The Central Visual System And Color

Behavioral studies of vision have led to a consensus view that the facts of human color vision can best be understood as reflecting the operation of three separate mechanisms. These mechanisms are conceived as implying the presence of three parallel channels of information in the central visual system. Two of these are opponent channels reflecting, respectively, the mutual antagonisms of red and green and of yellow and blue. The third is a nonopponent channel that provides achromatic information (the luminance mechanism). The spectrally opponent and nonopponent cells of the type described previously were first recorded approximately 40 years ago not in the retina but in the lateral geniculate nucleus (LGN), which is a large thalamic structure that serves as a relay site situated in the central visual pathway between the retina and the visual cortex. Following the discovery of these cells it was immediately recognized that there is a compelling analogy between the physiological results and the behavioral conception of the color mechanisms. Thus, the L—M cells and the (L+M)—S cells

(Fig. 4) have many characteristic features that appear to be like those of the behaviorally defined red/green and yellow/blue color channels. Similarly, the L+M cells have the appropriate spectral sensitivity as well as many other features that appear to make them ideal candidates to serve as the luminance mechanism.

Over time, however, closer examination of the relationships between the standard behavioral model and the physiological results has revealed a lack of complete correspondence between the two, and this forced a reevaluation of the idea that the spectrally opponent cells of the retina and LGN directly represent the color mechanisms documented in behavioral experiments. To note just one problem, consider the perceptual observation that the spectrum takes on a distinctly reddish appearance in the short wavelengths (this can be seen in the hue naming results of Fig. 3C). Among color vision theorists, the conventional explanation for this fact is that the red/green color mechanism must receive some signals from the S cones. However, electrophysiological studies indicate that the L—M cells do not have an S cone input so they cannot directly account for this feature of red/green color vision. Many other disparities between the behavioral aspects of human color vision and the physiology of cells located early in the visual pathway have also been enumerated. The inevitable conclusion is that although the spectrally opponent cells clearly transmit the information necessary for color vision, the response properties of these cells cannot directly account for many features of human color vision. This means that there must be some further transformations of the signals provided by spectrally opponent cells and those transformations are assumed to take place somewhere in the visual cortex.

The picture of how color information is processed and elaborated in the visual cortex is still very sketchy. Researchers have tried to understand how color information is encoded by cortical neurons, and they have pursued the hypothesis that color processing might be localized to particular component regions of the cortex. Analysis of the response properties of single cortical cells has been avidly pursued. Unfortunately, the results have led to conflicting views as to how cortical cells encode color information. This is understandable since the task is far from simple. Much of the difficulty arises from the fact that although the responses of cells in the retina and in the LGN are only modestly dependent on the spatial and temporal features of the stimulus, the responses of cortical cells are very much conditioned by these properties. The consequence is that studies utilizing differing spatio-

temporal stimulus features have often reached quite different conclusions about the nature ofcortical color coding. What is known, both from single cell studies conducted on monkeys and imaging studies of human brains, is that color-selective responses can indeed be recorded in primary visual cortex (area V-1) and at other locations in the extrastriate cortex. Studies also make clear that transformations of the spectrally opponent responses of LGN cells do occur in V-1 and that such transformations arise, at least in part, from dynamic interactions between cortical cells. For example, neural feedback circuits may significantly amplify contributions from signals originating in S cones relative to those S cone signals recorded at more peripheral locations. Such a change brings the physiological picture more in line with behavioral measures of color vision. Similarly, at least some cells in V-1 are known to combine signals from the two classes of LGN spectrally opponent cells. This too brings the physiology closer to the coding scheme inferred from behavioral studies of color vision. Although the relationships between cortical codes for color and psychophysical models of color coding are not clear, significant progress is being made toward their rationalization.

For years it has been argued that the processing of color information can be localized in the extrastriate visual cortex. From studies of both monkey and human brains, regions in the lingual and fusiform gyrus (sometimes characterized as area V-4) have been identified as particularly important for color. In humans this area responds robustly to the presentation of stimuli designed to specifically probe color vision, and it has been reported that cells in monkey area V-4 exhibit some forms of color constancy. Providing additional support for the idea of localized cortical representation of color are clinical descriptions of patients who have suffered a loss of color vision as a result of cortical damage (cerebral achromatopsia). The nature of the color vision loss in cerebral achromatopsia is complex and beyond the scope of this article. For our purposes, the important fact is that although the locus of damage that results in cerebral achromatopsia varies significantly among described cases, it most often does include area V-4.

Stop Anxiety Attacks

Stop Anxiety Attacks

Here's How You Could End Anxiety and Panic Attacks For Good Prevent Anxiety in Your Golden Years Without Harmful Prescription Drugs. If You Give Me 15 minutes, I Will Show You a Breakthrough That Will Change The Way You Think About Anxiety and Panic Attacks Forever! If you are still suffering because your doctor can't help you, here's some great news...!

Get My Free Ebook


Post a comment