Cerebral Color Blindness:
An Acquired Defect in Hue Discrimination
Alan L. Pearlman, MD, Jennifer Birch, MPhil, and John C. Meadows, M D

In contrast to the traditional view that striate visual cortex (area 17) is surrounded by two homogeneous cortical
areas (areas 18 and 19), recent studies have shown that mammalian extrastriate visual cortex contains several
anatomically and functionally distinct subregions. One such region, the V-4complex of the rhesus monkey, is
highly specialized for the analysis of color information, suggesting that a lesion in a homologous region might
produce a defect in color vision while sparing other visual functions.
We have studied a patient whose clinical syndrome supports this suggestion: a 44-year-old man with normal color
vision suffered two cerebral infarctions that produced first a right and then a left superior homonymous quadrantanopia and also caused prosopagnosia, topographical disorientation, and severely impaired color vision. Computed
tomography demonstrated extensive lesions in both inferior occipital lobes in the territories of the lateral branches
of the posterior cerebral arteries, involving the lingual and medial occipitotemporal gyri bilaterally; these gyri
contain the inferior portion of striate cortex and segments of extrastriate visual cortex. The patient had no difficulty
in giving the correct color names associated with common objects presented either verbally or in outline drawings.
Standardized testing with the Farnsworth-Munsell 100-hue test, the Nagel anomaloscope, and a method that tests
for just-noticeable differences between monochromatic stimuli all showed that the patient’s ability to distinguish
one color from another was markedly impaired but not totally absent. In contrast, visual acuity, reading, visually
guided eye movements, and stereopsis were normal. Cells in the V-4 complex of monkey extrastriate cortex are
highly specialized for distinguishing one color from another; the hue discrimination deficit that was demonstrated
in this patient with cerebral color blindness indicates that a region or regions with similar function has been
damaged.
Pearlman AL, Birch J , Meadows JC: Cerebral color blindness: an acquired defect in hue discrimination.
Ann Neurol 5:253-261, 1979
The extensive analysis of the mammalian visual system carried out over the last several years has demonstrated clearly that individual neurons at each level
of the visual system can be categorized by carefully
describing the features of the visual stimuli that cause
them to respond. Each cell receives input directly or
indirectly from a restricted part of the retina known
as its receptive field, which in turn represents a restricted part of the visual world. The analysis of receptive field properties of individual cells at various
levels of the visual system has provided new insights
into the way in which neurons of the system are interconnected for processing visual information. For
example, many receptive fields of cells in the dorsal
laminae of the lateral geniculate nucleus in the monkey are organized in a concentric center-surround
fashion [37]. Lateral geniculate cells project in turn to
the striate cortex (area 17), where cells are highly
specialized to respond to bars and edges of a particular orientation and often also of a particular length

and moving in a specified direction. Most cells of the
striate cortex also receive input from exactly homologous parts of the two retinas, thus combining information from the two eyes that has up to that point
been kept separate [20]. There is increasing evidence
that individual cell layers within striate cortex are also
specialized, containing cells with particular receptive
field types that send their axons to different locations
[15, 28, 321.
Information about the further processing of visual
information in cortical regions beyond area 17 is now
becoming available. Although areas 18 and 19 [8, 91
in the prestriate cortex both appear to be homogeneous on cytoarchitecconic grounds, modern anatomical techniques combined with single cell recording
have demonstrated that extrastriate visual cortex is
divided into many regions in a wide range of mammals from mouse [35]to monkey [2, 3 , 4 1 , 4 2 ] , and
that each region contains a partial or complete representation of the contralateral visual hemifield. The

From the DeDartments of Neurology and Neurological Surgery

Accepted for publication July 20, 1978.

0364-5134/79/030253-09$01.25 @ 1978 by Alan L. Pearlman 2 5 3

full functional significance of these several different
representations of the visual field remains unclear,
but one rather attractive possibility is that each region is specialized to process particular aspects of the
visual array, either serially or in parallel with other
regions. Thus the extrastriate region immediately
adjacent to area 17, called the second visual area
(V-2), may be processing information necessary for
stereopsis, in that many V - 2 cells in the rhesus monkey respond only when the appropriately oriented
visual stimulus falls on slightly disparate parts of the
two retinas [ 2 I ] . Another extrastriate visual region
lies in the posterior bank of the superior temporal
sulcus in the rhesus monkey; it appears to be highly
specialized for the analysis of movement C42, 431.
Still another region, the V - 4 complex [39, 40, 41,
441, is distinguishable by its incoming connections
and is strikingly specialized for the analysis of color.
Compared to other visual regions, where color-coded
cells are rare, there is a marked preponderance of
cells in the V - 4 complex that are activated preferentially by a particular part of the visible spectrum.
Thus, while some contour information is maintained,
color information is more prominently emphasized in
the V 4 complex than in any other cortical region
thus far described, suggesting an important and
perhaps specific role for this region in color vision.
The presence of regions of cortex that are highly
specialized for processing specific aspects of vision
would suggest that a lesion within one such region
might produce a specific visual deficit while sparing
other aspects of vision [24].The case described here
lends support to this concept, in that two cerebral
infarctions led to lesions that severely disturbed color
vision but left many other visual functions intact. The
literature pertaining to the very rare disorders of
color vision resulting from cerebral lesions has recently been reviewed [29];the present paper reports
the findings of a more detailed study of a patient who
was described briefly in the appendix of that review.
Case Report
The patient, a right-handed Caucasian male, was 44 years
old when he became ill in 1966. He had been employed for
many years as a customs inspector and had passed a color
vision (lantern) test with no difficulty to obtain that job.
There was no family history of defective color vision. In
October, 1966, the patient noted sudden onset of visual
symptoms that were nonprogressive but that he could not
clearly characterize. However, these symptoms definitely
did not include the problems with color vision, face recognition, or topographical disorientation that were to become
prominent at a later stage. The patient was first seen at the
National Hospital for Nervous Diseases, Queen Square,
several weeks later, where he was found to have a right
superior homonymous quadrantanopia. Memory was moderately disturbed in that he could not correctly name the

254 Annals of Neurology Vol 5 No 3 March 1979

month, had difficulty with current events, and could not
learn an unfamiliar sentence. Cranial nerves, motor and
sensory functions, and language were not disturbed.
The Wechsler Adult Intelligence Scale showed a verbal IQ
of 8 4 and performance IQ of 99. The findings on laboratory examination included normal hemoglobin, white cell
count, sedimentation rate, skull and chest x-rays, and electroencephalogram. Bilateral vertebral angiogams and a left
carotid angiogram were carried out under general anesthesia; all were normal.
Immediately after the angiographic procedures the patient complained that his vision was worse, in particular
that he was unable to see colors, stating later that it was
“like seeing a black and white movie.” Examination showed
that he had developed a left superior quadrantanopia in
addition to the right superior quadrant defect initially present, to produce a superior altitudinal field defect. In addition, he was unable to recognize faces, had topographical
disorientation, and had severely impaired color vision, all
of which improved slightly over the next few weeks but
have since remained stable.
The patient was reexamined six years later, in 1972, at
age 50 1291. He had returned to a simple job that did not
require color discriminations, and he was able to find his
own way to work by carefully following a route along the
Thames. Otherwise he was unable to find his way about
without assistance and could not recognize his wife’s
face when she was dressed in unfamiliar clothing. Neurological examination showed a congruent superior altitudinal field defect (see Fig 3 in [29]), mild memory impairment, and very mild difficulty in naming everyday objects.
Visual acuity was 6/9 (20/30)in each eye with correction,
and he could read N4.5 (approximately equivalent to
Jl). The results of color vision testing (Ishihara and
Farnsworth-Munsell) were consistent with those obtained
in the more extensive reexamination to be reported here.
The patient was tested for this report in 1974, at age 52.
His general health had remained good, and the symptoms
referable to his cerebral infarcts in 1966 were unchanged.
Computed tomography in approximately t h e horizontal
plane demonstrated extensive lesions in the inferior occipital iobes bilaterally (Fig lA,B). Comparison of the C T scan
with horizontal sections of the human brain at comparable
levels (Fig 1) provides an indication of the gyri involved in
the lesions. O n the left there appears to be involvement of
the lingual gyrus, the lateral occipital gyri, and the posterior
aspect of the medial occipitotemporal gyrus (Fig 1A). The
lesion in the lingual gyrus on the left is also evident on the
higher cut shown in Figure IB, but it is relatively small at
this level. The cortex above the calcarine fissure (cuneus)
appears to be spared (Fig lB,C). A large occipital horn of
the left lateral ventricle is also evident (Fig 1B). The area of
apparent decreased density posteriorly on the left in Figure
1C is a linear artifact presumably due to motion of the head
with respiration. The lesion on the right is more extensive;
it appears to involve the lingual gyrus, the medial occipitotemporal gyms. the medial part of the optic radiations, and the caudal end of the parahippocampal gyrus (Fig
IA). The lesion in the lingual gyrus on the right is also
evident in Figure l B , but the cuneus has apparently been
spared (Fig lB,C). I t should be noted that a number of

F i g 1 . Top row: CT .rrun.r of the patient. taken in approximately the horizontal plane, progressing from the deepest rrrt (A)
to the most superficial (C).
Bottom row: Oritline drawings of
horizontal sections of the human brain at nearly the .ram levelr
as those shown in the top row. (Drawings were traced from
Hanaway, Scott, andstrother [ 1 8 ] , p p1.3. 15, und 19.)

structures near t h e lesions appear normal bilaterally o n the
C T scan; these include the lateral geniculate nucleus (Fig
l A ) , the cuneus (Fig lA), the splenium of the corpus
callosum (Fig l C ) , the cingulate gyrus (Fig l C ) , and the
precuneus (Fig 1C). The lesions correspond well to the
vascular territory of the lateral branch of the posterior CKrebral artery on both sides 1-34],The territory of the medial
branch of the posterior cerebral artery, which includes the
cuneus, has a normal appearance on the C T scan, accounting for the preservation of the inferior visual field.
We attempted an approximation of the cytoarchitectonic
regions involved in these lesions by superimposing horizontal lines that correspond to the planes of section shown
in the CT scan on Brodmann's summary diagram of cytoarchitectonic regions [9]. This procedure indicates that the
inferior aspect of the primary visual cortex (area 17) is involved bilaterally, as are the inferior portions of the extrastriate visual cortex (areas 18 and 19) on the medial
aspect of the right hemisphere. Area 18 is similarly involved o n the left, but it is difficult to tell whether or not
the lesion extends far enough anteriorly o n the left to in-

volve area 19. O n the right the lesion also involves area 35
in the caudal aspect of the parahippocampal gyrus.
Tests of visual function were carried out in the Department of Anatomy and Embryology, University College,
London, and in the Department of Optometry and Visual
Science, The City University, London. The test procedures
were explained to the patient and his informed consent was
obtained. Several additional tests are described in a separate report 171.

Results
lnformai Obsewations
A striking feature of patients with acquired cerebral
lesions leading to defective color vision is that most
complain spontaneously of the difficulty, many likening the problem to watching a black and white
movie o r television [ 2 9 ] . Our patient made a similar
statement; his other observations about his difficulty
were also quite informative. H e reported that the
color vision defect interfered considerably with several aspects of his life in which discriminations must
be made based on color rather than on other visual
features such as texture, shape, o r size. His wife
helped him to select his clothes each day. As an
amateur painter, he also relied on his wife to select
colors for him. He could not tell a green apple from a
red one, or an unripe tomato or banana from a ripe
one. Coins of similar size that differ in color had to be

Pearlman et al: Cerebral Color Blindness

255

examined carefully for differences in milling or inscription. His job as a customs officer was to inspect
the seals on containers in the holds of cargo ships, a
task that did not require color vision. His job capabilities were apparently limited more by his topographical disorientation and mild memory dysfunction than by his visual deficit.
The patient had no difficulty in correctly naming
the color of common objects either when asked directly (i.e., what color is a London bus?) or when
shown a black and white photograph or outline
drawing of a familiar object. But when he was asked
to color a page containing the outline drawings of
familiar fruits and vegetables, the patient’s profound
deficit in distinguishing colors immediately became
apparent (Fig 2). A page was selected from a child’s
coloring book of the sort that could easily be completed correctly by a nonartistic 6-year-old in less
than 5 minutes. The patient was given a set of eight
brightly colored felt-tip markers (red, orange, yellow,
green, blue, purple, brown, and black). During nearly
30 minutes of laborious efforts that included repeated comparisons of one pen with another, he selected the same brown pen at different times to color
carrots, mushrooms, and a tomato. The tomato stem
was colored blue, while the carrot tops and a head of
lettuce were colored blue and green (Fig 2). When he
had finished, the patient was asked to name the normal color of the various objects and stated unequivocally that the tomato should be red, the carrots
orange, and so forth, although he was much less certain that h e had accomplished the task correctly.
A further estimate of the patient’s deficit was obtained by asking him to identify colors when they
were presented free from any relationship to familiar
objects. A monochromatic, deeply saturated red (645
or 670 nm) rectangle projected on a white background was called “washed-out red” or “very pale
pink.” A wide range of monochromatic violet, blue,
and green rectangles (435 to 540 nm) were described
as “very pale blue.” Colors between these two ranges
(560 to 600 nm) were variously called “very light
beige” or “light orangey” or “no color at all, really.”
Less saturated monochromatic rectangles were often
described as “very pale and washed out” or “murky
white.” Thus the patient’s visual spectrum is roughly
divided into two regions that he describes with words
that signify desaturated reds or blues. These two regions appear to be separated by a portion of the
spectrum that is nearly achromatic.
Standardized Testing

It is apparent from these relatively simple observations that the patient’s deficit is not an inability to
match a seen color with its spoken name [14]or to
use color names correctly in purely verbal tasks [23].

256 Annals of Neurology Vol 5 No 3 March 1979

F i g 2 . The patient’s attempt to color a page from a childs coloring book with thejilt-tip pens shown at the top. He t w k approximately 30 minutes t o complete the task.

His visual world contains very little color; as a result
he has profound difficulty in telling one color from
another. We therefore undertook a series of standardized color vision tests in order to describe and
document this deficit in more detail.

PSEUDOISOCHROMATIC PLATES. The standard tests
of color vision that are most readily available in a
clinical setting are the pseudoisochromatic plate tests
that have been designed to screen for and diagnose
the various types of hereditary deficiency of retinal
cone pigments that lead to defective color vision. The
tests are based on the observation that it is possible to
prepare colored surfaces reflecting a mixture of
wavelengths that appear different in color to a normal
trichromatic observer, but look to be identical to a
dichromatic individual with a particular defect in
retinal cone photopigment. Thus the plates are called
pseudoisochromatic, since they are isochromatic to

those with defective color vision but not to normal
subjects.
Our patient did remarkably well with these tests.
Although his answers were sometimes slow and
hesitant, he made only two minor errors on the
Ishihara test (1968 edition) and failed only two
low-threshold, red-green screening plates in the A 0
H-R-R test (American Optical Company, Hardy,
Rand, and Rittler, second edition). He also failed
the low-threshold red-green screening plates of the
TMC test (Tokyo Medical College, first edition) but
did well on the remainder of the series. In sum, the
patient made only a few errors on the pseudoisochromatic plate tests, and these were generally on the
low-threshold screening plates. These results are
nondiagnostic and thus are helpful only in a negative
sense. The presumed basis for the patient’s ability to
visualize the figures on pseudoisochromatic plates,
and the implications of these results for clinical testing, are considered in the Discussion.
100-HUE TEST. Another
test that is often available in a clinical setting, the
Farnsworth-Munsell test [ 131 (Munsell Color Company, Baltimore, MD), consists of 85 (originally 100)
discs of different hues but of equal saturation and
brightness. The 85 discs are contained in four elongated trays, each containing about 21 discs, to comprise the four series of the test: one series from red to
yellow, the second from yellow to blue-green, the
third from blue-green to blue, and the fourth from
blue to purple-red. Each tray contains a fixed colored
disc at each end that is the same as the first or last disc
in the neighboring series. The subject’s task is to
arrange the movable discs in each tray in a continuous
color series. Each of the movable discs is numbered
on the back; the error score for a given disc is the
sum of the differences between the number of that
disc and the numbers of the chosen adjacent discs
[25]. A score of 2 is thus the minimum for any disc
and so does not constitute an error. The scores for all
discs are plotted on a polar graph in which each axis
line represents a particular colored disc and the distance from t h e center represents the error score for
that disc (Fig 3).
O n this test our patient showed very poor hue discrimination throughout the entire range (Fig 3), as he
did when tested earlier [29]. The deficit was not limited to any particular part of the spectrum, as it is in
most cases of genetic color vision deficiency. Indeed, the massive defect apparent on this test might
well have looked a great deal worse if the subject had
been given discs from the whole spectrum to order at
the same time, rather than from a limited range.
When given a group of 16 more saturated discs
( H - 1 6 test) [31] from the entire spectrum, he placed

FARNSWORTH-MUNSELL

F i g 3 . The Farnsworth-Munsell 100-hue test. Each filled circle
indicates the patient’s score on a giwn disc (the sum of the diyferences between the number of that diic and the numbers of the
chosen adjacent discs). The broken line is an example of an arerage, normalsubject, taken from the test manual for comparison.
The numbered axis lines correspond to the disc numbers, and the
letters indicate the color of the discs in that part of the range. For
clarity, only every tenth disc number and axis line are shozun.

them into three approximately equal groups that he
called pale pink, pale blue, and darker blue, and he
was unable to place them in any order within a group.
NAGEL ANOMALOSCOPE. The anomaloscope is another test for distinguishing patients with different
types of congenital color deficiency from normal
subjects and from each other. The Nagel anomaloscope presents a monochromatic yellow field that the
subject is asked to match in appearance with a mixture of monochromatic red and green lights. A normal trichromat will accept only one mixture of red
and green as a match for the yellow field; dichromats
can match any mixture of the red and green with the
yellow simply by varying the luminosity of the yellow
field or of the mixture [30]. Our patient successfully
matched a wide range of settings of the red-green
primaries to the monochromatic yellow. The average
fraction of red in the red-green mixture (0.45 5 0.36;
normal value, 0.55) and the average intensity of the
monochromatic yellow needed to match (0.17 +0.07; normal value, 0.17) at first glance merely suggest normal matching end-points with abnormally
wide matching ranges. Closer examination of the
pattern of these matches (Fig 4) demonstrates that
any given red-green mixture was matched with only

Pearlman et al: Cerebral Color Blindness

257

ROD
MoNOCH,ROMAT

Patient’s Hue Discrimination as
Tested by Just-Noticeable
Difference Method

Test

i i i ’

>

o

I
.1

I

.4

I

.6

.B

10

REOlGREEN MIXTURE SCALE

F i g 4 . Matches on the Nagelanomaloscope made by the patient
(filled circles) compared w i t h those made by congenital dichromats and a rod monochromat (curvesfor the dichromats and
the monochromat are from Grntzner [ I 71). The ordinate indicates the normalized intensity of the monochromatic yellow light
that the sirbject adjusted to match the red-green mixture set by
the experimenter; the abscissa gives the normalized fraction of
red in the red-green mixture [ 4 ] .

one intensity of the monochromatic yellow except at
the extreme red end of the scale, where the subject
saw a color difference and was unable to make a
match. The plot of our patient’s anomaloscope results
(Fig 4) indicates that on this test his matching behavior is indistinguishable from that of an extreme
anomalous trichromat. On the other hand, the apparent absence of a blue mechanism and the presence of
relatively normal red and green mechanisms, as measured in this patient by the Wald technique [7], has
suggested a diagnosis of acquired tritanopia. However, as Birch-Cox pointed out [7], this diagnosis
is not consistent with all the observations. The term
deuteranomalous tritanopia has been used to describe
somewhat similar test results (albeit with narrower
anomaloscope matching ranges) in 2 subjects with
congenital color vision defects [ 10, 361; it may also be
used to describe the type of color loss acquired in
certain diseases, such as diabetic retinopathy, that
damage large areas of the peripheral retina. Thus
several different diagnostic terms are partially applicable to this patient, although none is completely
compatible with the test results. At our present level
of understanding of color vision defects resulting
from central nervous system lesions, it is probably
best to avoid the terms that have previously been
devised to characterize the congenital color vision
deficiencies.
A quantitative assessment
of the patient’s ability to discriminate hue was obtained by a method that employs just-noticeable differences (JNDs) for hue. In this test the subject
views a bipartite circular test field of 2.5 degrees,
with each half-field illuminated by monochromatic
light. One half-field remains constant while the examiner gradually changes the hue of the other half-

A(nm)

JNDUp

JNDDown

Mean AA

5 30
580
600
615
630

570
604
608
680
675 t

460 4
5 69
590
602
605

>55
17
9
39
>3 5

A = Wavelength; JND = just-noticeable difference for hue.

field until the subject reports that the two fields are
different. The subject then attempts to match the two
fields by adjusting only the brightness of the test
field. If the subject is unable to make a match by
adjusting only the brightness, the difference i n wavelength of the two fields is taken as the JND for hue
[6, 381.
Our subject was presented with five different
wavelengths (A in the Table); for each of these, the
wavelength of the other half-field was gradually varied toward longer wavelengths UND up in the Table)
or shorter wavelengths UND down in the Table)
until the subject reported that the color was different
and that he could not match the two half-fields by a
change in brightness. The mean JND (mean Ah in the
Table) for each test wavelength is then taken as
(Ahup
AXdown)/2. The arrows indicate that for
test wavelengths of 530 and 630 nm, the subject
reached the limits of the test apparatus (460 and 675
nm) without detecting a difference in hue. Our subject’s hue discrimination curve, taken from the Table,
is plotted in Figure 5, along with that of normal subjects for comparison [38]. The points representing
mean Ah for 580 and 600 nm are shown, and the
curve is drawn to indicate the positions of points that
are off-scale. It is evident from the Table and Figure
5 that our patient’s hue discrimination is very poor
throughout the spectrum, although somewhat less so
in the range between 580 and 620 nm.

+

Tests of Other V i s u a l Functions
In contrast to the patient’s profound difficulty with

color vision, several other tests of visual function
were surprisingly normal.

HUE DISCRIMINATION.

258

Annals of Neurology Vol 5 No 3 March 1979

ACUITY AND READING. The patient’s visual acuity
was 6/9 (20/30) in each eye with correction, and he
could read fine print with no difficulty (OD N6, 0s

N5).
EYE MOVEMENTS. There was a full range of eye
movement when the patient was tested either

20

L

’I

I

16

Fig 5 . Hue discrimination. The broken line shows the patient’s
hue di.tcriminationcurve; the solid line is the averagefor sweral normal subjects as given b~ Wright and Pitt L381. (See text
for details.)

monocularly or binocularly, and the cover test
showed no ocular muscle imbalance. N o difficulty
was apparent in the patient’s ability to fixate and follow a smoothly moving target, nor in his ability to
move his eyes quickly from one visual target to another. Optokinetic nystagmus was present and symmetrical when a striped tape was moved to the patient’s right or to his left.
Three tests were used to
assess the patient’s binocular single vision and
stereopsis. The first, the synoptophore (Clement
Clarke, London), is an optical device that allows the
subject to view separate two-dimensional slides with
each eye. O n this test the patient was able to fuse two
nondisparate images normally when they were presented at optical infinity and to maintain fusion
through 10 degrees of convergence and 5 degrees of
divergence, a normal range for his age. H e was also
able to judge correctly the relative “depth” of several
objects on a pair of slides when the objects had varying retinal disparities, indicating normal stereopsis.
Normal stereopsis was also evident on a widely used
test that presents disparate two-dimensional images
to each eye by means of Polaroid vectograph cards
(Titmus Optical Company, Petersburgh, VA). Tests
for gross stereopsis (grasping a housefly’s wings that
appear above the page and selecting the forwardappearing animal in a series) were normal. The quantitative test series indicated a stereo acuity threshold
of 50 seconds of arc, a value at the high end of the
normal range.
Finally, the patient was tested with a series of
random-dot stereograms [22]. When these targets are
viewed monocularly, each appears as a random array
of small light and dark squares in which no form or
depth is apparent. The two random-dot patterns are
identical except that in one pattern the dots in the
central region have been displaced laterally with reFUSION AND STEREOPSIS.

spect to the same region in the other pattern. There
are no monocular cues to depth in these patterns;
with binocular fusion a central figure stands out vividly in front of or behind the surround. The central
figure is apparent only when viewed stereoscopically,
since information about the central figure is contained only in the relationship between the two sets
of random dots. The patterns are printed in pale red
and green inks and are viewed with a green filter over
one eye and a red filter over the other so that each
eye sees only one pattern, as in the synoptophore and
Polaroid vectograph tests. The patient viewed several
random-dot stereograms (from Julesz [22]: Figs
1.0-1, 2.4-1, 8.1-2a and b, 8.3-3, and 8.3-4). H e was
able to describe the central figure and saw it as
standing away from the surround, again indicating
normal stereopsis.

Discussion
In a recent review of disturbed color perception resulting from acquired cerebral lesions, Meadows [29]
drew attention to the distinction between syndromes
that might be considered perceptual and those in
which perception of color is intact but the ability to
use the names of colors is markedly impaired [14,
231. The former syndromes have been called cerebral
achromatopias or cerebral color blindness, and the latter
color anomias or color agnosias. The distinctions are
rather straightforward; the patient we have studied
exemplifies the features of cerebral color blindness.
He suddenly found himself in a world containing very
little color, and complained spontaneously of the
change. H e did poorly on tests that require matches
between similar colors or distinctions between different colors, but he had no difficulty in giving the
correct color names associated with common objects.
Patients with color anomias or agnosias, on the other
hand, can perform color matches and distinguish
different colors, but they cannot associate the correct
color name with colors they can obviously see.
The testing that has been carried out on previously
described patients with cerebral color blindness has
been limited [ l , 16, 291. It is important to stress that
the pseudoisochromatic plate tests that are commonly used in a clinical setting for testing color vision
are not adequate for detecting or defining this syndrome; these tests were designed to detect subjects
with a deficiency in a particular retinal cone pigment.
Our subject with cerebral color blindness was able to
recognize the figures on these plates, probably because his hue discrimination was poor but not absent, whereas patients with more severe syndromes
are unable to do so [16, 291. The figures on the
pseudoisochromatic plates, which contain rather
large color differences [25], were evident to our patient, whereas the distinctions between discs of the

Pearlman et al: Cerebral Color Blindness

259

Farnsworth-Munsell 100-hue test, in which the color
differences between discs are very small [25],presented much greater difficulty. It is hue discrimination that is impaired in this syndrome, and tests for it
must provide more subtle distinctions than are present on pseudoisochromatic plates. The FarnsworthMunsell 100-hue test is a suitable modality that
is widely available and easily administered. Assessing hue discrimination by JNDs for monochromatic stimuli gives a quantitative assessment of the
deficit but requires rather specidzed apparatus. Perhaps as more is learned about the function of cortical
areas in color vision, more specific tests will be devised to study them.
Although this case of cerebral color blindness represents a very unusual clinical syndrome, its implications for our understanding of the function of the
visual system are of considerable interest. There is
certainly a great deal of evidence that the processing
of visual information takes place in a serial, or hierarchical, fashion at several stages in the visual system,
including the retina [12] and striate cortex [20]. O n
the other hand, the presence of highly specialized
segments of extrastriate visual cortex that are definable both anatomically and physiologically [411
suggests the prospect that certain aspects of visual
processing, such as color vision, may take place in
parallel with other aspects of processing. O n e consequence of an arrangement that includes some degree
of parallel processing of different aspects of vision
is that a lesion involving a region that is highly
specialized for a particular function ought to produce
a deficit in that particular visual function and not in
others. The patient we have studied as well as other
cases of a similar sort [16, 291 lend support to this
concept. As we have shown, many of our patient’s
visual abilities, including visual acuity, stereopsis, and
visually guided eye movement, have remained intact
despite lesions that caused a severe defect in hue
discrimination. Thus it would seem that analysis of
color takes place, at least in part, in a distinct cortical
region or regions where lesions can occur without
interfering in a major way with many other visual
functions.
The question then arises as to where this “color
region” or regions might be located in the human
cerebral cortex. Two lines of evidence are available on
this question, the first from experimental studies of
the rhesus monkey and the second from clinical and
pathological observations in humans. T h e V - 4 complex in the rhesus monkey prestriate cortex contains
a large proportion of cells whose responses are
color-specific. The V - 4 complex is made up of at least
two subregions, and possibly more: one subregion
lies in the anterior bank of the lunate sulcus [39,40],

260

Annals of Neurology

Vol 5

No 3 March 1979

the other in the lateral aspect of the posterior bank of
the superior temporal sulcus [44].Although the response characteristics of cells in both subregions are
quite similar, the subregions are distinguishable on
the basis of their afferent connections [39, 441.
Neuroanatomical definition of the V-4 complex and
its subregions has required the use of degeneration
techniques and radioactive amino acid tracer methods [41, 441, since none are distinguishable on
cytoarchitectonic grounds. The available evidence
suggests that the requisite lesion in man ought to be
found in extrastriate visual cortex (areas 18 and 19 of
Brodmann [8 , 91). However, detailed analysis of visual areas in the monkey has had a relatively short
history, so that regions other than the V - 4 complex,
either within o r beyond areas 18 and 19, are certainly
not excluded.
The 3 autopsied patients with cerebral color blindness [19, 26, 271 and the C T scan in our patient
provide evidence that a region important in the analysis of color lies in the anteroinferior aspect of the
occipital lobe in man, within the medial occipitotemporal (fusiform) o r lingual gyri or both. According to
Brodmann’s classification [ 9 ] , the human cytoarchitectonic area 18 surrounds striate cortex (area 17)
as it does in the monkey; a portion of area 18 is
within the lingual gryus, and a segment of area 19 is
in the medial occipitotemporal gyrus. Thus there is
evidence from human material that the region in
question may be within areas 18 and 19. However,
the evidence linking the regions of monkey prestriate
cortex that are specialized for color processing with
homologous regions in humans cannot at this time be
taken to be anything more than suggestive. The reasons for caution are several: First, the evidence in
man comes from a description of the gyri involved by
lesions, not from detailed cytoarchitectonics. The
gyral pattern of the cortex is highly variable and does
not correspond at all well to cytoarchitectonic borders [ 5 , 111. Second, although the cytoarchitectonic
criteria that define area 17 and set it apart from area
18 are very distinct, those that define the border
between areas 18 and 19 are very much less so and
are therefore a matter of some controversy [33].Finally, delineation of the subregions of extrastriate visual cortex in the monkey has required experimental
techniques that cannot be applied to human material.
The important point to be made at this stage is that a
region of extrastriate visual cortex containing neurons that are highly specialized for distinguishing one
color from another has been identified in the monkey, and that a syndrome of severely impaired color
vision occurs in man to indicate that this or a similar
region involved in hue discrimination has been damaged.

_____

-

Supported by a grant from the Science Research Council, UK, to
Dr Semir Zeki, by a grant from the US National Eye Institute (EY
00621). and by a Royal Society of Medicine Foundation Travelling
Fellowship Award to Dr Pearlman.
We are grateful to Dr Semir Zeki, University College, London,
who generously made his laboratory available and facilitated the
study in many other ways; to Dr Joseph Hanaway of the Department of Neurology and Dr Mokhtar Gad0 of the Department of
Radiology, Washington University School of Medicine, who provided valuable aid in interpreting the CT scan; and to Drs Mathew
Alpern and Nigel Daw, who made many helpful comments on the
manuscript. We thank Janice Jones, who typed the manuscript, and
Sue McConnell, who made the figures.

References
1. Albert ML, Reches A, Silverberg R: Hemianopic colour
blindness. J Neurol Neurosurg Psychiatry 38:546-549,
1975
2. Allman JM, Kaas JH: The organization of the second visual
area (VII) in the owl monkey: a second order transformation
of the visual hemifield. Brain Res 76:247-265, 1974
3. Allman JM, Kaas JH: A crescent-shaped cortical visual area
surrounding the middle temporal area (MT) in the owl monkey (Aotus trivirgutus). Brain Res 81:199-213. 1974
4. Alpern M: What is it that confines in a world without color?
Invest Ophthalmol 13:648-674, 1974
5. Bailey P, von Bonin G: The Isocortex of Man. Urbana, IL,
University of Illinois Press, 1951
6. Birch J. Wright WD:Colour discrimination. Phys Med Biol
6:3-24, 1961
7. Birch-Cox J: A case of acquired tritanopia. Mod Probl
Ophthalmol 171325-330, 1976
8. Brodmann K: Beitrage zur histologischen Localisation der
Grosshirnrinde: 111. Die Rinderfelder der niederen Affen. J
Psycho1 Neurol (Leipzig) 4:177-226, 1905
9. Brodmann K Vergleichende Lokalisationslehre der Grosshirnrinde. Leipzig, Barth, 1909
10. Cole BL, Henry G H , Nathan J: Phenotypical variations of
uitanopia. Vision Res 6:301-313, 1966
11. Connolly CJ: External Morphology of the Primate Brain.
Springfield, IL, Thomas, 1950
12. Dowling JE: Organization of vertebrate retinas. Invest
Ophthalmol 9655-680, 1970
13. Farnsworth D: The Farnsworth-Munsell 100-hue and
dichotomous test for color vision. J Opt SOCAm 33:568-578,
1943
14. Geschwind N, Fusillo M: Color-naming defects i n association
with alexia. Arch Neurol 15:137-146, 1965
15. Gilbert CD, Kelly JP: The projections of cells in different
layers of the cat’s visual cortex. J Comp Neurol 163:Xl-106,
1975
16. Green GJ. Lessell S: Acquired cerebral dyschromatopsia.
Arch Ophthalmol 95:121-128, 1977
17. Grutzner P Acquired color vision defects, in Jameson D,
Hurvich LM (eds): Handbook of Sensory Physiology, Vol
VII/4, Visual Psychophysics. Berlin, Springer, 1972
18. Hanaway J, Scott WR, Strother CM: Atlas of the Human
Brain and the Orbit for Computed Tomography. St Louis,
Warren H. Green, 1977
19. Heidenhain A: Beitrag zur Kenntnis der Seelenblindheit.
Monatsschr Psychiatr Neurol 65:6I-I 16, 1927

20. Hubel D H , Wiesel TN: Receptive fields and architecture of
monkey striate cortex. J Physiol (Lond) 195:215-243, 1968
21. Hubel D H , Wiesel T N : Cells sensitive to binocular depth in
area 18 of t h c macaque monkey cortex. Nature 225:41-42,
1970
22. Julesz B: Foundations of Cyclopean Perception. Chicago,
University of Chicago Press, 1971
23. Kinsbourne M, Warrington EK: Observations on colour agnosia. J Neurol Neurosurg Psychiatry 27:296-299, 1964
24. Kleist K: Sensory Aphasia and Amusia: the Myeloarchitectonic Basis. Translated by FJ Fish and JB Stanton. London,
Oxford University Press, 1969
25. Lakowski R Theory and practice of colour vision testing: a
review. Part 2. Br J Ind Med 26:265-288, 1969
26. Len2 G: Zwei Sectionsfalle doppelseitiger zentrale Farbenhemianopsie. Zbl Ges Neurol Psychiatr 71:135-186,
1921
27. MacKay G, Dunlop JC: The cerebral lesions in a case of complete acquired colour blindness. Scot Med Surg J 5:503-512,
1899
28. Mangini N, Pearlman AL: Laminar organization of mouse
primary visual cortex. Assoc Res Vis Ophthalmol Abstr,
1977, P 138
29. Meadows JC: Disturbed perception of colours associated with
localized cerebral lesions. Brain 97:615-632, 1974
30. Padgham CA, Saunders JE: The Perception of Light and Colour. New York, Academic, 1975
31. Paulson HM: Color vision testing in the United States Navy.
Visual Science. Bloomington, IN, Indiana University Press,
1968, p 164
32. Schiller PH, Finlay BL, Volman SF: Quantitative studies of
single cell properties of monkey striate cortex. J
Neurophysiol 29:1115-1156, 1976
33. von Bonin G, Bailey P: The Neocortex of Mucuru mrduttu.
Urbana, IL, University of Illinois Press, 1947
34. Waddington MM: Atlas of Cerebral Angiography with Anatomic Correlation. Boston, Little, Brown, 1974
35. Wagor E, Mangini N, Pearlman A L A retinotopic map of
mouse extrastriate visual cortex. Neurosci Abstr 3:580,
1077
36. Wald G: Defective color vision and its inheritance. Proc Natl
Acad Sci USA 55:1347-1363, 1966
37. Wiesel T N , Hubel DH: Spatial and chromatic interactions in
the lateral geniculate body of the rhesus monkey. J
Neurophysiol 29:1115-1156, 1966
38. Wright WD, Pitt FHG: Hue discrimination in normal
colour-vision. Proc Phys SOC(Lond) 46469-473, 1934
39. Zeki S M Cortical projections from two prestriate areas in the
monkey. Brain Res 34:19-35, 1971
40. Zeki SM: Colour coding in rhesus monkey prestriate cortex.
Brain Res 53:422-427, 1973
41. Zeki SM: The mosaic organization of the visual cortex in the
monkey, in Bellairs R, Gray EG (eds): Essays on the Nervous System. Oxford, England, Clarendon, 1974. chap 13, pp
327-343
42. Zeki SM: Functional organization of a visual area in the posterior bank of the superior temporal sulcus of the rhesus monkey. J Physiol (Lond) 236:549-573, 1974
43. Zeki SM: Cells responding to changing image size and disparity in the cortex of the rhesus monkey. J Physiol (Lond)
242:827-841, 1974
44. Zeki SM: Colour coding in the superior temporal sulcus of
rhesus monkey visual cortex. Proc R SOC Lond [Biol]
197:195-223, 1977

Pearlman er al: Cerebral Color Blindness

261