Foundations Of Cognitive Science

Problem of Representation

The physical symbol system hypothesis (Newell, 1980; Newell & Simon, 1976) is attractive because the underlying syntactic processes preserve semantic interpretations assigned to the symbols (Haugeland, 1985).  That is, according to Haugeland’s Formalist’s Motto, if you take care of the syntax, then the semantics will take care of itself.  Underlying this notion is that the symbols manipulated by a physical symbol system are content-bearing – they designate states of affairs in the real world.

The problem of representation (Cummins, 1989) concerns explaining how symbols designate – by what means is it possible for symbols to bear content?  Cummins argues that this is a fundamental problem for cognitive science, and points out that there seem to be only four different approaches to solving it: similarity (symbols are similar in appearance to the content they represent), covariance (a causal relation exists between states of affairs in the world and symbolic states of affairs), adaptational role (Millikan, 1984), and functional role (mental contents are individuated functionally – a form of methodological solipsism).

It is important to distinguish between the problem of representation (singular) and the problem of representations (plural) – they are very different problems, both discussed by Cummins (1989), and both given their own entry in this dictionary.


  1. Cummins, R. (1989). Meaning And Mental Representation. Cambridge, MA: MIT Press.
  2. Haugeland, J. (1985). Artificial Intelligence: The Very Idea. Cambridge, MA: MIT Press.
  3. Millikan, R. (1984). Language, Thought, and Other Biological Categories.  Cambridge, MA: MIT Press.
  4. Newell, A. (1980). Physical symbol systems. Cognitive Science, 4, 135-183.
  5. Newell, A., & Simon, H. A. (1976). Computer science as empirical inquiry - Symbols and search. Communications of the ACM, 19(3), 113-126.

(Added October 2010)