The segmentation of visual form

Abstract

The argument of this work is that, despite the massive body of literature that has accumulated in the decades since the discovery of 'gestalt' as the ruling principle of perception, little genuine progress in solving the problem posed by the visual perception of form has been made. This state of affairs is attributed, moreover, to a fundementally inadequate formulation of this problem. It is not enough merely to revise this or that theory, or this or that experimental design, if the argument is correct; rather, it is necessary to revise the formulation of the form problem upon which theory and experimental design rest. Thus, the reformulation suggested is that (a) form is the unit which segments space, and consequently that (b) the problem posed by this unit is essentially that of its segmentation/formation of space, rather than that of its recognition/conservation through change in space; the former is the primary, the latter the secondary, (psycho-physical) problem posed by the visual perception of form. This work also contains a segmentation (spatial/holistic) theory of form, and five experiments designed to test this theory against current recognition (dimensional/analytic) theories of form (for example, see Corcoran, 1971); these experiments are all concerned with different facets of the role played by contour in visual perception, and they provide some evidence for the former, and against the latter, type of theory. It should be pointed out that both in the main body of the text, and in an appendix, it is argued that segmentation is primarily two-dimensional rather than three-dimensional: two-dimensional 'figure'form is primary over three-dimensional 'object' form in perceptual development, and indeed, the latter is constructed from the former. (This hypothesis is part of a more general point of view about cognition, namely that there is an a priori spatial system which is used to process perceptual input, and establish in it the spatial structure of perceptual experience, but one whose conceptual implications and properties become available for symbolisation and thinking when it is freed from the task of perceptual processing by being lifted out of perception into a visual form of representation which Bruner terms 'ikonic' (See Bruner et al., 1966).)