Patients with cerebral achromatopsia, a perceptual disorder caused by
ventromedial occipital brain damage, can be completely unable to arran
ge colours in chromatic sequence and fail most conventional tests of c
olour blindness. A possible explanation for cerebral achromatopsia is
that the colour-opponent parvocellular (P) channel has been selectivel
y and totally destroyed at the level of visual cortex, leaving vision
to be mediated by the broad-band magnocellular (M) channel. The persis
tence of normal occipital visually evoked potentials, and preserved se
nsitivity to isoluminant chromatic gratings indicates that if this hyp
othesis is correct the destruction must occur beyond the striate corte
x. We have shown that an achromatopsic subject can detect chromatic bo
rders and construct shape from colour, and that he can even perceive t
he apparent direction of motion of a phase shifted isoluminant chromat
ic grating where perceived direction depends on knowing the sign of th
e colour diffence, i.e., which colour is which in the stripes. This an
d other evidence suggests that perhaps only one part of the cortical P
channel has been destroyed. Does the critical area involved in achrom
atopsia correspond to cortical area V4 of monkeys, often implicated in
processing wavelength? When Visual Area 4 is totally ablated in monke
ys they have only a mild colour discrimination impairment and easily s
olve the colour ordering and colour selection tasks that an achromatop
sic patient finds impossible. However, monkeys with ventromedial damag
e rostral to Area V4 do perform like achromatopsic patients, suggestin
g that the role of V4 in the perception of colour is still unclear and
that the colour area of the human brain does not correspond to area V
4.