Complex Symbol-Processing in Conposit, A Transiently Localist Connectionist Architecture

  • John A. Barnden
Part of the The Springer International Series In Engineering and Computer Science book series (SECS, volume 292)


Two unusual primitives for the structuring of symbolic information in connectionist systems were discussed in [9]. The primitives are called Relative-Position Encoding (RPE) and Pattern-Similarity Association (PSA). The present article shows that the primitives are powerful and convenient for effecting cognitively sophisticated connectionist symbol processing. Specifically, it shows how RPE and PSA are used in a connectionist implementation of Johnson-Laird’s mental model theory of syllogistic reasoning [23] [24] [25]. The symbol processing achieved is therefore at the level of complexity to be found in existing, detailed information-processing theories in cognitive psychology. This system is called Conposit/SYLL, but for brevity it will often be referred to here as Conposit. To be exact, Conposit is a general framework for implementing rule-based systems in connectionism, and Conposit/SYLL is just one instance of it. (The name “Conposit” is derived from “Connectionist POSI-Tional encoding.” Conposit/SYLL is a major extension beyond the preliminary version described in [2]).


Mental Model Head Register Tentative Conclusion Command Signal Connectionist System 
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]
    Barnden, J.A. (1984). On short-term information processing in connectionist theories. Cognition and Brain Theory, 7(1), pp. 25–59.Google Scholar
  2. [2]
    Barnden, J.A. (1989). Neural-net implementation of complex symbol-processing in a mental model approach to syllogistic reasoning. In Procs. 11th Int. Joint Conf. on Artificial Intelligence. San Mateo, CA: Morgan Kaufmann.Google Scholar
  3. [3]
    Barnden, J.A. (1990). Syllogistic mental models: exercising some connectionist representation and control methods. Memoranda in Computer and Cognitive Science, No. MCCS-90-204, Computing Research Laboratory, New Mexico State University, Las Cruces, NM 88003.Google Scholar
  4. [4]
    Barnden, J.A. (1991). Encoding complex symbolic data structures with some unusual connectionist techniques. In J.A. Barnden & J.B. Pollack (Eds.), Advances in Connectionist and Neural Computation Theory, Vol. 1. Norwood, N.J.: Ablex Publishing Corp.Google Scholar
  5. [5]
    Barnden, J.A. (1992a). 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.Google Scholar
  6. [6]
    Barnden, J.A. (1992b). Connectionism, structure-sensitivity, and system-aticity: refining the task requirements. Memoranda in Computer and Cognitive Science, No. MCCS-92-227, Computing Research Laboratory, New Mexico State University, Las Cruces, NM 88003.Google Scholar
  7. [7]
    Barnden, J.A. (1993a). On using analogy to reconcile connections and symbols. In D.S. Levine & M. Aparicio (Eds), Neural Networks for Knowledge Representation and Inference, pp.27–64. Hillsdale, N.J.: Lawrence Erlbaum Associates.Google Scholar
  8. [8]
    Barnden, J.A. (1993b). Time phases, pointers, rules and embedding. Behavioral and Brain Sciences, 16(3), pp.451–452. Invited Commentary (on Shastri and Ajjanagadde’s “From Simple Associations to Systematic Reasoning”).CrossRefGoogle Scholar
  9. [9]
    Barnden, J.A. & Srinivas, K. (1991). Encoding techniques for complex information structures in connectionist systems. Connection Science, 3(3), pp.263–309.CrossRefGoogle Scholar
  10. [10]
    Barnden, J.A. & Srinivas, K. (1992). Overcoming rule-based rigidity and connectionist limitations through massively-parallel case-based reasoning. Int. J. Man-Machine Studies, 36, pp.221–246.CrossRefGoogle Scholar
  11. [11]
    Barnden, J.A. & Srinivas, K. (1993). Temporal winner-take-all networks: a time-based mechanism for fast selection in neural networks. IEEE Trans. Neural Networks, 4(5), pp.844–853.CrossRefGoogle Scholar
  12. [12]
    Barnden, J.A., Srinivas, K. & Dharmavaratha, D. (1990). Winner-take-all networks: time-based versus activation-based mechanisms for various selection goals. In Procs. IEEE International Symposium on Circuits and Systems, New Orleans, May 1990.Google Scholar
  13. [13]
    Blank, D.S., Meeden, L.A. & Marshall, J.B. (1992). Exploring the symbolic/subsymbolic continuum: a case study of RAAM. In Dinsmore, J. (Ed.), The Symbolic and Connectionist Paradigms: Closing the Gap. Hillsdale, N.J.: Lawrence Erlbaum. pp. 113–148.Google Scholar
  14. [14]
    Chalmers, D.J. (1990). Syntactic transformations on distributed representations. Connection Science, 2(1&2), pp.53–62.CrossRefGoogle Scholar
  15. [15]
    Charniak, E. & Santos, E. (1987/1991). A connectionist context-free parser which is not context-free, but then it is not really connectionist either. In Procs. 9th Annual Conference of the Cognitive Science Society. Hillsdale, N.J.: Lawrence Erlbaum. A revised version appears in J.A. Barnden & J.B. Pollack (Eds.), Advances in Connectionist and Neural Computation Theory, Vol. 1. Norwood, N.J.: Ablex Publishing Corp., March 1991.Google Scholar
  16. [16]
    Chrisman, L. (1991). Learning recursive distributed representations for holistic computation. Connection Science, 3(4), pp.354–366.CrossRefGoogle Scholar
  17. [17]
    Chun, H.W., Bookman, L.A. & Afshartous, N. (1987). Network Regions: alternatives to the winner-take-all structure. In Procs. Tenth Int. Joint Conf. On Artificial Intelligence, pp.380–387. Los Altos, CA: Morgan Kaufmann.Google Scholar
  18. [18]
    Dyer, M.G. (1991). Symbolic NeuroEngineering and natural language processing: a multilevel research approach. In J. A. Barnden & J.B. Pollack (Eds.), Advances in Connectionist and Neural Computation Theory, Vol. 1. Norwood, N.J.: Ablex Publishing Corp. pp.32–86.Google Scholar
  19. [19]
    Feldman, J. A. & Ballard, D. H. (1982). Connectionist models and their properties. Cognitive Science, 6, pp.205–254.CrossRefGoogle Scholar
  20. [20]
    Fodor, J.A. & Pylyshyn, Z.W. (1988). Connectionism and cognitive architecture: a critical analysis. In S. Pinker & J. Mehler (Eds.), Connections and symbols, Cambridge, Mass.: MIT Press, and Amsterdam: Elsevier. (Reprinted from Cognition, 28, 1988.)Google Scholar
  21. [21]
    Grossberg, S. (1988). Nonlinear neural networks: principles, mechanisms, and architectures. Neural Networks, 1, 17–61.CrossRefGoogle Scholar
  22. [22]
    Hinton, G.E. (1990). Mapping part-whole hierarchies into connectionist networks. Artificial Intelligence, 46(1–2), pp.47–75.CrossRefGoogle Scholar
  23. [23]
    Johnson-Laird, P.N. (1983). Mental models: towards a cognitive science of language, inference and consciousness. Cambridge, Mass.: Harvard University Press.Google Scholar
  24. [24]
    Johnson-Laird, P.N. & Bara, B.G. (1984). Syllogistic inference. Cognition, 16(1), 1–61.CrossRefGoogle Scholar
  25. [25]
    Johnson-Laird, P.N. & Byrne, R.M.J. (1991). Deduction. Hove, U.K.: Lawrence Erlbaum.Google Scholar
  26. [26]
    Lippmann, R.R. (1987). An introduction to computing with neural nets. IEEEASSP Magazine, 4, 4–22.Google Scholar
  27. [27]
    Oakhill, J.V. & Johnson-Laird, P.N. (1985). The effects of belief on the spontaneous production of syllogistic conclusions. The Quarterly J. of Experimental Psych., 37A, pp.553–569.Google Scholar
  28. [28]
    Oakhill, J.V., Johnson-Laird, P.N., & Garnham, A. (1989). Believability and syllogistic reasoning. Cognition, 31(2), pp.117–140.CrossRefGoogle Scholar
  29. [29]
    Plate, T. (1991). Holographsreduced representations. Tech. Report CRG-TR-91-1, Dept. of Computer Science, University of Toronto, Canada M5S 1A4.Google Scholar
  30. [30]
    Pollack, J.B. (1990). Recursive distributed representations. Artificial Intelligence, 46(1–2), pp.77–105.CrossRefGoogle Scholar
  31. [31]
    Shastri, L. & Ajjanagadde, V. (1993). 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.Google Scholar
  32. [32]
    Smolensky, P. (1990). Tensor product variable binding and the representation of symbolic structures in connectionist systems. Artificial Intelligence, 46(1–2), pp. 159–216.CrossRefMATHMathSciNetGoogle Scholar
  33. [33]
    Stolcke, A. & Wu, D. (1992). Tree matching with recursive distributed representations. TR-92-025, Computer Science Division, University of California, Berkeley, CA 94704.Google Scholar
  34. [34]
    Touretzky, D.S. (1990). BoltzCONS: dynamic symbol structures in a connectionist network. Artificial Intelligence, 46(1–2), pp. 5–46.CrossRefMathSciNetGoogle Scholar
  35. [35]
    Yuille, A.L. & Grzywacz. (1989). A winner-take-all mechanism based on presynaptic inhibition feedback. Neural Computation, 1, pp.334–347.CrossRefGoogle Scholar

Copyright information

© Kluwer Academic Publishers 1995

Authors and Affiliations

  • John A. Barnden
    • 1
  1. 1.Computing Research Laboratory and Computer Science DepartmentNew Mexico State University Las CrucesNew Mexico

Personalised recommendations