The Role of Base Rates in Category Learning

Previous researchers have discovered perplexing inconsistencies in how human subjects appear to utilize knowledge of category base rates when making category judgments. In particular, Medin and Edelson (1988) found an “inverse base rate effect” in which subjects tended to select a rare category when tested with a combination of conflicting cues, and Gluck and Bower (1988) reported apparent “base rate neglect” in which subjects tended to select a rare category when tested with a single symptom whose objective diagnosticity was equal for all categories. In this article I suggest that two principles underlie those effects: First, base rate information is learned and consistently deployed during all training and testing cases. Second, the dominant effect of base rates is to cause frequent categories to be learned before rare categories, so that the common categories are encoded in terms of their typical features, and the rare categories are encoded by whichever features distinguish them relative to the already-learned, common categories. Four new experiments are reported which confirm those principles. Experiment 1 replicates and extends the basic inverse base rate effect and establishes a standard for subsequent experiments and modeling. Experiment 2 shows that pre-training on a subset of categories has comparable effects as giving those categories high base rates during training. Experiment 3 shows that apparent “base rate neglect” is simply an attenuated case of the “inverse base rate effect,” hence any model of one effect should account for the other. Experiment 4 uses probabilistic categories, and shows that apparent “base rate neglect” can be obtained even for cues that occur in less than 50% of the rare category’s exemplars, and that cues that are not associated with any category decrease the base-rate bias. The principles are formalized in a new connectionist model that modifies attention to stimulus features based on the network’s current knowledge. Quantitative fits to the empirical data are reported, and provide additional support for the principles.

[1]  Baruch Fischhoff,et al.  When Do Base Rates Affect Predictions , 1981 .

[2]  D. Medin,et al.  Problem structure and the use of base-rate information from experience. , 1988, Journal of experimental psychology. General.

[3]  Richard C. Atkinson,et al.  Stimulus Sampling Theory , 1967 .

[4]  J. Kruschke,et al.  ALCOVE: an exemplar-based connectionist model of category learning. , 1992, Psychological review.

[5]  D. Medin,et al.  Sensitivity to changes in base-rate information , 1991 .

[6]  G. Bower,et al.  From conditioning to category learning: an adaptive network model. , 1988, Journal of experimental psychology. General.

[7]  W. Estes,et al.  Base-rate effects in category learning: a comparison of parallel network and memory storage-retrieval models. , 1989, Journal of experimental psychology. Learning, memory, and cognition.

[8]  John K. Kruschke,et al.  Human Category Learning: Implications for Backpropagation Models , 1993 .

[9]  John R. Anderson.  The Adaptive Character of Thought , 1990 .

[10]  A. Markman.  LMS rules and the inverse base-rate effect: Comment on Gluck and Bower (1988). , 1989 .

[11]  R. Rescorla,et al.  A theory of Pavlovian conditioning : Variations in the effectiveness of reinforcement and nonreinforcement , 1972 .

[12]  CONNECTIONISM AND HUMAN LEARNING - CRITIQUE OF GLUCK AND BOWER (1988) , 1990 .

[13]  L. Beach,et al.  Experience and the base-rate fallacy. , 1982, Organizational behavior and human performance.

[14]  John K. Kruschke,et al.  Investigations of an Exemplar-Based Connectionist Model of Category Learning , 1992 .

[15]  John K. Kruschke,et al.  Three principles for models of category learning , 1993 .

[16]  H. Eysenck.  THINKING , 1958 .

[17]  Douglas B. Lenat,et al.  On the thresholds of knowledge , 1987, Proceedings of the International Workshop on Artificial Intelligence for Industrial Applications.

[18]  Douglas L. Medin,et al.  Context theory of classification learning. , 1978 .

[19]  R. Nosofsky.  Attention, similarity, and the identification-categorization relationship. , 1986, Journal of experimental psychology. General.

[20]  R. Nosofsky.  Relation between the Rational Model and the Context Model of Categorization , 1991 .

[21]  A. Scott,et al.  The Analysis of Categorical Data from Complex Sample Surveys: Chi-Squared Tests for Goodness of Fit and Independence in Two-Way Tables , 1981 .

[22]  J. L. Myers,et al.  Modeling Probabilistic Categorization Data: Exemplar Memory and Connectionist Nets , 1994 .

[23]  R. Nosofsky,et al.  Combining exemplar-based category representations and connectionist learning rules. , 1992, Journal of experimental psychology. Learning, memory, and cognition.

[24]  David R. Shanks.  Connectionist Accounts of the Inverse Base-rate Effect in Categorization , 1992 .

[25]  A. Tversky,et al.  On the psychology of prediction , 1973 .

[26]  S. Lewandowsky.  Base-rate neglect in ALCOVE: A critical reevaluation. , 1995 .

[27]  J. Koehler.  The Base Rate Fallacy Myth , 1993 .

[28]  A. Tversky.  Features of Similarity , 1977 .

[29]  D R Shanks,et al.  Connectionism and the Learning of Probabilistic Concepts , 1990, The Quarterly journal of experimental psychology. A, Human experimental psychology.