Cobweb and variants are computational models of learning of and categorization with hierarchical category structure, which were inspired by and accounted for well-known psychological categorization phenomena of basic level effects, typicality effects, and fan effects. Fisher and Yoo (1993) are an endpoint in that modeling work, and that paper (below) very briefly summarizes the earlier accounts of these phenomena using Cobweb, and extends the basic framework to problem-solving. This paper's real goal was to treat problem-solving as an extended categorization task, with coarse accounts of the transition from novice to expert problem solving (i.e., the transition from naive reliance on surface features to increasing reliance on the "optimal" features, be they deep/causal or surface); the "optimal" features are posited to be those that optimize a tradeoff between cue validity and cost necessary to infer them.

  • Fisher, D., & Yoo, J. (1993). "Problem solving, categorization, and concept learning: A unifying view," in G. Nakamura, R. Taraban, & D. Medin (eds.), The Psychology of Learning and Motivation, 29, San Diego, CA: Academic Press, 219--255.

    The paper by Fisher and Langley (1990) goes into more depth about the accounts of basic level, typicality, fan effects, and interactions between them under the basic categorization framework. An important point relevant to all cognitive models is that they should go beyond the known data by making predictions about behavior under circumstances that have not yey been tested, thus pointing the way for subsequent psychological study.

  • Fisher, D., & Langley, P. (1990). "The Structure and Formation of Natural Categories," in G. Bower (ed.), The Psychology of Learning and Motivation, 26, San Diego, CA: Academic Press, 241--284.

    Earlier papers that led to the latter works above include:

  • Fisher, D. (1988). "A Computational Account of Basic Level and Typicality Effects", Proceedings of the Seventh National Conference on Artificial Intelligence. Minneapolis, MN: Morgan Kaufmann, 233--238.

  • Silber, J., & Fisher, D. (1989). "A Model of Natural Category Structure and its Behavioral Implications," Proceedings of the Eleventh Annual Conference of the Cognitive Science Society, Ann Arbor, MI: Lawrence Erlbaum, 884--891.

    More recent work by one of my graduate students looked at some of the phenomena above and other phenomena using naive Bayesian classifiers and variants on these. This work culminated in his dissertation (Frey, 2003).

  • Frey, L., & Fisher, D. (2003). "Augmented Naive Bayesian Model of Classification Learning," Proceedings of the Twenty-Fifth Annual Conference of the Cognitive Science Society Boston, MA (on CD).

    My interest in connections between cognitive psychology, cognitive science, and artificial intelligence was a major theme in my research from 1987-1995. This book and other collaborative efforts are indicative of these interests:

  • Fisher, D., Pazzani, M., & Langley, P. (eds.), (1991). Concept Formation: Experience and Knowledge in Unsupervised Learning. San Mateo, CA: Morgan Kaufmann, 472 pages, 15 chapters.

  • Billman, D., Fisher, D., Gluck, M., Langley, P., & Pazzani, M. (1990). "Computational Models of Category Learning" (symposia), Proceedings of the Twelfth Annual Conference of the Cognitive Science Society, Boston, MA: Lawrence Erlbaum, 989--996.