An Internal Report for Connectionists

  • Noel E. Sharkey
  • Stuart A. Jackson
Part of the The Springer International Series In Engineering and Computer Science book series (SECS, volume 292)


Although there is a considerable diversity of representational formalisms in the Connectionist literature, most of the excitement about representation has been concerned with the idea of distributed representations. Dissatisfied with the Symbolic tradition, and in search of the new, many cognitive theorists began to infiltrate connectionism in search of a new theory of mind. Like the Classicists, these theorists required that a constructed, analytic theory of mind postulate complex structured representations; it is only by having structural similarities among representations that we can account for the systematic nature of human thought. The Classical view is that the systematic relations between representations are necessarily based on the composition of similar syntactic elements. Likewise, some of the representational types found in the connectionist literature satisfy this requirement, only by virtue of the fact that they are similar to Classically conceived symbolic representations. For example, the structure of complex expression may be maintained in vector frames consisting of explicit tokens or complex expressions that are essentially passed around a net [18]. Only distributed representations offered the promise of a novel representational scheme that could underpin a connectionist theory of cognition; a scheme that relied upon the assumption that structural similarity relations can be captured in a connectionist net by systematic vectorial distances between the distributed representations created.


Decision Boundary Hide Unit Output Unit Decision Space Connectionist School 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    Aizawa, K. (1992) Review of Philosophy and Connectionist Theory. W. Ramsey, S.P. Stich & D.E. Rumelhart (Eds). In Mind and Language, 7.Google Scholar
  2. [2]
    Baldwin, A. (1993) The role of connectionist representation in necessary inference for natural language processing. Unpublished PhD. thesis, University of Exeter.Google Scholar
  3. [3]
    Chalmers, D.J. (1990) Why Fodor and Pylyshyn were wrong: The simplest refutation. Proceedings of the Twelfth Annual Conference of the Cognitive Science Society, 340–347.Google Scholar
  4. [4]
    Collins, A.M. & Quillian, M.R. (1969) Retrieval time from semantic memory. Journal of Verbal Learning and Verbal Behavior, 8, 240–247.CrossRefGoogle Scholar
  5. [5]
    Dolan, C.P. & Smolensky, P. (1989) Tensor product production system: a modular architecture and representation. Connection Science, 1.1., 53–68.CrossRefGoogle Scholar
  6. [6]
    Elman, J.L. (1989). Representation and structure in connectionist models. CRL Technical Report 8903, Center for Research in Language, University of California, San Diego, CA.Google Scholar
  7. [7]
    Fodor, J.A. (1975) The Language of Thought. New York: Crowell.Google Scholar
  8. [8]
    Fodor, J.A. & McLaughlin, B. (1990) Connectionism and the problems of systematicity: Why Smolensky’s solution doesn’t work. Cognition, 35, 183–204.CrossRefGoogle Scholar
  9. [9]
    Fodor, J.A., & Pylyshyn, Z.W. (1988). Connectionism and cognitive architecture: A critical analysis. Cognition, 28, 2–71.CrossRefGoogle Scholar
  10. [10]
    van Gelder, T.( 1989) Classical questions, radical answers: Connectionism and the structure of mental representations. Horgan, T. & Tienson J. (Eds) Connectionism and the Philosophy of Mind.Google Scholar
  11. [11]
    van Gelder, T. (1990) Compositionality: A connectionist variation on a classical theme. Cognitive Science, 14, 355–384.CrossRefGoogle Scholar
  12. [12]
    van Gelder, T. (1992) Defining ‘Distributed Representation’. Connection Science, 4(3,4), 175–192.CrossRefGoogle Scholar
  13. [13]
    Hinton, G.E. (1989) Connectionist learning procedures. Artificial Intelligence, 40, 184–235.CrossRefGoogle Scholar
  14. [14]
    McCulloch W.S. & Pitts W.H. (1943) A logical calculus of ideas immanent in nervous activity. Bulletin of Mathematical Biophysics, 5, 115–133.MATHCrossRefGoogle Scholar
  15. [15]
    McDermott, D. (1976) Artificial intelligence meets natural stupidity. SIGART Newsletter, no. 57, 4–9.Google Scholar
  16. [16]
    Pollack, J.B. (1990) Recursive distributed representations. Artificial Intelligence, 46, 77–105.CrossRefGoogle Scholar
  17. [17]
    Rumelhart, D.E., Hinton, G.E. & Williams, R.J. (1986) Learning representations by back-propagating errors. Nature, 323, 533–536.CrossRefGoogle Scholar
  18. [18]
    Sharkey, N.E. (1991) Connectionist representation techniques. AI Review, 5,3, 143–167.Google Scholar
  19. [19]
    Sharkey, N.E. (1992) The Ghost in the Hybrid: A Study of Uniquely Connectionist Representations. AISB Quarterly. 10–16.Google Scholar
  20. [20]
    Sharkey, N.E. & Jackson, S.A. (In press) Three horns of the representational trilemma. In V. Honavar & L. Uhr (Eds) Artificial Intelligence and Neural Networks: Steps towards Principled Integration. Volume 1: Basic Paradigms; Learning Representational Issues; and Integrated Architectures. Cambridge, MA: Academic Press.Google Scholar

Copyright information

© Kluwer Academic Publishers 1995

Authors and Affiliations

  • Noel E. Sharkey
    • 1
  • Stuart A. Jackson
    • 1
  1. 1.Computer Science DepartmentUniversity of SheffieldSheffieldUK

Personalised recommendations