Many visual systems also have the ability to distinguish between different parts of the visual field on the basis of different distributions of wavelengths of light reaching the eye from various places in the field of view. The perceptual correlate of this ability to distinguish light of different wavelength is called colour vision. Humans are just one of many species which have this ability — so when we see different colours, it is because the eye is viewing a scene which reflects different distributions of wavelengths in the visible spectrum (about 400 to 700 nanometres).
The basis for this ability to see different colours is that we have three different kinds of cone receptor cells in the retina, each responding best to different parts of the visible spectrum. They are often referred to as L, M, and S cones (long-, medium-, and short-wave sensitive). The relationship between the absorption of these cones and human colour vision is well understood and forms the basis of colour reproduction systems such as colour TV. What is less well understood is what happens to the information from the cones once it gets distributed in the visual cortex. Briefly, what seems to be happening is as follows:
• The signals from the three cone types get transmitted as opponent pairs, such as L−M and Y−B where Y=L+M (see Parraga, Troscianko, and Tolhurst 2002 for a description of what this implies for the coding of natural images).
• There is also a luminance (intensity) representation from the combined L+M cone activity.
• There has been debate about the extent to which colour and luminance information is used in the encoding of visual attributes such as movement — see points below.
• Human vision contains neurons which respond to rapid changes in light and are often assumed to be insensitive to colour (see Troscianko et al. 1996);
• The chromatic pathways mentioned above have a rather sluggish response to rapid changes and are therefore considered to be poor encoders of movement (see Troscianko et al. 1996).
The argument about the above aspects of human vision has been informed as a result of the existence of a rare and highly interesting type of brain damage. Patients suffering from damage to the visual cortex (e.g. as a result of stroke or a blow to the head) occasionally report that they see the world in 'shades of grey' after the damage has occurred. Their retinae are completely unaffected by this, and for at least some of the people it is known that their colour vision was normal before the onset of the damage. The damage is therefore in the brain, not the eye. The condition is called cerebral (or central) achromatopsia, to distinguish it from an inability to see colour because there is only one receptor type in the retina, as can also occur (see Meadows 1974, Mollon et al. 1980).
Various research groups set out to study such individuals, and a curious finding began to emerge. While it seems universally true that the people reported seeing only shades of grey, they could perform above chance in tasks that required some colour discrimination. For example, if two colours of equal luminance were placed side by side, it was found that achromatopsic individuals knew that there was something different about the two halves of the display, without being aware of the colours. Confusingly, some patients were also able to name a few colours correctly, again apparently without 'seeing' them. Thus, evidence was emerging that there was some colour information in the brain, without this information allowing people to have conscious perception of colour (see Heywood, Cowey, and Newcombe 1991, Troscianko et al. 1996, Cavanagh et al. 1998, Heywood, Kentridge, and Cowey 1998).
We were fortunate to be able to study three such patients: HJA, WM, and JPC (Troscianko et al. 1996, Cavanagh et al. 1998). They were all achromatopsic in that they reported seeing the world in 'shades of grey' but they had intact retinae. We devised experiments in which the patients had to respond 'same' or 'different' while looking at a TV display in which the two halves of the screen were either the same or different. The differences could be of luminance, or colour, or both luminance and colour. This allowed us to study the contribution which colour information made to their performance level, without at any point requiring the patient to perceive the colour consciously. As in other such research, we found that performance was better when there was a colour and a luminance difference, rather than a colour difference alone.
We were keen to find out how this 'unconscious' colour information was being transmitted in the visual system. We added 'noise' consisting of random variations in luminance which were either unchanging in time (static noise) or alternating 25 times per second (dynamic noise). This noise had a dramatic, but different, effect on the patients. HJA's ability to respond to colour information was unimpaired with dynamic noise but got worse with static noise. This is as expected from what is generally known about colour vision – since colour information is insensitive to rapid change (see above), putting in such rapid change should not impair colour-based performance. However, patient WM showed a very different set of results: his performance dropped to chance with dynamic noise. We were at a loss to understand why this should be, since the noise should not be visible to neurons encoding colour information. We hypothesized that there must be another stream of 'colour' information in vision that responds to rapid changes, but does not reach conscious perception.
What might be the function of such additional information? It seemed likely (given its fast response) that it might be involved in movement perception. We therefore tested WM in a brain-imaging system using functional magnetic resonance imaging (fMRI) (see brain imaging: methods), expecting to find an augmented response, with colour information present, of the area known as V5 which is known to encode motion signals. We found no such augmented response (Troscianko et al. 1997). However, psychophysical experiments on several achromatopsic patients including WM and JPC showed that their perception of motion was indeed much affected by colour information. Thus, it seemed that we were on the right track but that area V5 was not the locus of the effect.
This issue was finally settled by some further fMRI work by a group at Harvard (Hadjikhani et al. 1998). By requiring participants to look at alternating colours, or colours alternating with a grey field and producing coloured after-images, they showed that there may be a separate area of the visual cortex (which they named V8). This area seems to be involved in the conscious perception of colour. So the hypothesis is that this may be the area which is damaged in cerebral achromatopsia. This leaves intact the ability of the 'unconscious' colour information to assist with the encoding of movement, but this is not likely to occur in V5, but rather may occur in a different extra-striate area. Together with the Harvard group, we confirmed that patients WM and JPC had a largely intact contribution of colour to motion perception. Therefore, the function of this unconscious colour information in our cortex is to enhance motion information. To perceive colour consciously, we require another separate cortical area, and it is this which is damaged in cerebral achromatopsia.
The study of individuals who lack colour perception has therefore taught us a surprising amount about the neural basis of colour information.
(Published 2004)
— Tom Troscianko
- Bibliography
- Cavanagh, P., Henaff, M.-A., Michel, F., Landis, T., Troscianko, T., and Intriligator, J. (1998). 'Complete sparing of high-contrast color input to motion perception in cortical color blindness'. Nature Neuroscience, 1.
- Hadjikhani, N., Liu, A. K., Dale, A. M., Cavanagh, P., and Tootell, R. B. H. (1998). 'Retinotopy and color sensitivity in human cortical area V8'. Nature Neuroscience, 1.
- Heywood, C. A., Cowey, A., and Newcombe, F. (1991). 'Chromatic discrimination in a cortically colour blind observer'. European Journal of Neuroscience, 3.
- — — Kentridge, R. W., and Cowey, A. (1998). 'Form and motion from colour in cerebral achromatopsia'. Experimental Brain Research, 123.
- Meadows, J. C. (1974). 'Disturbed perception of colours associated with localized cerebral lesions'. Brain, 97.
- Mollon, J. D., Newcombe, F., Polden, P. G., and Ratcliffe, G. (1980). 'On the presence of three cone mechanisms in a case of total achromatopsia'. In Mollon et al. (eds.), Colour Vision Deficiencies, vol v.
- Parraga, C. A., Troscianko, T., and Tolhurst, D. J. (2002). 'Spatio-chromatic properties of natural images and human vision'. Current Biology, 12.
- Troscianko, T., Davidoff, J., Humphreys, G., Landis, T., Fahle, M., Greenlee, M., Brugger, P., and Phillips, W. (1996). 'Human colour discrimination based on a non-parvocellular pathway'. Current Biology, 6.
- — — Greenlee, M. W., Brugger, P., Freitag, P., Kraemer, F. M., and Radue, E. W. (1997). 'An fMRI investigation of a patient with cerebral achromatopsia: evidence for a role of chromatic extrastriate mechanisms in motion encoding'. Perception, 26 (suppl.).
