Neural Mechanisms for Color Vision in the Retina

As noted previously, the human retina contains multiple types of cone photopigment. Multiple cone types are a necessary but not sufficient basis to support a color vision capacity. In addition, there must also be an appropriately organized nervous system. The extraction of information that allows for color vision begins in the neural networks of the retina.

In addition to the photoreceptors, there are four other major classes of nerve cell in the retina: bipolar, ganglion, horizontal, and amacrine cells. These form intricately organized vertical and horizontal possessing networks arrayed through the thickness of the retina with bipolar and ganglion cells serving as the principal vertical pathway and the other two types providing a rich array of horizontally organized connections. Each of the four types of cell in turn consists of discrete subtypes. Currently, a majority of the latter are poorly defined, both structurally and functionally, but it is believed that it will eventually be possible to characterize as many as 50 distinct types of cells in the primate retina. The main business of the networks formed by these cells is to perform the first stages in the processing of the retinal image. These tasks include the analysis of local spatial and temporal variations and the regulation of visual sensitivity. The processing of color information proceeds in the context of all these other analyses.

Because each cone type behaves univariantly, the extraction of color information requires the comparison ofthe signals from cone types containing different photopigments. There are two principal ways in which cone signal information is combined in the nervous system: additively (spectrally nonopponent) or sub-tractively (spectrally opponent). Cells of the former type sum signals from the L and M cone types (L + M). Because they do not respond differentially to wavelength differences irrespective of the relative intensity, cells so wired cannot transmit information useful for the production of color vision. The response properties of spectrally opponent cells are produced by convergence of excitatory and inhibitory signals onto recipient nerve cells. The cone signal combinations are classified into two main groups: L—M and (L + M)—S (for each of these types the signs can be reversed, i.e., there are both L—M and M—L types). Response profiles for opponent and nonopponent cells are shown in Fig. 4. It can be seen that, unlike the nonopponent cells, spectrally opponent cells will yield different responses to different wavelengths of light irrespective of the relative intensities of those lights. They thus can transmit information that may be useful for supporting color vision. With regard to color information, the output cells of the retina (the ganglion cells) are classified into three groups: those that transmit nonopponent information into the central nervous system, those carrying M—L information, and those whose response patterns are (M + L)—S (Fig. 4). Much is known about the anatomy of such cells and the input pathways that yield these different response patterns.

Several structural and functional properties of the retina map are directly mapped into the quality of human color vision. For example, S cones are sparsely distributed across the retina and are absent entirely from its very central portion (the fovea). A consequence is that the color information contributed from S cone signal pathways is lost for stimuli that are very small (i.e., trichromacy gives way to dichromacy under such viewing conditions). The connection pathways from the L and M cones also vary across the retina. Although the picture is still somewhat controversial, in general there is an observed decrease in the relative potency of L/M spectrally opponent signals toward the peripheral parts of the retina. This decrease is presumed to be a factor in the gradual decline in sensitivity of red/green color vision for stimuli located away from the direction of gaze. Finally, due to variations in the neural circuitry, there are significant differences in the spatial and temporal sensitivities of the different types of retinal cells. One consequence of this is that our color vision becomes progressively more restricted for regions in the visual scene that are very small and/or are changing very rapidly. Thus, human color vision is trichromatic only for relatively large and slowly changing stimuli, giving way to dichromatic color vision as the space/time components of the stimulus are increased and, eventually, one can lose color vision entirely (i.e., become monochromatic) for very small and/or very rapidly changing stimuli.

Breaking Bulimia

Breaking Bulimia

We have all been there: turning to the refrigerator if feeling lonely or bored or indulging in seconds or thirds if strained. But if you suffer from bulimia, the from time to time urge to overeat is more like an obsession.

Get My Free Ebook


Post a comment