Abstract
Glass patterns are made by taking an array of randomly positioned dots, and pairing each with another according to a predefined geometric rule. Glass patterns have been used in numerous studies to study how local signals are integrated to generate a coherent percept of form. In light of several studies suggesting that form and color are treated independently in the primate brain, it is of interest to study the chromatic selectivity of the physiological mechanisms supporting Glass pattern detection.
We recorded extracellularly from neurons in area V4 of an awake, passively viewing monkey while the neurons' receptive field was stimulated with different types of Glass patterns (circular, radial, parallel, and hyperbolic), or by an array of randomly oriented dot pairs. All patterns were made of 500 dot pairs presented on a dark background. A given pattern was presented for 1.6 sec, and was redrawn 4 times during that interval to avoid local retinal adaptation. Among 51 individual neurons or multiunit clusters, 29 showed preferential responses to one type of Glass pattern compared to the random array. Each form selective neurons was then tested with its preferred pattern presented in different, equiluminant colors. Of the 29 form selective neurons, 24 showed a clear preference for a specific color: their response to the preferred color was at least 1.5 times that for a white pattern. We conclude that V4 neurons represent a likely substrate for the processing of global form. Moreover, our data show that color and form are not treated by distinct neural populations within V4.