How Big Is A Stimulus?

Abstract

Abstract The last few years have seen a great increase in the popularity of organizational approaches to the study of perception, more or less explicitly indebted to Gestalt theory, and of direct or stimulus-informational approaches, explicitly indebted to the work of Gibson (1950, 1979). The mere fact that the appearance of a pattern cannot in general be completely predicted from the appearances of its parts presented in isolation does not by itself provide a basis for reviving anything like Gestalt theory. And the mere fact that what has been called higher-order stimulus information about distal (object) properties are normally provided by the environment, and that such information is sometimes used by the normal viewer, by itself implies nothing whatsoever about the processes of perception. Helmholtz discussed quite explicitly what later was taken by the Gestaltists as the law of good continuation; he discussed at length the importance of motion-produced stimulation and of active exploration in vision and in touch as providing what was later to be called the invariant information in the stimulus transformations (Cassirer, 1944; Gibson, 1950, 1979). Moreover, Helmholtz stressed that as far as our perceptions are concerned, the distal properties are usually directly perceived (i.e., we are not aware of the activities of the individual receptors, nor of the processes by which the percepts are constructed). If one is to go beyond (or oppose) Helmholtz in these respects, therefore, it will have to be in terms of evidence about the ways in which larger spans of stimulus information are used, not merely by showing that they are present in stimulation, or that they are used.