Rethinking mental representation through the epistemology of modeling

Abstract

The problems associated with classical symbolism need not entail abandoning representationalism. While representational theories of mind have been widely challenged by embodied and action-oriented approaches, representationalism remains alive and well in empirical research on higher cognition, where the notion of conceptual representation appears indispensable. I argue that a reconciliation between representationalism and action-oriented theories is possible through a pragmatist interpretation of mental model research. Mental models are flexible knowledge structures that integrate information at multiple levels of abstraction to serve diverse epistemic and pragmatic functions. By combining this research with the epistemology of scientific modeling, I propose replacing symbolic referentialism with pragmatist inferentialism, an account of representation grounded in notions of inference and adequacy for purpose, rather than reference and truth conditions. This framework dissolves the divide between philosophical and psychological theories of concepts and aligns more closely with current empirical research on higher cognition than either classical symbolism or its radical anti-representational alternatives.