Abstract

In this analysis, I address the recent debate over the functional characteristics of mental imagery, in which the question is whether the format for imagistic representation is pictorial or sentential. I offer an account of imagery and cognition as an important aspect of the theory of mental representation, one which serves as a test for prevalent views of the nature of explanation in cognitive science. ;The first chapter articulates a functionalist theory of internal states. I argue for the need to individuate mental states by virtue of their formal features, but in a way which is consistent with the fact that natural representational systems are self-interpreting. To that end, I describe a basic capacity for imaging which underlies any mental representation. ;In the second chapter, I discuss recent experiments and argue that a correct interpretation of the data cannot be determined on methodological grounds. The substantive argument against mental pictures is set forth and shown to rest upon two distinct assumptions. ;The third chapter aims to show that the argument that mental pictures are inherently ambiguous fails as an argument for an exclusively sentential format. I argue further that sentential states alone are insufficient to account for all rational behavior. The conclusion is that conceptual considerations do not preclude nondiscursive images, and a view of cognition as broader than computation favors them. I propose a logical foundation for an account of access to mental images-under-descriptions. ;The fourth chapter sets forth principles governing operations in which images might occur naturally. I argue that such an account depends upon the possibility of pictorial attitudes, mental features which are theoretically important for the notion of representational competence. ;In the fifth chapter, I show the relevance of the general account described in the previous chapter by applying it to diverse contexts: animal representation, hypothesis in perception, and self-representation