On the Interaction of Prior Knowledge and Stimulus Structure in Category Learning

Abstract

Contemporary theories of categorization propose that concepts are coherent in virtue of being embedded in a network of theories about the world. Those theories function to pick out some of the many possible features of a set of objects as most salient for purposes of classification, a process that is complex and still poorly understood (Murphy & Medin, 1985). Part of what makes this account incomplete is a lack of information as to (1) what makes a feature salient on a given occasion and (2) how feature salience interacts with category structure to determine the course of learning. We report on the results of three studies of category learning using complex schematic drawings to show that (1) the contrast set defined by one's initial encounters with category exemplars can be a source of individual differences in feature salience assignments; (2) such effects are short-lived in the face of clear evidence about actual feature diagnosticity; and (3) more robust prior hypotheses interact with category structure to either enhance learning or impede it. The enhancement occurs when the hypothesis emphasizes category-relevant features, even if the hypothesis is in fact incorrect. A hypothesis that assigns high salience to irrelevant features impedes learning. Learning does occur as feedback concerning category structure leads to enhanced salience for relevant features. Salience of irrelevant features remains high, however, suggesting that such learning as occurs involves augmentation and not total revision of the (incorrect) prior hypothesis.

Categorical perception of familiar objects

“…Categorization can also be acquired by selectively attending to the relevant dimension along which the categories differ (Nosofsky, 1984) or to certain features by which they differ (Livingston & Andrews, 1995). Categorization may also occur through a process called differentiation.…”

A common structure for concepts of individuals, stuffs, and real kinds: More Mama, more milk, and more mouse

“…More recently Medin et al (1993) have shown that how a dimension of variation is coded for purposes of establishing similarity depends on what portion of the dimension is available for examination. Livingston and Andrews (1995) have demonstrated that whether a particular dimension is salient for a classification may depend on how instances that vary on that dimension are presented to the concept learner. These results, among others, suggest that whether a given pair of items is identified as belonging to the same category depends upon the contrast set within which they are encountered.…”

“…The concepts we have acquired prior to that encounter, along with our goals and purposes at the time, also matter. The human capacity to control the direction of attention makes it possible to treat some members of a set to the exclusion of others, thereby changing the (psychological) context of potentially confusable alternatives and altering the relative salience and judged importance of various dimensions of variation (see, e.g., Livingston & Andrews 1995;Medin et al 1997). These selections constitute hypotheses about which groupings will be the right ones for our purposes, and are tested in various ways against facts about the consequences of deploying the concept-in-the-making in pursuit of those purposes.…”

The effects of alphabet and expertise on letter perception.

“…Another is that the feature set itself is actually fixed, but that the ranking or relative importance of features can be affected by learning and experience (e.g. Haider & Frensch, 1996; Livingston & Andrews, 1995). It is not clear how, or even if, it is possible to distinguish between these two interpretations of the flexibility view.…”