Optical velocity patterns, velocity-sensitive neurons, and space perception: a hypothesis.

A hypothesis is put forward of how global patterns of optical flow, as discussed by Gibson, Johansson, and others, could be processed by relatively simple physiological mechanisms. It is suggested that there may exist motion-sensitive cells in the visual system which operate on the optical flow over the retina, and, in so doing, structure the visual field in terms of distinct surfaces that move and/or lie at varying distances from the observer. First, concepts of static and dynamic perspective relative to a sphere centered about the eye are developed, partly on the basis of the work of Gordon. It is pointed out that the velocity flow pattern has a very simple form making it amenable to analysis by relatively low-level mechanisms. Next a higher-order variable of optical flow, the 'convexity', is defined; under the assumption of a rigid environment, convexity is shown to be related to relative depth. It is then postulated that velocity-sensitive cells having center surround organization could be linked in such a way as to define a higher-order cell, the convexity cell, having functional properties that make it sensitive to the convexity function. The response profile of a layer of such cells would provide an efficient structuring of the visual field in terms of distinct optical surfaces. Relevant evidence is briefly discussed. Lastly, the optical flow patterns corresponding to rotations of the observer are considered. It is shown that the convexity cell is insensitive to rotations and in consequence responds in an invariant fashion to aspects of the optical flow which are related to the surrounding environment.