ABSTRACT
Connectionist models are widely used in the cognitive sciences, and well beyond. This is so despite the fact that some critics have charged that we can’t learn about cognition using connectionist models (Fodor and Pylyshyn, 1988). Although researchers who use connectionist models have offered a number of defenses of their methods (Smolensky, 1988; McClelland, 1988), and there is growing empirical evidence suggesting that these models have been successful in advancing cognitive science, there is no consensus on how they work. This chapter explores the epistemic roles played by connectionist models of cognition, and offers a formal analysis of how connectionist models explain.
