Cue competition in human categorization: Contingency or the Rescorla-Wagner Learning Rule? Comment on Shanks (1991).
Eric R. Melz, Patricia W. Cheng, Keith J. Holyoak, Michael R. Waldmann · Journal of Experimental Psychology Learning Memory and Cognition · 1993
Shanks (1991) reported experiments that show selective-learning effects in a categorization task, and presented simulations of his data using a connectionist network model implementing the Rescorla-Wagner (R-W) theory of animal conditioning. He concluded that his results (a) support the application of the R-W theory to account for human categorization, and (b) contradict a particular variant of contingency-b ased theories of categorizatio n. We examine these conclusions. We show that the asymptotic weights produced by the R-W model actually predict systematic deviations from the observed human learning data. Shanks claimed that his simulations provided good qualitative fits to the observed data when the weights in the networks were allowed to reach their asymptotic values. However, analytic derivations of the asymptotic weights reveal that the final weights obtained in Shanks' Simulations 1 and 2 do not correspond to the actual asymptotic weights, apparently because the networks were not in fact run to asymptote. We show that a contingency-based theory that incorporates the notion of focal sets can provide a more adequate explanation of cue competition than does the R-W model. Shanks (1991) described three experiments in which subjects were asked to play the role of medical diagnosticians. After being presented with a series of case histories (patterns of patients' symptoms associated with various fictitious diseases), subjects were asked to rate how strongly they associated each symptom with each disease, using a 0-100 rating scale. Subjects' association ratings consistently varied with the relative predictiveness of each symptom for the disease as defined by the Rescorla and Wagner (1972) model (hereinafter the R-W model) rather than with the cue validity of the symptoms (i.e., the probability of the disease given a symptom; Shanks, 1991, Experiments 1-3) or with the contingency of the symptoms (i.e., the difference between the probability of the disease given the presence of a symptom and that probability given the absence of the symptom; Shanks, 1991, Experiments 2-3). For example, consider the design of Shanks's (1991) Experiment 2. (Summaries of the experimental designs of Experiments 1-3 are presented in Table 1.) In Shanks's (1991)