Skip to main content
SpringerLink
Log in

High-level reasoning, computational challenges for connectionism, and the Conposit solution

  • Published:
Applied Intelligence Aims and scope Submit manuscript

Abstract

Sophisticated symbol processing in connectionist systems can be supported by two primitive representational techniques calledRelative-Position Encoding (RPE) andPattern-Similarity Association (PSA), and a selection technique calledTemporal-Winner-Take-All (TWTA). TWTA effects winner-take-all selection on the basis of fine signal-timing differences as opposed to activation-level differences. Both RPE and PSA are for the encoding of highly temporary associations between representations. RPE is based on the way activation patterns are positioned relative to each other within a network. Under PSA, two patterns are temporarily associated if they have (suitable) subpatterns that are (suitably) similar. The article shows how particular versions of the primitives are used to good effect in a system called Conposit/SYLL. This is a connectionist implementation of a slightly simplified version of a complex existing psychological theory, namely Johnson-Laird's account of syllogistic reasoning. The computational processes in this theory present a major implementational challenge to connectionism. The challenge lies in the mutability, multiplicity, and diversity of the working memory structures, and the elaborateness of the processing needed for them. Conposit/SYLL's techniques allow it to meet the challenge. The implementation of symbolic processing in Conposit/SYLL is an interesting application of connectionism partly because it significantly affects the design of the symbolic processing level itself. In particular, it encourages the use of associative as opposed to pointer-based data structures, and the use of random as opposed to ordered iteration over sets of data structures. In addition, the article discusses Conposit/SYLL's somewhat unusual variable-binding approach.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  1. J.A. Barnden, “On short-term information processing in connectionist theories,”Cognition and Brain Theory, 7(1), pp. 25–59, 1984.

    Google Scholar 

  2. J.A. Barnden, “Diagrammatic short-term information processing by neural mechanisms.”Cognition and Brain Theory, 7, (3 & 4), pp. 285–328, 1985.

    Google Scholar 

  3. J.A. Barnden, “Neural-net implementation of complex symbol-processing in a mental model approach to syllogistic reasoning,” inProcs. 11th Int. Joint Conf. on Artificial Intelligence, San Mateo, CA: Morgan Kaufmann, pp. 568–573, 1989.

    Google Scholar 

  4. J.A. Barnden, “Syllogistic mental models: exercising some connectionist representation and control methods,” Computing Research Laboratory, New Mexico State University, Las Cruces, NM 88003,Memoranda in Computer and Cognitive Science, No. MCCS-90-204, 1990.

  5. J.A. Barnden, “Encoding complex symbolic data structures with some unusual connectionist techniques,” in J.A. Barnden and J.B. Pollack (Eds.),Advances in Connectionist and Neural Computation Theory, Vol. 1, Norwood, N.J.: Ablex Publishing Corp., pp. 180–240, 1991.

    Google Scholar 

  6. J.A. Barnden, “Connectionism, generalization and propositional attitudes: a catalogue of challenging issues”, in J. Dinsmore (Ed.),The Symbolic and Connectionist Paradigms: Closing the Gap, Hillsdale, N.J.: Lawrence Erlbaum, pp. 149–178, 1992.

    Google Scholar 

  7. J.A. Barnden, “Connectionism, structure-sensitivity, and systematicity: refining the task requirements,” Computing Research Laboratory, New Mexico State University, Las Cruces, NM 88003,Memoranda in Computer and Cognitive Science, No. MCCS-92-227, 1992.

  8. J.A. Barnden, “Time phases, pointers, rules and embedding,”Behavioral and Brain Sciences, 16(3), pp. 451–452.

  9. J.A. Barnden, “On using analogy to reconcile connections and symbols,” in D.S. Levine and M. Aparicio (Eds.),Neural Networks for Knowledge Representation and Inference, Hillsdale, N.J.: Lawrence Erlbaum Associates, pp. 27–64, 1994.

    Google Scholar 

  10. J.A. Barnden, “Complex symbol-processing in Conposit, a transiently localist connectionist architecture,” in R. Sun and L. Bookman (Eds.),Computational Architectures for Integrating Neural and Symbolic Processes, Kluwer, pp. 21–68, 1994.

  11. J.A. Barnden and J.B. Pollack (Eds.),Advances in Connectionist and Neural Computation Theory, Vol. 1, Norwood, N.J.: Ablex Publishing Corp., 1991.

    Google Scholar 

  12. J.A. Barnden and K. Srinivas, “Encoding techniques for complex information structures in connectionist systems,”Connection Science, 3(3), pp. 263–309, 1991.

    Google Scholar 

  13. J.A. Barnden and K. Srinivas, “Overcoming rule-based rigidity and connectionist limitations through massively-parallel case-based reasoning,”Int. J. Man-Machine Studies, Vol. 36, pp. 221–246, 1992.

    Google Scholar 

  14. J.A. Barnden and K. Srinivas, “Temporal winner-talke-all networks: a time-based mechanism for fast selection in neural networks,”IEEE Trans. Neural Networks, 4(5), pp. 844–853, 1993.

    Google Scholar 

  15. D.S. Blank, L.A. Meeden, and J.B. Marshall, “Exploring the symbolic/subsymbolic continuum: a case study of RAAM“, in J. Dinsmore (Ed.),The Symbolic and Connectionist Paradigms: Closing the Gap, Hillsdale, N.J.: Lawrence Erlbaum, pp. 113–148, 1992.

    Google Scholar 

  16. D.J. Chalmers, “Syntactic transformations on distributed representations,”Connection Science, 2(1 & 2), pp. 53–62, 1990.

    Google Scholar 

  17. E. Charniak and E. Santos, “A connectionist context-free parser which is not context-free, but then it is not really connectionist either”, inProcs. 9th Annual Conference of the Cognitive Science Society, Hillsdale, N.J.: Lawrence Erlbaum, 1987. A revised version appears in J.A. Barnden and J.B. Pollack (Eds.),Advances in Connectionist and Neural Computation Theory, Vol. 1, Norwood, N.J.: Ablex Publishing Corp., pp. 123–134, 1991.

    Google Scholar 

  18. L. Chrisman, “Learning recursive distributed representations for holistic computation,”Connection Science, 3(4), pp. 354–366, 1991.

    Google Scholar 

  19. H.W. Chun, L.A. Bookman, and N. Afshartous, “Network Regions: alternatives to the winner-take-all structure,” inProcs. Tenth Int. Joint Conf. On Artificial Intelligence, Los Altos, CA: Morgan Kaufmann, pp. 380–387, 1987.

    Google Scholar 

  20. M.G. Dyer, “Symbolic Neuro Engineering and natural language processing: a multilevel research approach,” in J.A. Barnden and J.B. Pollack (Eds.),Advances in Connectionist and Neural Computation Theory, Vol. 1, Norwood, N.J.: Ablex Publishing Corp., pp. 32–86, 1991.

    Google Scholar 

  21. S.E. Fahlman,NETL: A system for representing real-wrpold knowledge, Cambridge, MA: MIT Press, 1979.

    Google Scholar 

  22. J.A. Feldman and D.H. Ballard, “Connectionist models and their properties,”Cognitive Science, 6, pp. 205–254, 1982.

    Google Scholar 

  23. J.A. Fodor and Z.W. Pylyshyn, “Connectionism and cognitive architecture: a critical analysis,” in S. Pinker and J. Mehler (Eds.),Connections and symbols, Cambridge, Mass.: MIT Press, 1988, pp. 3–71. (Reprinted fromCognition, 28, 1988.)

    Google Scholar 

  24. S. Grossberg, “Nonlinear neural networks: principles, mechanisms, and architectures,”Neural Networks, 1, pp. 17–61, 1988.

    Google Scholar 

  25. G.E. Hinton, “Mapping part-whole hierarchies into connectionist networks,”Artificial Intelligence, 46(1 & 2), pp. 47–75, 1990.

    Google Scholar 

  26. G.E. Hinton (Ed.),Connectionist symbol processing. Cambridge, MA: MIT Press, 1991.

    Google Scholar 

  27. P.N. Johnson-Laird,Mental Models: Towards a Cognitive Science of Language, Inference and Consciousness, Cambridge, Mass.: Harvard University Press, 1983.

    Google Scholar 

  28. P.N. Johnson-Laird and B.G. Bara, “Syllogistic inference,”Cognition, 16(1), pp. 1–61, 1984.

    Google Scholar 

  29. P.N. Johnson-Laird and R.M.J. Byrne,Deduction. Hove, U.K.: Lawrence Erlbaum, 1991.

    Google Scholar 

  30. T.E. Lange, “Hybrid connectionist models: temporary bridges over the gap between the symbolic and subsymbolic,” in J. Dinsmore (Ed.),The Symbolic and Connectionist Hillsdale, N.J.: Paradigms; Closing the Gap, Lawrence Erlbaum, pp. 237–289, 1992.

  31. R.R. Lippmann, “An introduction to computing with neural nets,”IEEE ASSP Magazine, 4, pp. 4–22, 1987.

    Google Scholar 

  32. J.V. Oakhill and P.N. Johnson-Laird, “The effects of belief on the spontaneous production of syllogistic conclusions,”The Quarterly J. of Experimental Psych., 37A, pp. 553–569, 1985.

    Google Scholar 

  33. T. Plate, “Holographic reduced representations,” Dept. of Computer Science, University of Toronto, Canada M5S 1A4, Tech. Report CRG-TR-91-1, 1991.

  34. J.B. Pollack, “Recursive distributed representations,”Artificial Intelligence, 46 (1 & 2), pp. 77–105, 1990.

    Google Scholar 

  35. L. Shastri and V. Ajjanagadde, “From simple associations to systematic reasoning: a connectionist representation of rules, variables, and dynamic bindings using temporal synchrony,”Behavioral and Brain Sciences, 16(3), pp. 417–494, 1993.

    Google Scholar 

  36. P. Smolensky, “Tensor product variable binding and the representation of symbolic structures in connectionist systems,”Artificial Intelligence, 46(1–2), pp. 159–216, 1990.

    Google Scholar 

  37. A. Stolcke and D. Wu, “Tree matching with recursive distributed representations,” Computer Science Division, University of California, Berkeley, CA 94704, TR-92-025, 1992.

    Google Scholar 

  38. R. Sun, “On variable binding in connectionist networks,”Connection Science, 4(2), pp. 93–124, 1992.

    Google Scholar 

  39. D.S. Touretzky, “BoltzCONS: dynamic symbol structures in a connectionist network,”Artificial Intelligence, 46(1 & 2), pp. 5–46, 1990.

    Google Scholar 

  40. A.L. Yuille and Grzywacz, “A winner-take-all mechanism based on presynaptic inhibition feedback,”Neural Computation, 1, pp. 334–347, 1989.

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Computing Research Laboratory & Dept. of Computer Science, Dept. 3CRL, New Mexico State University, Box 30001, 88003-8001, Las Cruces, NM

    John A. Barnden

Authors
  1. John A. Barnden
    View author publications

    You can also search for this author in PubMed Google Scholar

Rights and permissions

Reprints and permissions

About this article

Cite this article

Barnden, J.A. High-level reasoning, computational challenges for connectionism, and the Conposit solution. Appl Intell 5, 103–135 (1995). https://doi.org/10.1007/BF00877228

Download citation

  • Issue Date:

  • DOI: https://doi.org/10.1007/BF00877228

Keywords

Search

Navigation