Spatial and Chromatic Interactions in the Lateral Geniculate Body of the Rhesus Monkey1

THE RECEPTORS AND NERVE CELLS that make up the visual pathway must convey and interpret information on both the form and the color of retinal images. In higher mammals little is known about the degree to which nerve cells are specialized for handling these types of information. In a visual stimulus the importance of spatial attributes, and especially of dark-light contours, first became obvious with the discovery by Hartline (20 ) of lateral inhibition in the Limulus, a type of study that was extended to mammals when Kuffler (28) demonstrated t,hat the receptive fields of retinal ganglion cells in the cat are subdivided into a center and an opponent surround. The opponent principle, in which spatially separated excitatory and inhibitory regions are pitted against each other, has now been observed for retinal ganglion cells in the frog (1), the lizard (9), the rabbit (3), the rat (4), the ground squirrel (33), and the monkey (24). Similar effects have been seen in the lateral genie ulate bod .y and visual cortex in the cat (23, 25, 26), and also recently at these levels in the monkey. In 1958 De Valois and his collaborators (15) observed geniculate cells in the macaque monkey that were excited by one set of wavelengths and inhibited by another, making it apparent that in higher mammals the spectral composition of the stimulus was also an important variable. Similar opponent-color effects have since been described in the primate at the level of the retin .a1 ganglion cell (24), and in the visual cortex (34). I n the cat the absence or rarity of oppon .ent-color mecha nisms (19, 36 > may be related to an inferior ability to discriminate color (31, 32, 40); indeed since Svaetichin’s (41) original observation of opponent-color responses in the fish retina (Spotentials), similar response patterns have been seen only in animals thought to have good color vision. Given the existence of two opponent mechanisms system, one for the spatial variable and the other for in the monkey visual color, it is natural to ask whether these occupy the same channels, or are confined to separate groups of cells. In the goldfish it is clear from the work of Wagner, MacNichol, and Wolbarsht (43, 45) that opponent-color and opponent-spatial