63ournal ofNeurology, Neurosurgery, and Psychiatry 1996;61:638-643

Behavioural and electrophysiological chromaticand achromatic contrast sensitivity in an

achromatopsic patient

Charles A Heywood, Julian J Nicholas, Alan Cowey

AbstractObjectives-In cases ofincomplete achro-matopsia it is unclear whether residualvisual function is mediated by intact stri-ate cortex or results from incompletelesions to extrastriate cortical visualareas. A patient with complete cerebralachromatopsia was tested to establish thenature of his residual vision and to deter-mine the integrity of striate cortex func-tion.Methods-Behavioural contrast sensitiv-ity, using the method of adjustment, andaveraged visually evoked cortical poten-tials were measured to sinusoidally mod-ulated chromatic and achromaticgratings in an achromatopsic patient anda normal observer. Eye movements weremeasured in the patient using a Skalarinfrared monitoring system.Results-The patient's chromatic con-trast sensitivity was normal, indicatingthat despite his dense colour blindness hisoccipital cortex still processed informa-tion about spatial variations in hue. Hissensitivity to achromatic gratings wasdepressed particularly at high spatial fre-quencies, possiby because of his jerk nys-tagmus. These behavioural results werereinforced by the nature of visuallyevoked responses to chromatic andachromatic gratings, in which total colourblindness coexisted with an almost nor-mal cortical potential to isoluminantchromatic gratings.Conclusions-The results show that infor-mation about chromatic contrast is pre-sent in some cortical areas, and coded ina colour-opponent fashion, in the absenceofany perceptual experience of colour.

Department ofExperimentalPsychology, Universityof Oxford, South ParksRoad, OxfordOXI 3UD, UKJ J NicholasA CoweyDepartment ofPsychology, ScienceLaboratories, SouthRoad, DurhamDH1 3LE, UKC A HeywoodCorrespondence to:Dr C A Heywood,Department of Psychology,University of Durham,Science Laboratories, SouthRoad, Durham, DH1 3LE,UK.

(7 Neurol Neurosurg Psychiatry 1996;61:638-643)

Keywords: achromatopsia; chromatic contrast sensitiv-ity; visual evoked potentials

Cerebral achromatopsia is a severe distur-bance of colour vision caused by ventromedialoccipital brain damage. -3 The impairment isnot, indirectly, of retinal origin as measure-

ment of two colour incremental thresholds4shows intact trichromatic mechanisms5 andthe characteristic shape of the spectral sensitiv-ity function indicates residual colour opponentprocessing.6 At necropsy the brains of achro-matopsic patients invariably show damage inthe region of the lingual and caudal fusiform

gyrus in the ventromedial aspect of the occipi-tal lobes.7 Positron emission tomography stud-ies in normal subjects show an area in thesame region, where there is a selective increasein regional blood flow when subjects passivelyview a coloured Mondrian pattern.8 However,other studies involving functional neuroimag-ing by PET suggest that the processing ofcolour is not restricted to a single extrastriatevisual area.910 Such findings have led to thehypothesis that cerebral achromatopsia resultsfrom damage to one or more extrastriate visualareas which are specialised for the processingof colour.3 '1An alternative hypothesis, which stems from

studies of monkeys,'2 is that achromatopsiacan result from damage anywhere along achannel of processing that originates in thecolour opponent PB retinal ganglion cells thatinnervate anatomically and functionally segre-gated regions of striate cortex (V1) via the par-vocellular layers of the dorsal lateral geniculatenucleus (dLGN). This pathway, called the Pchannel, then becomes elaborated in someextrastriate cortical areas (V2 and V4) whichproject to inferior portions of the temporallobe. This pathway can be distinguished fromits partner, the M channel, which arises fromP retinal ganglion cells that project to magno-cellular layers of the dLGN which, in turn,project to distinct regions of Vl which con-tinue this stream of processing into the parietallobe via cortical areas V2, V3, and MT, as wellas, less intensively, into the temporal lobe.The cortical segregation of the P and M chan-nels has been shown with cytochrome oxidasestaining, when the P channel is associated withthe cytochrome oxidase rich "blobs" and"interblobs" of Vl, and the thin and paleinterstripes of V2. The M channel is primarilyrestricted to layer 4B of Vi and the thickstripes of V2. Response properties of singleneurons along the two pathways are consistentwith the notion of a functional segregationsuch that the P channel is concerned with pro-cessing chromatic information in addition tohigh spatial and low temporal frequencies.Cells on the M channel are especially respon-sive to motion and are more sensitive thanthose of the P channel to low contrast.Sensitivity to the orientation of a visual stimu-lus is not an exclusive property of either path-way and hence it may be supposed thatinformation about form is carried by bothpathways.

If achromatopsia is the result of selectivedisturbance of the P channel before its inner-vation of striate cortex then it should be

638

group.bmj.com on January 15, 2014 - Published by jnnp.bmj.comDownloaded from

Heywood, Nicholas, Cowey

accompanied by impairments of the discrimi-nation of form, texture, and dot position.Neurotoxic lesions confined to the parvocellu-lar layers of the dLGN result is severe disrup-tion of the processing of these, or comparable,attributes.'2 The demonstration that achro-matopsia customarily exists in the absence ofsuch accompanying disorders argues againstthis view.

Achromatopsic patients perform poorly,even randomly, on the Farnsworth-Munsell100 hue test, which requires the ordering ofsome isoluminant coloured chips on the basisof their chromaticity. However, the severity ofthe colour disturbance is variable and manypatients are dyschromatopsic-for example,showing a relatively greater disturbance forblue/greens than for reds' 1314 and some preser-vation of colour naming. The ability to namefigures or trace the figures embedded in theIshihara pseudoisochromatic plates is alsovariable. Some patients fail to identify any ofthe figures'5-'7 whereas others identify many ofthem.' '4 It has been proposed that the severityof cerebral achromatopsia is related to theextent to which the striate cortex, particularlyits chromatic compartments, has been com-promised. '4 The latter authors report a patientin whom poor performance of colour orderingand identification was accompanied by normalcolour contrast sensitivity, intact acuity, aclear visible visual evoked potential to an iso-luminant chromatic chequerboard, and anintact ability to detect colour differences in anoddity task. Moreover, eight out of nineIshihara plates were identified. They con-cluded that residual chromatic abilities weremediated by intact striate cortex. An alterna-tive explanation3 is that the lesion to the cru-cial areas of the lingual and fusiform gyri isincomplete. This interpretation would bestrengthened by examination of the integrity offunction of the striate cortex in a patient withcomplete achromatopsia, who should lack anormal, or even any, visually evoked potentialin response to isoluminant sinuosoidally mod-ulated chromatic gratings. It is the results ofsuch an examination that we report here.

Case historyThis has been published in detail elsewhere6 18 19and is summarised here. The patient is a 44year old man, who was fit and well until hehad idiopathic encephalitis as a 22 year oldpolice cadet. As a result, he had permanentsevere bilateral cerebral damage. He has severeobject agnosia, cannot recognise familar facesand has topographic disorientation and a dis-turbance in semantic memory. Routine physicalexamination showed normal Snellen visualacuity in each eye and a dense left homony-mous hemianopia with macular sparing.Before his illness he had normal colour visionas assessed by the Ishihara plates. Since his ill-ness he is unable to match or name colourspresented in his remaining visual hemifield.His colour ordering on the Farnsworth-Munsell 100 hue test is random, yielding anerror score of 1245.

Brain MRI shows the extent of the damage.6There is extensive ventromedial damage inboth hemispheres which includes the lingualand fusiform gyri. In the right hemisphere thesecond, third, and fourth temporal gyri arecompletely destroyed as is the pole of the tem-poral lobe. There is also damage to theparahippocampal gyrus. There is considerabledamage to the occipital lobe with sparing ofthe caudal tip of the calcarine cortex (thiscould explain the macular sparing in his leftvisual field). In the left hemisphere the pole ofthe temporal lobe, the parahippocampal andthe fourth temporal gyri are totally destroyedas is the area of the mesial occipitotemporaljunction. The dorsal part of both hemispheresshows relatively little damage compared withthe ventral part. In the right hemisphere thewhite matter under the inferior half of the infe-rior parietal lobe is damaged. In the left hemi-sphere there is no evidence of damage either inthe white or the grey matter of the parietallobe.

MethodsCONTRAST SENSITIVITYHorizontal sine wave gratings, subtending50 x 5° of visual angle from a viewing dis-tance of 1 m, were presented in the centre of ahigh resolution (Eizo, 1024 x 1024 pixels,frame rate 100 Hz), display, which subtended220 x 160 of visual angle. Gamma correction,to equate the isoluminant ratio of the red andgreen guns, was carried out in software inwhich luminance was measured with aMinolta luminance meter LS- 110. Iso-luminant chromatic gratings were displayed bymodulating the red and green guns inantiphase with 12 bit precision against a uni-form yellow background of the same meanluminance (33 C/dM2). Achromatic (strictlyspeaking, monochromatic) yellow/black grat-ings were displayed by modulating the guns inphase. Achromatic contrast is defined as thedifference between the maximum and mini-mum luminance of the grating, divided bytheir sum. Chromatic contrast is defined asequal to the luminance contrast of each con-stituent grating, if the two gratings are mono-chromatic and of equal contrast.

Measurements of contrast thresholds weremade using the method of adjustment in thepatient and an age/sex matched control sub-ject who had normal colour vision and aSnellen acuity of 6/6. The procedure was asfollows. The subject sat in a darkened room,to which he was adapted, facing the screenand was instructed to maintain fixation on asmall cross which was permanently displayed.A brief tone indicated the appearance of agrating, presented at 100% contrast immedi-ately to the right of the fixation point. Thisensured that the grating was not presented inthe left hemianopic field of the patient. andtherefore subtended the same visible visualangle in the right hemifield of both subjects.By adjusting a hand held switch the subjectcould decrease the contrast in steps of 0 9times the current contrast, or increase it by a

639

group.bmj.com on January 15, 2014 - Published by jnnp.bmj.comDownloaded from

Behavioural and electrophysiological chromatic and achromatic contrast sensitivity in an achromatopsic patient

factor of 1 1. The subject was asked to makethe adjustment until the grating became invis-ible. Gratings were presented at several spatialfrequencies (sine wave modulated gratingreversal). For each condition of static ordynamic gratings, the spatial frequency of thegrating was randomly selected from the stimu-lus set from trial to trial. At each spatial fre-quency three estimates of threshold weremade such that on the second and third pre-sentations the initial contrast was either twoor three times higher than the threshold deter-mined on the previous presentation-that is,the grating is always visible at the beginning ofthe series. Sensitivity to stationary chromaticgratings of 0-5, 1, 2, 4, 8, and 10 cycles/!was measured. For achromatic measurementsan additional frequency of 16 cycles/! wasused.

VISUAL EVOKED POTENTIALSVisual evoked potentials (VEPs) wererecorded in response to reversing achromaticand chromatic sine wave gratings in thepatient and the control. The gratings sub-tended 50 x 50, were viewed from a distanceof 90 cm, and were produced by a Pluto IIgraphics device connected to a PC and dis-played on a Microvitec monitor (768 x 576pixels, 50 Hz frame rate). The red and greenguns were modulated with 8 bit precision andgamma correction was again carried out insoftware. Chromatic and achromatic gratingswere constructed by modulating the red andgreen guns in antiphase and in phase respec-tively. The mean screen luminance was main-tained at 13-2 C/dM2. To ensure that thegratings were isoluminant to the patient, thepeak luminances of the red and green gunswere selected on the basis of flicker photometryand were shown to be indistinguishable to thepatient in a three choice oddity task6. Evokedpotentials were recorded in response to 6-25Hz (12-5 reversals/s) pattem reversal of grat-ings of 0 5, 1, 2, and 4 cycles/! each at a con-trast of 0-48.

Steady state evoked potentials wererecorded from 9 mm Ag/AgCl cup electrodes.

The source and reference electrodes wereplaced in the midline 5 cm above the inion and12 cm above the nasion respectively, and aforehead electrode acted as the subject's earth.Electrode impedance was less than 5 kW. Thesignal was amplified by 10 000 by a differentialamplifier (Digitimer, NL104) and filteredbelow 0 1 Hz and above 25 Hz (40 dB/decadeattenuation). The filtered signal was digitised(CED, 1401) and averaged over a period ofone second by a PC, sampling at 256 sam-ples/s. A minimum of 128 sweeps contributedto each average. The experimenter could viewboth the running average and the raw data toallow for on line artefact rejection. The ampli-tude of the second harmonic response wasmeasured by Fourier analysis, in which thebandwidth of each spectral component was1-0 Hz.The subjects were instructed to maintain

their fixation along a vertical line at theextreme left of the display. This restricted thestimulus to the right visual field and ensuredthat the normal subject's results were compa-rable with those of the patient, whose leftvisual field is hemianopic.

EYE MOVEMENT RECORDINGSThe patient's eye movements were measuredwith the Skalar infrared monitoring systemand recorded by a computer with a 250 Hzmetrabyte analogue-to-digital converter.Analysis was performed off line. Eye move-ments were recorded in very dim surround-ings. The patient was seated 114 cm from anarray of LEDs arranged in an isovergent plane(both eyes and all LEDs lay on the circumfer-ence of an imaginary circle of radius 57 cm).Both eyes were open. Calibration was per-formed with seven calibration points. Thepatient first fixated a central target for 30 sec-onds, followed by a target at 12° in his rightvisual field. Calibration was repeated with acentral target followed by a target appearing at120 in his left hemianopic field. This was invis-ible but he knew it was the mirror imagearrangement of the first calibration. Other,more eccentric positions were then used.

Figure 1 Contrastsensitivity to achromatic(left) and chromatic(right) gratings for thepatient (triangles) and thecontrol subject (circles).

Achromatic

100

10

1 10Spatial frequency (cycles!0)

100

Chromatic

1 10

Spatial frequency (cycles/!)

100

10

a,rG)

CD40C.)

1

100

640

1

group.bmj.com on January 15, 2014 - Published by jnnp.bmj.comDownloaded from

Heywood, Nicholas, Cowey

Figure 2 Left: steadystate evoked potentials tocontrast reversingisoluminant chromaticgratings for the patient (A)and the control subject(B). Right: The powerspectra of the evokedpotentials displayed on theleft, after a Fouriertransform. The arrowindicates the 2nd harmonicresponse, which wasgreatest in each case.

0.2 1 A

0.1 -

0-0 -

-0.1 -

> -0.2cU 0-

E

-0-1-

0.1-

0.010

0-005-

,-2 0.4 0.6 0.8 1.0Time (s)

0-00

0.02-B

0-01-

_2.

0 0.80-2 0-4 0-6 0-8 1.0 0-00Time (s)

6-25 12-50 018-75Frequency (Hz)

6.25 12-50 018-75Frequency (Hz)

ResultsCONTRAST SENSITIVITYFigure 1 shows the achromatic and chromaticcontrast sensitivity functions of the patient anda normal observer. The spatial frequency cutoff, in which sensitivity is extrapolated to uni-tary contrast, is 16 cycles/! for the patient and40 cycles/! for the normal observer for achro-matic gratings. The threshold value for thepatient corresponds to a Snellen acuity ofbetween 6/9 and 6/12 and is very differentfrom his measured Snellen acuity, using highcontrast letters, of 6/4-5 + 1 (left eye) and6/4-5-1 (right eye). Although the sensitivityfunction is bandpass, there is a reduction insensitivity at all spatial frequencies tested. Inremarkable contrast with the results for achro-matic gratings, the general shape and ampli-tude of the chromatic contrast sensitivityfunction for the patient and the control subjectare similar.

VISUAL EVOKED POTENTIALSFigures 2 and 3 show representative averaged

steady state potentials recorded from the con-

trol and the patient for achromatic and chro-matic gratings of 2 cycles!0 respectively. Inaddition, figures 2 and 3 display the results ofFourier transforms showing the form of powerspectra, clearly indicating the response to begreatest in the second harmonic (12-5 Hz) ineach case. The results for other spatial fre-quencies tested were essentially identical tothose displayed. The visual evoked potentialsrecorded in the patient were similar in shapeto those recorded in the control and althoughthey were generally of decreased amplitude, asecond harmonic response was evident inevery case.

EYE MOVEMENT RECORDINGSFigure 4 shows characteristic eye movementrecords of the patient. They show that a veryfine nystagmus is present. These are in therange of micromovements of the eye that are

normally seen. However, they are super-

imposed on square wave jerks of 0 6° ampli-tude. The fast phase is to the right and the

Figure 3 Left: steadystate evoked potentials tocontrast reversingachromatic gratings for thepatient (A) and the controlsubject (B). Right: Thepower spectra of the evokedpotentials displayed on theleft, after a Fouriertransform. The arrowindicates the 2nd harmonicresponse, which wasgreatest in each case.

0.21 A

0.1-

0.0-

-0.1-

a)

E

0u

0.010 -

0.005 -

0.2 0.4 0.6 0.8 1.0Time (s)

0.21 B

0.1

0.0

-0.1-

-0.2

0.00

n-nln

0.005 -

I . . 1.

0.2 0.4 0.6 0.8 1.0 0.00

Time (s)

v

6.25 12.50 18.75Frequency (Hz)

25.00

V

6.25 12.50 18.75 25.00Frequency (Hz)

25-00

25.00

641

v.v I v1

7

group.bmj.com on January 15, 2014 - Published by jnnp.bmj.comDownloaded from

Behavioural and electrophysiological chromatic and achromatic contrast sensitivity in an achromatopsic patient

Figure 4 (A)representative eyemovements of the patient tothe left (upwards) andright (downwards) whilehe was attempting to fixatea light emitting diodedirectly in front ofhim.The record is taken from amuch longer sequence thatlastedfor one minute. (B)A similar record at higherspatial and temporalresolution to illustrate hisfine nystagmus.

0.8

0, 0-6

c 0.4

°0)-C. 0.2

E) 0-0

DI- 0.20)4) 0.4

Dg' 0.6in 0.8

0)

a.)E

0

0)0)0)

0

maximu

of 20/s.

DiscussThe prepreviouscan beever, th(was dyscolour 1

abilitiesences incompletin an od(ity to iability oiences cathe integfrom theHow

sinuosoi

distinguipresentechromatspatial:threshol1 0 cyclbsignal i1itself direducedtain chrishown bnation.that thecessing.opponenless brig]tive lumieffect,21yellow bof the ki

-A and green. Indeed, when asked to indicatehow he sees bands in the chromatic grating hepointed to the "yellows".Our patient had a reduced achromatic con-

M A--2--l-- n I| s X I l Av AM , trast sensitivity at all spatial frequencies tested,but especially at the higher frequencies, thenhis performance was paradoxically bad whencompared with his normal Snellen acuity. Thesimplest, and likely, explanation is that theretinal image of grating stimuli is smeared by

2 4 6 8 10 his clear jerk nystagmus in which the velocityTime (s) of 20/s is in the range in which sensitivity to

high frequencies would be expected to beB diminished. Thus the smearing will be greater

the finer the grating. Snellen letters contain allspatial frequencies and will therefore be muchless affected. However, it should be noted thatthe jerks are interspersed with periods of rela-tively stable fixation which might be sufficientfor detection of high spatial frequencies. Theorigin of the nystagmus is unknown but itsinfluence could be tested by measuring achro-matic contrast sensitivity to high intensity,

1g01|21|4 16 18ltachistoscopically presented gratings to reduce0 0-2 0-4 0-6 0-8 1-0 1-2 1-4 1-6 1.8 or even eliminate retinal smearing. Our elec-

Time (s) trographic measurements of nystagmus didnot disclose its principal direction, if any.Should there be one, the image of gratings the

im slow phase velocity is of the order orientation of which is parallel to the directionof nystagmus should be much sharper thanthat of gratings oriented orthogonally. Theabsence of any effect of his jerk nystagmus on

ion chromatic contrast sensitivity-at first sight!sent findings confirm and extend a paradoxical-is a simple consequence of the, report that colour contrast sensitivity fact that the latter has too low a spatial fre-unaffected in achromatopsia.'4 How- quency cut off to be disrupted by the fine nys-e patient examined by those authors tagmus.,chromatopsic in the sense that the Our findings show that the patient has aoss was incomplete and his residual normal chromatic contrast sensitivity functionincluded detection of colour differ- and normal chromatic VEPs. He has extensive

l an oddity task. As the patient was bilateral ventromedial damage which includesely unable to detect colour differences the lingual and fusiform gyri. All this suggestsdity task,6yet retained normal sensitiv- that his residual ability to detect chromaticisoluminant chromatic gratings, the gratings does not arise from an incompletefsome patients to detect colour differ- lesion of the lingual and fusiform gryi and isnnot necessarily be attributed solely to mediated instead either by extrastriate visualJrity of striate cortex and is dissociable areas in the dorsal part of the cerebral cortexdetection of chromatic boundaries. which show little damaged compared with thedoes our patient detect isoluminant central part of the cerebral cortex or by intactdal red/green gratings when he cannot striate cortex.ish between the same reds and greens Saito et al22 have shown that there ared as patches on the screen? Although numerous cells in area MT, a prominent part-ic aberration could contribute at high of the socalled broad bandM channel, that arefrequencies any aberration is sub- not silenced at any chromatic luminance ratio.d at low spatial frequencies of 05 and Thus it is possible that the ability of thees/0.20 Furthermore, as any luminance patient to detect the chromatic borderntroduced by chromatic aberration between isoluminant colours6, which presum-iminishes as chromatic contrast is ably underlies his ability to detect chromatic, chromatic aberration could not sus- gratings, depends on intact extrastriate visualomatic contrast at the normal levels areas in the parietal lobe. However, area MTy the patient. There is a better expla- has a negliglible parvocellular input,23 and it isHis spectral sensitivity curve shows difficult to reconcile the completely normalpatient retains colour-opponent pro- chromatic contrast sensitivity function in the

iOne consequence of red-green colour patient (at the temporal and spatial frequen-icy is that a yellow stimulus appears cies tested) with an intact magnocellular chan-ht than red or green of the same objec- nel in the absence of a parvocelluar channelinance. The patient shows this normal for the following reasons. Most magnocellularwhich in normal observers makes the cells do have a null point at isoluminance andands in a red/green sinusoidal grating of the cells that do not, the responses at isolu-ind we used look dimmer than the red minance are less than at any other colour ratio.

642

(

group.bmj.com on January 15, 2014 - Published by jnnp.bmj.comDownloaded from

Heywood, Nicholas, Cowey

This is consistent with the response propertiesof single neurons in the magnorecipient layersof the striate cortex, which show a deep mini-mum to isoluminant red/green borders and,like the magnocellular layers of the dLGN,show broad band spectral properties.An alternative explanation for the normal

chromatic contrast sensitivity in an achro-matopsic patient such as ours is that there isstill substantial colour opponent processing.6This is presumably mediated by the undam-aged parts of the P channel which reside in theintact striate cortex and is consistent with thenormal chromatic VEPs in the striate cortexwhich is mediated by the P pathway. Therehave been recent demonstrations of a dissocia-tion between the ability to detect an isolumi-nant chromatic pattern and to discriminatecolour differences in cases of complete2' orincomplete achromatopsia.24 It is possible thatthe "blob" pathway of Vl mediates chromaticdiscrimination whereas the "interblob" path-way processes colour differences to detect iso-luminant patterns without conveying thenature of the hue of which the pattern is com-posed. Thus partial striate involvement, inwhich the "blob" arm of the P channel isselectively compromised would result inachromatopsia. The integrity of the"interblob" pathway of Vl, with its intact pro-jections to extrastriate cortex, could thenaccount for residual visual processes.We are preparing to examine, by functional

magnetic neuroimaging, the surviving extras-triate areas involved in the patient's residualprocessing of chromatic signals.

We thank the patient for his enthusiastic cooperation. Theresearch was supported by a Human Capital and MobilityGrant (HC and M: CHRX CT 930261) and the award of aVision Research Fellowship by the Wellcome Trust to JJN.

1 Meadows JC. Disturbed perception of colours associatedwith localized cerebral lesions. Brain 1974;97;615-32.

2 Plant GT. Disorders of colour vision in diseases of the ner-vous system. In: Foster DH, ed. Inherited and acquiredcolour vision deficiences: fundamental aspects and clinicalstudies. Vision and visual dysfunction. Vol 7. Basingstoke:Macmillan Press, 1991:173-98.

3 Zeki S. A century of cerebral achromatopsia. Brain 1990;113:1721-77.

4 Stiles WS. Mechanisms of colour vision. London: AcademicPress, 1978.

5 Mollon JD, Newcombe F, Polden PG, Ratcliffe G. On thepresence of three cone mechanisms in a case of totalachromatopsia. In: Colour vision deficiencies. Vol V.Bristol: Hilger, 1980:130-5.

6 Heywood CA, Cowey A, Newcombe F. Chromatic dis-crimination in a cortically colour blind observer. Eur JNeurosci 1991;3:802-12.

7 Damasio A, Yamada T, Damasio H, Corbett J, McKee J.Central achromatopsia: behavioral, anatomic, and physio-logic aspects. Neurology 1980;30:1064-71.

8 Zeki SM, Watson JDG, Lueck CJ, Friston KJ, Kennard C,Frackowiak RSJ. A direct demonstration of functionalspecialization in human visual cortex. J Neurosci 1991;11:641-9.

9 Corbetta M, Miezin FM, Dobmeyer S, Shulman GL,Petersen SE. Selective and divided attention duringvisual discriminations of shape, color and speed: func-tional anatomy by positron emission tomography. JNeurosci 1991;11:2383-402.

10 Gulyis B, Roland PE. Cortical fields participating in formand colour discrimination in the human brain.Neuroreport 1991;2:585-8.

11 Mackay G. Dunlop, JC. The cerebral lesions in a case ofcomplete acquired colour-blindness. Scottish Medical andSurgical Journal 1899;5:513-2.

12 Schiller PH, Logothetis NK, Charles ER. Functions of thecolour-opponent and broad-band channels of the visualsystem. Nature. 1990;343:68-70.

13 Pearlman AL, Birch J, Meadows JC. Cerebral color blind-ness: an acquired defect in hue discrimination. AnnNeurol 1979;5:253-61.

14 Victor JD, Maiese K, Shapley R, Sidtis J, Gazzaniga MS.Acquired central dyschromatopsia: analysis of a case withpreservation of color discrimination. Clin Vision Sci1989;4: 183-96.

15 Green GJ, Lessell S. Acquired cerebral dyschromatopsia.Arch Ophthalmol 1977:95:121-8.

16 Albert ML, Reches A, Silverberg R. Hemianopic colorblindness. J Neurol Neurosurg Psychiatry 1975;38:546-9.

17 Heywood CA, Wilson B, Cowey A. A case study of corticalcolour "blindness" with relatively intact achromatic dis-crimination. J Neurol Neurosurg Psychiatry 1987;50:22-9.

18 Newcombe F, Ratcliff G. Agnosia: a disorder of objectrecognition. In: Michel F, Schott B, eds. Les syndromes dedisconnexion calleuse chez l'homme. Lyon: Colloque Inter-national de Lyon, 1975: 317-41.

19 Ratcliff G, Newcombe F. Object recognition: some deduc-tions from the clinical evidence. In: Ellis AW, ed.Normality and pathology of cognitive functions. London:Academic Press, 1982:147-7 1.

20 Cavanagh P, Anstis SM. Do opponent-color channels con-tribute to motion? Invest Ophthalmol Vis Sci 1986(suppl27): 291.

21 Heywood CA, Cowey A. Newcombe F. On the role of par-vocellular (P) and magnocellular (M) pathways in cere-bral achromatopsia. Brain 1994;1 17:245-54.

22 Saito H, Tanaka K, Isono H, Yasuda M, Mikami A.Directionally selective response of cells in the middletemporal area (MT) of the macaque monkey to themovement of equiluminous opponent color stimuli. ExpBrain Res 1989;75:1-14.

23 Ferrara VP, Nealey TA, Maunsell JHR. Mixed parvocellularand magnocellular geniculate signals in visual area V4.Nature 1992;358:756-58.

24 Barbur JL, Harlow, AJ, Plant G. Insights into the differentexploits of colour in the visual cortex. Proc R Soc Lond.Series B. 1994;258:327-34.

643

group.bmj.com on January 15, 2014 - Published by jnnp.bmj.comDownloaded from

doi: 10.1136/jnnp.60.6.638 1996 60: 638-643J Neurol Neurosurg Psychiatry

 C A Heywood, J J Nicholas and A Cowey sensitivity in an achromatopsic patient.chromatic and achromatic contrast Behavioural and electrophysiological

http://jnnp.bmj.com/content/60/6/638Updated information and services can be found at:

These include:

References http://jnnp.bmj.com/content/60/6/638#related-urls

Article cited in:

serviceEmail alerting

the box at the top right corner of the online article.Receive free email alerts when new articles cite this article. Sign up in

Notes

http://group.bmj.com/group/rights-licensing/permissionsTo request permissions go to:

http://journals.bmj.com/cgi/reprintformTo order reprints go to:

http://group.bmj.com/subscribe/To subscribe to BMJ go to:

group.bmj.com on January 15, 2014 - Published by jnnp.bmj.comDownloaded from