Academic journal article Psychonomic Bulletin & Review

Featural Selective Attention, Exemplar Representation, and the Inverse Base-Rate Effect

Academic journal article Psychonomic Bulletin & Review

Featural Selective Attention, Exemplar Representation, and the Inverse Base-Rate Effect

Article excerpt

Selective attention plays a central role in theories of category learning and representation. In exemplar theory, selective attention has typically been formalized as operating uniformly across entire stimulus dimensions. Selective featural attention operating within dimensions has been recognized as a conceptual possibility, but relatively little research has focused on evaluating it. In the present research, we explored the usefulness of selective featural attention in the context of exemplar representation. We report the results of embedding the feature-to-category relations typically associated with the inverse base-rate effect-a classic and paradoxical category-learning result-within a perceptual category-learning task using a category structure with three multivalued feature dimensions. An exemplar model incorporating featural selective attention accurately accounted for the inverse base-rate effect that occurred but failed to do so with only dimensional attention.

(ProQuest: ... denotes formulae omitted.)

Assigning objects to categories and then adaptively using those categories is one of the foundational components of cognition. Exemplar models (Kruschke, 1992; Medin & Schaffer, 1978; Nosofsky, 1986) have been very successful in accounting for human categorization, especially in perceptual categorization tasks in which participants learn to categorize stimuli with features from multivalued feature dimensions. Exemplar models represent categories by storing category instances, and new items are categorized on the basis of their similarity to the stored exemplars. One reason that exemplar models have been so successful is that they include selective attention mechanisms that give more attention to the aspects of the stimuli that differentiate categories and less attention to those aspects that do not. Selective attention changes the perceived similarity between the category instances in a way that makes instances within a category more similar to each other than to instances in other categories. This facilitates both category learning and use.

For perceptual category-learning tasks with stimuli composed of features from multivalued feature dimensions (Figure 1), there are two basic kinds of selective attention: dimensional and featural. Dimensional attention allows some stimulus dimensions to influence responding more than others. For example, the rectangle height dimension in Figure 1 might influence responding more than the circle diameter dimension because the rectangle dimension is more important for differentiating category instances. However, the key aspect of dimensional attention is that all values along the dimension (i.e., features) must be attended equally-for example, the high, medium, and low rectangles in Figure 1. In contrast, featural selective attention allows different features within a given dimension to be differentially influential-for example, the high rectangle more than the low one.

The dominant exemplar models have emphasized dimensional rather than featural attention (the context model, Medin & Schaffer, 1978; the generalized context model [GCM], Nosofsky, 1986; ALCOVE, Kruschke, 1992). But, although the potential for featural selective attention has been acknowledged conceptually, little research has focused on directly evaluating its usefulness in the context of exemplar representation.

Aha and Goldstone's (1990, 1992) addition of contextsensitive, feature-specific attention to Nosofsky's (1986) GCM is the most well-known formalization of featural selective attention in an exemplar model. They showed that this additional mechanism allowed the model to learn faster and fit the results of several categorization experiments better than it did with only dimensional attention. In addition, Medin and Edelson (1988) and Nosofsky, Clark, and Shin (1989) explored exemplar variants of featurespecific attention, and there have recently been some interesting eyetracking studies of featural attention (Blair, Watson, Walshe, & Maj, 2009). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.