Human learning of elemental category structures: Revising the classic result of Shepard, Hovland, and Jenkins (1961).

Abstract

The findings of Shepard, Hovland, and Jenkins (1961) on the relative ease of learning 6 elemental types of 2-way classifications have been deeply influential 2 times over: 1st, as a rebuke to pure stimulus generalization accounts, and again as the leading benchmark for evaluating formal models of human category learning. The litmus test for models is the ability to simulate an observed advantage in learning a category structure based on an exclusive-or (XOR) rule over 2 relevant dimensions (Type II) relative to category structures that have no perfectly predictive cue or cue combination (including the linearly-separable Type IV). However, a review of the literature reveals that a Type II advantage over Type IV is found only under highly specific experimental conditions. We investigate when and why a Type II advantage exists to determine the appropriate benchmark for models and the psychological theories they represent. A series of 8 experiments link particular conditions of learning to outcomes ranging from a traditional Type II advantage to compelling non-differences and reversals (i.e., Type IV advantage). Common interpretations of the Type II advantage as either a broad-based phenomenon of human learning or as strong evidence for an attention-mediated similarity-based account are called into question by our findings. Finally, a role for verbalization in the category learning process is supported.

Evaluating Amazon's Mechanical Turk as a Tool for Experimental Behavioral Research

“…A similar result was reported by Love (2002) [45] who found only a marginal Type II advantage compared to the Type IV problem in a related design. In a series of experiments, Kurtz et al [47] have argued that the Type II advantage can be explained by the extent to which instructions emphasize verbal rules.…”

Human Category Learning: Toward a Broader Explanatory Account

“…In addition, we were inspired to seriously question the most surprising finding in the traditional SHJ (Shepard, Hovland, & Jenkins) ordering: the tremendous ease with which learners acquire a nonlinearly separable and seemingly unnatural category structure in Type IIthe XOR rule that requires learners to separate, for example, white squares and black triangles from white triangles and black squares. In a series of behavioral studies (Kurtz, Levering, Stanton, Romero, & Morris, 2013), we found compelling evidence that the classic ordering actually depends on the inclusion of task instructions that encourage rule formation. Specifically, Type II is not significantly faster than Type IV under neutral instructions, but it is learned more quickly when the experimental task explicitly encourages hypothesis testing or selective attention.…”

Combination or Differentiation? Two theories of processing order in classification

“…One possible explanation of this result within the framework of Combination Theory is that incidental conditions promote simple additive combination rules (all that is required for Type IV) over more esoteric combination rules, such as exclusive-or. However, given a recent report that, even under intentional conditions, Type IV is sometimes easier than Type II (Kurtz et al, 2013), further research is probably required.…”