functions within
specified parts of the visual field can be accomplished in monkeys who,
have been trained to maintain fixation (Wurtz 1969).Thus, the combination
of this method and the use of lesions based on our new understanding
of prestriate anatomy offers a promising approach for the
study of the processing characteristics and functional organization of
prestriate cortex.
Summary and Conclusions
The hypothesis that has been guiding our research is that appreciation
of an object’s qualities and of its spatial location depends on the processing
of different kinds of visual information in the inferior temporal
and posterior parietal cortex, respectively. From an initial concern with
these higher-order visual areas, our research has proceeded backward.
Having first obtained strong support from ablation studies for the postulated
dichotomyof temporal and parietal function, we next examined
the pathways through which visual information reaches these two cortical
areas. Converging evidence from a variety of sources-behavioral
and electrophysiologicaldisconnectionexperiments as well as anatomical
and metabolic mapping studies-indicated that both the temporal
and parietal visual areas depend on corticocortical inputs relayed from
striate through prestriate cortex. The results of prestriate ablation
studies, however, conflicted with this conclusion, since they commonly
indicated only modest deficits in both visual recognition and spatial
orientation. The solution to this puzzle emerged from a series of interrelated
behavioral, electrophysiological, and anatomical experiments
that demonstrated that every case of functional sparing after prestriate
damage can be directly attributed to the continued relay of information
through viableprestriate remnants. To determine the locus of these prestriate
remnants, we turned to a study of the cortical efferents from all
parts of striate cortex. The results revealed that all prestriate lesions to
date, no matter how massive, have failed to include varying extents of
tissue that receive direct projections from posterior and medial striate
cortex, the parts representing peripheral vision. Having delineated the
projections of striate cortex, which were found to consist of three separate
re-representations of the visual field, we are now in a position to
proceed in a forward direction and follow these projections, with the
inferior temporal and posterior parietal cortex as our targets. In the
course of this endeavor, a major goal will be to determine where within
Ungerleider, Mishkin
579
the complex of prestriate cortex the two corticalvisual systems begin to
diverge. On the assumption that both systems can indeed be followed
stepwise to our target areas, not only in the temporal but also in the
parietal lobe, a major question for the future will be how the object and
spatial information carried in these two separated systems are subsequently
integrated into a unified visual percept.
Acknowledgments
We gratefully acknowledge the contribution of past members of the
Laboratory of Neuropsychology at the National Institute of Mental
Health: Eiichi Iwai, Charlene Drew Jarvis, Michael,E. Lewis, Walter G.
Pohl, and Marvin Snyder. We also wish to thank our collaboratorsfrom
other laboratories: Charles G. Gross and David B. Bender at Princeton
University, Edward G. Jones at Washington University, Charles Kennedy
and Louis Sokoloff at NIMH, and Blair H. Turner at Howard University.
Skillful technical assistance was provided by Margaret E.
Knapp, Bertha McClure, and George Creswell. The research was supported
in part by the National Institutes of Health under postdoctoral
fellowship EY05009.
References
Bailey, P., G. von Bonin, E. W. Davis, H. W. Garol, and W. S. McCulloch. 1944. ”Further
observations on association pathways in the brain of Macaca mulatta.” 1,NeuropathoZ.
Exp. Neurol. 3: 413-415.
Bates, J. A. V., and G. Ettlinger. 1960. ”Posterior biparietal ablations in the monkey.”
Arch. Neurol. 3: 177-192.
Benson, D. F., J. Segarra, and M. L. Albert. 1974. “Visual agnosia-prosopagnosia.” Arch.
Neurol. 30: 307-310.
Blake, L., C. D. Jarvis, and M. Mishkin. 1977. ”Pattern discrimination thresholds after
partial inferior temporal or lateral striate lesions in monkeys.” Brain Res. 120: 209-220.
Blum, J. S. 1951. “Cortical organization in somesthesis: Effects of lesions in posterior
associative cortex in somatosensory function in Macaca mulatta.” Comp. Psychol. Monogr.
20: 219-249.
Brody, B. A., and K. H. Pribram. 1978. ”The role of frontal and parietal cortex in cognitive
processing: Tests of spatial and sequence functions.” Brain 101: 607-633.
Butters, N., C. Soeldner, and P.Fedio. 1972. ”Comparison of parietal and frontal lobe
spatial deficits in man: Extrapersonal vs. personal (egocentric) space.” Percept. Motor
Skills 34: 27-34.
Chow, K. L. 1952.”Further studies on selective ablation of associative cortex in relation to
visually mediated behavior.” 1.Comp. Physiol. Psychol. 45: 109-118.
Two Cortical Visual Systems
580
Cowey, A., and C. G. Gross. 1970. ”Effects of fovealprestriate and inferotemporallesions
on visual discrimination by rhesus monkeys.” Exp. Brain Xes. 11:128-144.
Cragg, B. G., and A. Ainsworth. 1969. ”The topography of the afferentprojections in the
circumstriate visual cortex of the monkey studied by the Nauta method.” Vision Res. 9:
733-747.
Daniel, P. M., and D. Whitteridge. 1961. “The representation of the visual field on the
cerebral cortex in monkeys.” J. Physiol. 159: 203-221.
Dean, P. 1976. ”Effects of inferotemporal lesions on the behavior of monkeys.” Psychol.
Bull. 83: 41-71.
Denny-Brown, D.; and R. A. Chambers. 1958. ”The parietal lobe and behavior.” Res.
Publ. Assoc. Res. Nerv. Ment. Dis. 36: 35-117.
De Renzi, E., and P. Faglioni. 1967. ”The relationship between visuo-spatial impairment
and constructional apraxia.” Cortex 3: 327-342.
Desimone, R., J. Fleming, and C. G. Gross. 1980. ”Prestriate afferentsto inferior temporal
cortex: An HRP study.” Brain Res. 184: 41-55.
Ettlinger, G., and J.E. Kalsbeck. 1962. “Changes in tactile discrimination and in visual
reaching after successive and simultaneous bilateral posterior parietal ablations in the
monkey.” J. Neurol. Neurosurg. Psychiatr. 25: 256-268.
Ettlinger, G., and J. Wegener. 1958. ”Somaesthetic alternation, discrimination, and
orientation afterfrontaland parietal lesions in monkeys.” Q. J. Exp. Psychol. 10: 177-186.
Ettlinger, G., H. B. Morton, and A. Moffett. 1966. ”Tactile discrimination performancein
the monkey: The effect of bilateral posterior parietal and lateral frontal ablations, and of
callosal section.” Cortex 2: 5-29.
Evarts, E. V. 1952.”Effects of ablation of prestriate cortexin auditory-visual associationin
monkey.” J. Neurophysiol. 15: 191-200.
Faugier-Grimaud, S., C. Frenois, and D. G. Stein. 1978. ”Effects of posterior parietal lesions
on visually guided behavior in monkeys.” Neuropsychologia 16: 151-168.
Fink, R. P., and L. Heimer. 1967. ”Two methods for selective silver impregnation of degenerating
axons and their synaptic endings in the centralnervous system.” Brain Res. 4:
369-374.
Flechsig, P.1896. Gehirn und Seele. Leipzig: von Veit.
. 1920. Anatomie des menschlichen Gehirns und Rückenmarks auf myelogenetischer
Grundlage. Leipzig: Thieme.
Gattass, R., and C. G. Gross. 1979. ”A visuotopically organized area in the posterior
superior temporal sulcus of the macaque.” Invest. Ophihalmol. (ARVO suppl.) 18: 184.
Gattass, R., J. H. Sandell,and C. G. Gross. 1979. ”V2 in the macaque: Visuotopic organization
and extent.” Soc. Neurosci. Abstr. 5: 786.
Ungerleider, Mishkin
581
Gross, C. G. 1973. ”Visual functions of inferotemporal cortex.” In Handbook of Sensory
Physiology, VII13, R. Jung, ed. Berlin: Springer.
Gross, C. G., and M. Mishkin. 1977. ”The neural basis of stimulus equivalence across
retinal translation.” In Lateralization in the Nervous System, S . Harnad et al., eds. New
York: Academic.
Gross, C. G., C. E. Rocha-Miranda, and D. B. Bender. 1972. ”Visualproperties of neurons
in inferotemporal cortex of the macaque.” J. Neurophysiol. 35: 96-111.
Hartje, W., and G. Ettlinger. 1973. “Reaching in light and dark after unilateral posterior
parietal ablations in the monkey.” Cortex 9: 344-352.
Heilman, K. M., D. N. Pandya, and N. Geschwind. 1970. ”Trimodal inattention following
parietal lobe ablations.” Trans. Amer. Neurol. Assoc. 95: 259-261.
Held, R. 1968. ”Dissociation of visual functions by deprivation and rearrangement.”
Psychol. Forsch. 31: 338-348.
Hubel, D. H., and T. N. Wiesel. 1968. “Receptive fields and functional architecture of
monkey striate cortex.” J. Physiol. 195: 215-243.
.1970. ”Stereoscopicvision in macaque monkey.” Nature 225: 41-42.
Hyvärinen, J., and A. Poranen. 1974. ”Function of the parietal associative area 7 as revealed
from cellular dischargesin alert monkeys.” Brain 97: 673-692.
Ingle, D. 1967. ”Twovisual mechanismsunderlying the behavior of fish.” Psychol. Forsch.
31: 44-51.
Iwai, E., and M. Mishkin. 1968. ”Two visual foci in the temporal lobe of monkeys.” In
Neurophysiological Basis of Learning and Behavior, N. Yoshii and N. A. Buchwald, eds.
Osaka University Press.
Jarvis, C. D., and M. Mishkin. 1975. “Responses of inferior temporal neurons to visual
discriminanda.” Soc. Neurosci. Abstr. 1: 61.
.1977. ”Responses of cells in the inferior temporalcortex of monkeys during visual
discrimination reversal.” Soc. Neurosci. Abstr. 3: 564.
Jarvis, C. D., M. Mishkin, M. Shinohara, O. Sakurada, M. Miyaoka, and C. Kennedy.
1978. ”Mapping the primate visual system with the [14C]2-deoxyglucosetechnique.” Soc.
Neurosci. Abstr. 4: 632.
Jones, B., and M. Mishkin. 1972. ”Limbic lesions and the problem of stimulusreinforcement
associations.”Exp. Neurol. 36: 362-377.
Jones, E. G., and T. P. S. Powell. 1970. ”An anatomical study of convergingsensory pathways
within the cerebral cortex of the monkey.” Brain 93: 793-820.
Keating, E. G. 1975. ”Effectsof prestriate and striate lesions on the monkey’s ability to
locate and discriminate visual forms.” Exp. Neurol. 47: 16-2.5.
Two CorticalVisual Systems
582
Keating, E. G., and J. A. Horel. 1972. ”Effects of prestriate and striate lesions on performance
of simple visual tasks.” Exp. Neurol. 35: 322-336.
Kennedy, C., M. H. Des Rosiers, J. W. Jehle,M. Reivich, F. Sharp, and L. Sokoloff. 1975.
”Mapping of functional neural pathways by autoradiographic survey of local metabolic
rate with [14C]deoxyglucose.”Science 187: 850-853.
Kennedy, C., C. D. Jarvis, O. Sakurada, and M. Mishkin. 1978. “A delineation of the
visually responsive loci of the temporal lobe by means of [14C]deoxyglucose.”Neurology
28: 366.
Kimura, D. 1963. ”Right temporal-lobe damage.” Arch. Neurol. 8:264-271.
Kuypers, H. G. J. M., M. K. Szwarcbart, M. Mishkin, and H. E.. Rosvold. 1965. ”OCcipitotemporal
corticocorticalconnections in the rhesus monkey.” Exp. Neurol. 11: 245-
262.
LaMotte, R. H.,-andC. Acuña. 1978. ”Defects in accuracy in reaching after removal of
posterior parietal cortex in monkeys.” Brain Res. 139: 309-326.
Lansdell, H. 1968. ”Effectof extent of temporal lobe ablationson two lateralized deficits.”
Physiol. Behav. 3: 271-273.
Lashley, K. S. 1948. “The mechanism of vision. XVIII. Effects of destroying the visual
’associative areas’ of the monkey.” Genet. Psychol. Monogr. 37: 107-166.
Lund, J. S., R. D. Lund, A. E. Hendrickson, A. H. Bunt, and A. F. Fuchs. 1975. “The
origin of efferentpathways from the primary visual cortex, area 17, of the macaque monkey
as shown by retrograde transport of horseradish peroxidase.” J.Comp. Neurol. 164:
287-304.
Maunsell, J. H. R., J. L. Bixby, and D. C. Van Essen. 1979. “The middle temporal area
(MT)in the macaque: Architecture, functionalproperties, and topographicorganization.”
Soc. Neurosci. Abstr. 5: 796.
McFie, J., M. F. Piercy, and O. L. Zangwill. 1950. ”Visual-spatialagnosiaassociatedwith
lesions of the right cerebral hemisphere.” Brain 73: 167-190.
Meadows, J. C. 1974. ”The anatomical basis of prosopagnosia.” J.Neural. Neurosurg.
Psychiatr. 37: 489-501.
Mesulam, M.-M., G. W. Van Hoesen, D. N. Pandya, and N. Geschwind. 1977. “Limbic
and sensory connectionsof the inferiorparietal lobule (areaPG) in the rhesus monkey: A
study with a new method for horseradish peroxidase histochemistry.” Brain Res. 136:
393-414.
Meyer, D. R., H. F. Harlow, and H. W. Ades. 1951. ”Retention of delayed responses and
proficiencyin oddity problems by monkeyswith preoccipital ablations.”Amer. J.Psychol.
44:391-396.
Milner, A. D., E. M. Ockleford, and W. Dewar. 1977. ”Visuo-spatialperformance following
posterior parietal and lateral frontal lesions in stumptail macaques.” Cortex 13: 350-
360.
Ungerleider, Mishkin
583
Milner, B. 1958. “Psychological defects produced by temporal-lobe excision.” Res. Publ.
Assoc. Res. Nerv. Ment. Dis. 36: 244-257.
. 1968. ”Visual recognition and recall after right temporal-lobe excision in man.”
Neuropsychologia 6: 191-209.
Mishkin, M. 1966. ”Visual mechanisms beyond the striate cortex.” In Frontiers in
Physiological Psychology, R. Russell, ed. New York: Academic.
.1972. ”Cortical visual areas and their interactions.” In Brain and Human Behavior,
A. G. Karczmar and J.C. Eccles, eds. Berlin: Springer.
Moffett, A. M., and G. Ettlinger. 1970. ”Tactile discrimination performance in the monkey:
The effect of unilateral posterior parietal ablations.” Cortex 6: 47-67.
Moffett, A., G. Ettlinger, H. B. Morton, and M. F. Piercy. 1967. ”Tactile discrimination
performance in the monkey: The effect of ablation of various subdivisions of posterior
parietal cortex.” Cortex 3: 59-96.
Mountcastle, V. B., J. C. Lynch, and A. Georgopoulos. 1975. ”Posterior parietal association
cortex of the monkey: Command functions for operations within extrapersonal
space.” J. Neurophysiol. 38: 871-908.
Nauta, W. J. H., and P. A. Gygax. 1954.”Silverimpregnation of degenerating axonsin the
central nervous system: A modified technique.’’ Stain Tech. 29: 91-94.
Newcombe, F., and W. R. Russell. 1969. ”Dissociated visual perceptual and spatial deficits
in focal lesions of the right hemisphere.’’ J. Neurol. Neurosurg. Psychiat. 32: 73-81.
Pandya, D. N., and H. G. J.M. Kuypers. 1969. “Cortico-corticalconnectionsin the rhesus
monkey.” Brain Res. 13: 13-36.
Pasik, P., T. Pasik, W. S. Battersby, and M. B. Bender. 1958. ”Visual and tactual discriminations
by macaqueswith serialtemporal and parietal lesions.”1.Comp. Physiol. Psychol.
51: 427-436.
Petrides, M., and S. D. Iversen. 1979. “Restricted posterior parietal lesions in the rhesus
monkey and performance on visuospatial tasks.” Brain Res. 161: 63-77.
Pohl, W. 1973. ”Dissociation of spatial discrimination deficits following frontal and
parietal lesions in monkeys.” J. Comp. Physiol. Psychol. 82: 227-239.
Pribram, H. B., and J. Barry. 1956. ”Further behavioral analysis of parieto-temporopreoccipital
cortex.” J. Neurophysiol. 19: 99-106.
Pribram, K. H., D. N. Spinelli, and S. L.Reitz. 1969. ”The effects of radical disconnexion
of occipital and temporal cortex on visual behaviour of monkeys.’’ Brain 92: 301-312.
Ratcliff, G., and G. A. B. Davies-Jones.1972. ”Defectivevisual localizationin focal brain
wounds.” Brain 95: 49-60.
Ratcliff, G.,and F. Newcombe. 1973. “Spatial orientation in man: Effectsof left, right, and
bilateral posterior cerebral lesions.” J.Neurol. Neurosurg. Psychiat. 36: 448-454.
Two Cortical Visual Systems
584
Ratcliff, G., R. M. Ridley, and G. Ettlinger. 1977.”Spatial disorientation in the monkey.”
Cortex 13: 62-65.
Ridley, R. M., and G. Ettlinger.1973.”Visualdiscrimination performancein the monkey:
The activity of single cells in infero-temporal cortex.’’ Bruin Res. 55: 179-182.
. 1975. “Tactile and visuo-spatial discrimination performance in the monkey:
The effects of total and partial posterior parietal removals.” Neuropsychologia 13:
191-206.
Riopelle, A. J., H. F. Harlow, P..H. Settlage, and H. W. Ades. 1951. ”Performance of
normal and operated monkeys on visual learning tests.” J. Comp. Physiol. Psychol. 44:
283-289.
Robinson, D. L., M. E. Goldberg, and G. B. Stanton. 1978. ”Parietal association cortex in
the primate: Sensory mechanisms and behavioral modulations.” J. Neurophysiol. 41:
910-932.
Rocha-Miranda, C. E., D. B. Bender, C. G. Gross, and M. Mishkin. 1975. “Visual activation
of neurons in inferotemporalcortex depends on striate cortex and forebrain commissures.”
J.Neurophysiol. 38: 475-491.
Rockland, K. S. 1979. ”Cortical connections of the occipital lobe in the rhesus monkey.’’
Ph.D. diss., Boston University.
Rockland, K. S., and D. N. Pandya. 1979. “Laminar origins and terminations of cortical
connections of the occipital lobe in the rhesus monkey.” Bruin Res. 179: 3-20.
P
Rolls, E. T., S. J. Judge, and M. K. Sanghera. 1977.”Activityof neurones in the inferotemporal
cortex of the alert monkey.” Brain Res. 130: 229-238.
Rolls, E. T., D. Perrett, S. J. Thorpe, D. Puerto, A. Roper-Hall, and S. Maddison. 1979a.
”Responses of neurons in area 7 of the parietal cortex to objectsof different significance.”
Bruin Res. 169: 194-198.
Rolls, E. T., M. K. Sanghera, and A. Roper-Hall, 1979b. “The latency of activation of
neurones in the lateral hypothalamus and substantia innominata during feeding in the
monkey.” Bruin Res. 164: 121-135.
Sanghera, M. K., E. T. Rolls, and A. Roper-Hall. 1979.”Visual responses of neurons in the
dorsolateral amygdala of the alert monkey.’’ Exp. Neurol. 63: 610-626.
Schneider, G. E. 1967. ”Contrasting visuomotor functions of tectum and cortex in the
golden hamster.” Psychol. Forsch. 31: 52-62.
Seacord, L., C. G. Gross, and M. Mishkin. 1979. ”Role of inferior temporal cortex in
interhemispheric transfer.” Bruin Res. 167: 259-272.
Semmes, J. 1967. ”Manual stereognosis after brain injury.’’ In Symposium on Oral Sensution
und Perception, J. F. Bosma, ed. Springfield, Ill.: Thomas.
Semmes, J., S. Weinstein, L. Ghent, and H.-L. Teuber. 1963. ”Correlates of impaired
orientation in personal and extrapersonal space.” Bruin 86: 747-772.
Ungerleider, Mishkin
585
Spiegler, B. J., and M. Mishkin. 1978. ”Evidence for the sequential participation of inferior
temporal cortex and amygdala in stimulus-reward learning.” Soc. Neurosei. Abstr.
4: 263.
Stanton, G. B., W. L. R. Cruce, M. E. Goldberg, and D. L. Robinson. 1977.”Some ipsilateral
projections to areas PF and PG of the inferior parietal lobule in monkeys.” Neurosci.
Lett. 6: 243-250.
Stein, J. 1978.“Effects of parietal lobe cooling on manipulative behaviour in the conscious
monkey.” In Active Touch, G. Gordon, ed. Oxford: Pergamon.
Sugishita, M., G. Ettlinger, and R. M. Ridley. 1978. “Disturbance of cage finding in the
monkey.” Cortex 14: 431-438.
Sunshine, J., and M. Mishkin. 1975. ”A visual-limbic pathway serving visual associative
functions in rhesus monkeys.” Fed. Proc. 34: 440.
I Trevarthen, C. B. 1968. ”Two mechanisms of vision in primates.” Psychol. Forsch. 31:
299-337.
Ungerleider, L. G., and B. A. Brody. 1977. ”Extrapersonalspatial orientation: The role of
posterior parietal, anterior frontal, and inferotemporal cortex.” Exp. Neurol. 56: 265-280.
Ungerleider, L. G., and M. Mishkin. 1978a. ”Interactions of striate and posterior parietal
cortex in spatial vision.” Soc. Neurosci. Abstr. 4: 649.
. 197813.”The visual area in the superior temporal sulcus of Macaca mulatta: Location
and topographic organization.” Anat. Rec. 190: 568.
. 1979a. ”Three cortical projection fields of area 17 in the rhesus monkey.” Soc.
Neurosci. Abstr. 5: 812.
. 197910. “The striate projection zone in the superior temporal sulcus of Macaca
mulatta: Location and topographic organization.” J. Comp. NeuroZ. 188: 347-366.
Ungerleider, L. G., and K. H. Pribram. 1977. ”Inferotemporal versus combined
pulvinar-prestriate lesions in the rhesus monkey: Effects on color, object, and pattern
discrimination.” Neuropsychologia 15: 481-498.
Van Essen, D. C., J. H. R. Maunsell, and J. L. Bixby. 1979. ”Areal boundaries and topographic
organization of visual areas V2 and V3 in the macaque monkey.” Soc. Neurosci.
Abstr. 5: 812.
Weller, R. E., and J. H. Kaas. 1978.”Connections of striate cortex with the posterior bank
of the superior temporal sulcus in macaque monkeys.” Soc. Neurosci. Abstr. 4: 650.
Weller, R. E., J. Graham, and J. H. Kaas. 1979. ”Cortical connections of striate cortex in
macaque monkeys.” Invest. Ophthalmol. (ARVO suppl.) 18: 157.
von Bonin, G.,and P. Bailey. 1947. The Neocortex of Mpcaca Mulatta. Urbana: University
of Illinois Press.
von Bonin, G., H. W. Garol, and W. S. McCulloch. 1942. ”The functional organization of
the occipital lobe.” Bid. Symp. 7: 165-192.
Two Cortical Visual Systems I
Wilson, M. 1957. “Effects of circumscribed cortical lesions upon somesthetic and visual
discrimination in the monkey.” ]. Comp. Physiol. Psychol. 50: 630-635.
.1965.”Tactual discrimination learning in monkeys.’’Neuropsychologia 3: 353-361.
. 1978. ”Visual system: Pulvinar-extrastriate cortex.” In Handbook of Behavioral
Neurobiology, vol. 1:Sensory Integration. R. B. Masterton, ed. New York: Plenum.
Wilson, M., J. A. Stamm, and K. H. Pribram. 1960. ”Deficits in roughness discrimination
after posterior parietal lesions in monkeys.” J. Comp. Physiol. Psychol. 53: 535-539.
Wurtz, R. H. 1969.”Visual receptivefields of striatecortexneurons in awakemonkeys.”].
Neurophysiol. 32: 727-742.
&ki, S. M. 1969. ”Representation of centralvisual fields in prestriate cortex of monkey.”
Brain Res. 14: 271-291.
. 1971a. Tonvergent input from the striate cortex (area 17) to the cortex of the
superior temporal sulcus in the rhesus monkey.” Brain Res. 28: 338-340.
.1971b. ”Cortical projections from two prestriate areas in the monkey.” Brain Res.
34: 19-35.
.1976. “The projectionsto the superior temporal sulcus from areas 17 and 18in the
rhesus monkey.” Proc. Roy. Soc. B 193: 199-207.
586 Ungerleider, Mishkin