Constraining the use of constraints

Pylyshyn uses constraints to solve many of the problems associated with the inverse problem in vision. We are sympathetic to such an approach, and ind eed, we think that in many cases constraints allow tractable solutions to o therwise insoluble problems, Ne argue, however, that Pylyshyn has been too quick to assume that certain perceptual phenomena can be explained by appea ling to constraints embodied in the visual machinery. For several more comp lex perceptual phenomena it is not clear how one proceeds to look for const raints once simple constraints like rigidity have been abandoned.