COMPUTATIONAL MODELS OF VISUAL NEURONS SPECIALIZED IN THE DETECTION OF PERIODIC AND APERIODIC ORIENTED VISUAL-STIMULI - BAR AND GRATING CELLS

Computational models of periodic- and aperiodic-pattern selective cell s, also called grating and bar cells, respectively. are proposed. Grat ing cells are found in areas V1 and V2 of the visual cortex of monkeys and respond strongly to bar gratings of a given orientation and perio dicity but very weakly or not at all to single bars. This nonlinear be haviour, which is quite different from the spatial frequency filtering behaviour exhibited by the other types of orientation-selective neuro ns such as the simple cells, is incorporated in the proposed computati onal model by using an AND-type non-linearity to combine the responses of sim pie cells with symmetric receptive field profiles and opposite polarities. The functional behaviour of bar cells, which are found in the same areas of the visual cortex as grating cells, is less well ex plored and documented in the literature. In general, these cells respo nd to single bars and their responses decrease when further bars are a dded to form a periodic pattern. These properties of bar cells are imp lemented in a computational model in which the responses of bar cells are computed as thresholded differences of the responses of correspond ing complex (or simple) cells and grating cells. Bar and grating cells seem to play complementary roles in resolving the ambiguity with whic h the responses of simple and complex cells represent oriented visual stimuli, in that bar cells are selective only for form information as present in contours and grating cells only respond to oriented texture information. The proposed model is capable of explaining the results of neurophysiological experiments as well as the psychophysical observ ation that the perception of texture and the perception of form are co mplementary processes.