conducted. Forty years of research has assumed that category learning often involves learning to selectively attend to only those stimulus dimensions useful for classification. We confirmed that participants learned to allocate their attention optimally. We also found that learners tend to fixate all stimulus dimensions early in learning. This result obtained despite evidence that participants were also testing one-dimensional rules during this period. Finally, the restriction of eye movements to only relevant dimensions tended to occur only after errors were largely (or completely) eliminated. We interpret these findings as consistent with multiple-systems theories of learning which maximize information input in order to maximize the number of learning modules involved, and which focus solely on relevant information only after one module has solved the learning problem.

to all of the category's features. However, people's knowledge of real-world categories often consists of many "rote " features that are not related to their prior knowledge. Five experiments found that even minimal prior knowledge (1 knowledge-relevant feature and 5 rote features per exemplar) can facilitate category learning. Posttests revealed that although the knowledge aided learning, subjects also acquired the rote features that were not related to knowledge, contradicting predictions of an attentional expla-nation of the knowledge effect. The results of Experiment 6 suggested that subjects attempt to link even rote features to their knowledge.

Several theories have been proposed regarding how causal relations among features of objects affect how those objects are classified. The assumptions of these theories were tested in 3 experiments that manipulated the causal knowledge associated with novel categories. There were 3 results. The 1st was a multiple cause effect in which a feature’s importance increases with its number of causes. The 2nd was a coherence effect in which good category members are those whose features jointly corroborate the category’s causal knowledge. These 2 effects can be accounted for by assuming that good category members are those likely to be generated by a category’s causal laws. The 3rd result was a primary cause effect, in which primary causes are more important to category membership. This effect can also be explained by a generative account with an additional assumption: that categories often are perceived to have hidden generative causes.

Disentangling bottom-up and top-down processing in adult category learning is notoriously difficult. Studying category learning in infancy provides a simple way of exploring category learning while minimizing the contribution of top-down information. Three- to 4-month-old infants presented with cat or dog images will form a perceptual category representation for cat that excludes dogs and for dog that includes cats. The authors argue that an inclusion relationship in the distribution of features in the images explains the asymmetry. Using computational modeling and behavioral testing, the authors show that the asymmetry can be reversed or removed by using stimulus images that reverse or remove the inclusion relationship. The findings suggest that categorization of nonhuman animal images by young infants is essentially a bottom-up process. Few in cognitive science would dispute the argument that both bottom-up (i.e., perceptually driven) and top-down (i.e., conceptually driven) processes are involved in adult categorization. Numerous studies have discussed the relationship between these two mechanisms of categorization (e.g., French, 1995; Murphy & Kaplan, 2000; Schyns, Goldstone, & Thibaut, 1998). However, in adults, perceptual and conceptual processes are deeply intertwined, making them difficult to isolate and study independently (Goldstone & Barsalou, 1998).

The present research explored the role of working memory (WM) in unsupervised category learning, learning without an external tutor or even knowing that categories exist, by investigating its role using a pattern-sequence manipulation. A pattern-sequence manipulation compares learning when items from categories are presented together (blocked) versus when the items are presented in random order (mixed). Experiment 1 extended the pattern-sequence manipulation to assess category knowledge separate from paired-associate learning. Participants performed equally well on new and studied items, supporting the hypothesis that the pattern-sequence manipulation results in the acquisition of category information, not simply memory for item-feature associations. Experiment 2 introduced a WM factor, administering the method used in Experiment 1 to a group of high and low WM span participants. High WM span was predicted to interact with the pattern-sequence effect to produce greater learning when the items were blocked than mixed. There was reliable support for a role of WM span in the discovery and acquisition of category knowledge, but this role was different from the one hypothesized. The high WM span participants exhibited higher overall accuracies than the low WM span participants. This result supports a role for WM in unsupervised category learning, but did not benefit more from the pattern-sequence effect than did the low WM span participants as predicted. Implications for theories of category learning and WM are discussed.

iderable thematic or complexive responding were based on criteria that now seem very unclear (e.g., Olver &Hornsby, 1966). Finally, the study of word learning has yielded much evidence of children's ability to acquire taxonomic categories, especially at the basic level, even in the acquisition of their first words (Huttenlocher & Smiley, 1987; Markman, 1989; Rosch, Mervis, Gray, Johnson, & Boyes-Braem, 1976). In short, although it seems clear that children do not always perform very well in these tasks, the suggestion that they cannot understand taxonomic categories is not correct. The second part of the thematic-to-taxonomic shift hypothesis has not received as much attention, however. In fact, the argument that children can use taxonomic categories implies even more strongly that adults will do so, since this form of categorization is thought to be more powerful and advanced (Markman & Callanan, 1983). Furthermore, adults have often shown taxonomic sorting in tasks in which children

Subtyping as a knowledge preservation

Two experiments examined the impact of causal relations between features on categorization by adults and 5-6-year-old children. Participants learned about artificial categories containing instances with two causally related features and two non-causal features. They then selected the most likely category member from a series of novel test pairs. Classification patterns and logistic regression were used to diagnose the presence of independent effects of causal coherence, causal status and relational centrality. Adult classification was driven primarily by coherence when causal links were deterministic (Experiment 1), but showed additional influences of causal status and centrality when links were probabilistic (Experiment 2). Children’s classification was based primarily on causal coherence in both cases. These results suggest that the generative model [Rehder, B. (2003). A causalmodel theory of conceptual representation and categorization. Journal of Experimental Psychology: Learning, Memory, and Cognition, 29, 1141-1159] provides a good account of causal categorization in both children and adults. Children’s Causal Categorization 3 It is well established that causal knowledge plays an important role in adult categorization and

Research has shown that category learning is affected by (a) attention, which selects which aspects of stimuli are available for further processing, and (b) the existing semantic knowledge that learners bring to the task. However, little is known about how knowledge affects what is attended. Using eyetracking, we found that (a) knowledge indeed changes what features are attended, with knowledgerelevant features being fixated more often than irrelevant ones, (b) this effect was not due to an initial attentional bias toward relevant dimensions but rather emerged as a result of observing category members, and (c) this effect grew even after a learning criterion was reached, that is, despite the absence of error feedback. We argue that models of knowledge-based learning will remain incomplete until they include mechanisms that dynamically select prior knowledge in response to observed category members and which then directs attention to knowledge-relevant dimensions and away from irrelevant ones. Knowledge and Attention in Category Learning 3 How Prior Knowledge Affects Selective Attention During Category Learning:

Exemplar sequencing effects in incidental and intentional unsupervised category learning were investigated to illuminate how people form categories without an external teacher. Stimuli were perfectly separable into 2 categories based on 1 of 2 dimensions of variation. Sequencing of the first 20 training stimuli was manipulated. In the blocked condition, 10 Category A stimuli were followed by 10 Category B stimuli. In the intermixed condition, these 20 stimuli were ordered randomly. Experiment 1 revealed an interaction between learning mode and sequence, with better intentional learning for intermixed sequences but better incidental learning for blocked sequences. Experiment 2 showed that manipulating trial-to-trial variability along each dimension can impact intentional learning. Training sequences that emphasized variation along the category-relevant dimension resulted in better performance than sequences that emphasized variation along the category-irrelevant dimension. The results suggest that unsupervised category learning is influenced by the mode of learning and the order and nature of encountered exemplars.