Representation and constraints: the inverse problem and the structure of visual space

Visual space can be distinguished from physical space. The first is found in visual experience, while the second is defined independently of perception. Theorists have wondered about the relation between the two. Some investigators have concluded that visual space is non-Euclidean, and that it does not have a single metric structure. Here it is argued (1) that visual space exhibits contraction in all three dimensions with increasing distance from the observer, (2) that experienced features of this contraction (including the apparent convergence of lines in visual experience that are produced from physically parallel stimuli in ordinary viewing conditions) are not the same as would be the experience of a perspective projection onto a frontoparallel plane, and (3) that such contraction is consistent with size constancy. These properties of visual space are different from those that would be predicted if spatial perception resulted from the successful solution of the inverse problem. They are consistent with the notion that optical constraints have been internalized. More generally, they are also consistent with the notion that visual spatial structures bear a resemblance relation to physical spatial structures. This notion supports a type of representational relation that is distinct from mere causal correspondence. The reticence of some philosophers and psychologists to discuss the structure of phenomenal space is diagnosed in terms of the simple materialism and the functionalism of the 1970s and 1980s.