Representation in Perception and Cognition

There is disagreement over how the notion of representation should be understood in cognitive science. Many investigators equate representations with symbols, that is, with syntactically defined elements in an internal symbol system (see Fodor, 1975; Pylyshyn, 1984). Some have urged that representation be understood wholly in these terms (Fodor, 1980; Stich, 1983, chap. 8), ¹ while others have implicitly adopted this position (Newell & Simon, 1976; Newell, 1980). In recent years there have been two challenges to this orthodoxy. First, a number of philosophers have argued that “representation” should be understood in its classical sense, as denoting a “stands for” relation between representation and represented. These philosophers include those outside the symbolist orthodoxy (Dretske, 1988; Hatfield, 1988b; Searle, 1983) as well as those within (Fodor, 1987). Second, there has been a growing challenge to the orthodox view of representation under the banner of connectionism (Rumelhart & McClelland, 1986b; Smolensky, 1988). Although this connectionist challenge has been of sufficient weight to evoke emotionally charged rebukes from the symbolist camp (Fodor & Pylyshyn, 1988; Pinker & Prince, 1988), connectionists as a group have not articulated a conception of representation to replace the symbolist view. Nonetheless, most connectionists agree on the need for a nonsymbolic notion of representation (Feldman & Ballard, 1982; Hinton, McClelland, & Rumelhart, 1986; Smolensky, 1988; see also Kosslyn & Hatfield, 1984).