Douglas L. Medin

Context theory of classification learning

Most theories dealing with ill-defined concepts assume that performance is based on category level information or a mixture of category level and specific item information. A context theory of classificatio n is described in which judgments are assumed to derive exclusively from stored exemplar information. The main idea is that a probe item acts as a retrieval cue to access information associated with stimuli similar to the probe. The predictions of the context theory are contrasted with those of a class of theories (including prototype theory) that assume that the information entering into judgments can be derived from an additive combination of information from component cue dimensions. Across four experiments using both geometric forms and schematic faces as stimuli, the context theory consistently gave a better account of the data. The relation of the context theory to other theories and phenomena associated with ill-defined concepts is discussed in detail. One of the major components of cognitive behavior concerns abstracting rules and forming concepts. Our entire system of naming objects and events, talking about them, and interacting with them presupposes the ability to group experiences into appropriate classes. Young children learn to tell the difference between dogs and cats, between clocks and fans, and between stars and street lights. Since few concepts are formally taught, the evolution of concepts from experience with exemplars must be a fundamental learning phenomenon. The focus of the present article is to explore how such conceptual achievements emerge from individual instances.

The role of theories in conceptual coherence

The question of what makes a concept coherent (what makes its members form a comprehensible class) has received a variety of answers. In this article we review accounts based on similarity, feature correlations, and various theories of categorization. We find that each theory provides an inadequate account of conceptual coherence (or no account at all) because none provides enough constraints on possible concepts. We propose that concepts are coherent to the extent that they fit peoples background knowledge or naive theories about the world. These theories help to relate the concepts in a domain and to structure the attributes that are internal to a concept. Evidence of the influence of theories on various conceptual tasks is presented, and the possible importance of theories in cognitive development is discussed.

Respects for similarity.

This article reviews the status of similarity as an explanatory construct with a focus on similarity judgments. For similarity to be a useful construct, one must be able to specify the ways or respects in which two things are similar. One solution to this problem is to restrict the notion of similarity to hard-wired perceptual processes. It is argued that this view is too narrow and limiting. Instead, it is proposed that an important source of constraints derives from the similarity comparison process itself. Both new experiments and other evidence are described that support the idea that respects are determined by processes internal to comparisons

Concepts and Conceptual Structure

Research and theory on categorization and conceptual structure have recently undergone two major shifts. The first shift is from the assumption that concepts have defining properties (the classical view) to the idea that concept representations may be based on properties that are only characteristic or typical of category examples (the probabilistic view). Both the probabilistic view and the classical view assume that categorization is driven by similarity relations. A major problem with describing category structure in terms of similarity is that the notion of similarity is too unconstrained to give an account of conceptual coherence. The second major shift is from the idea that concepts are organized by similarity to the idea that concepts are organized around theories. In this article, the evidence and rationale associated with these shifts are described, and one means of integrating similarity-based and theory-driven categorization is outlined.

SUSTAIN: a network model of category learning.

SUSTAIN (Supervised and Unsupervised STratified Adaptive Incremental Network) is a model of how humans learn categories from examples. SUSTAIN initially assumes a simple category structure. If simple solutions prove inadequate and SUSTAIN is confronted with a surprising event (e.g., it is told that a bat is a mammal instead of a bird), SUSTAIN recruits an additional cluster to represent the surprising event. Newly recruited clusters are available to explain future events and can themselves evolve into prototypes-attractors-rules. SUSTAINs discovery of category substructure is affected not only by the structure of the world but by the nature of the learning task and the learners goals. SUSTAIN successfully extends category learning models to studies of inference learning, unsupervised learning, category construction, and contexts in which identification learning is faster than classification learning.

Sinning Saints and Saintly Sinners The Paradox of Moral Self-Regulation

The question of why people are motivated to act altruistically has been an important one for centuries, and across various disciplines. Drawing on previous research on moral regulation, we propose a framework suggesting that moral (or immoral) behavior can result from an internal balancing of moral self-worth and the cost inherent in altruistic behavior. In Experiment 1, participants were asked to write a self-relevant story containing words referring to either positive or negative traits. Participants who wrote a story referring to the positive traits donated one fifth as much as those who wrote a story referring to the negative traits. In Experiment 2, we showed that this effect was due specifically to a change in the self-concept. In Experiment 3, we replicated these findings and extended them to cooperative behavior in environmental decision making. We suggest that affirming a moral identity leads people to feel licensed to act immorally. However, when moral identity is threatened, moral behavior is a means to regain some lost self-worth.

Family resemblance, conceptual cohesiveness, and category construction

Abstract Many people have argued that natural categories are organized in terms of a family resemblance principle. Members of family resemblance categories tend to share properties with each other but have no properties that are singly necessary and jointly sufficient (defining) for category membership. This paper reports seven experiments using a sorting task to evaluate the conditions under which people prefer to construct categories according to a family resemblance principle. The first set of studies followed the typical practice of defining family resemblance in terms of independent sets of matching and mismatching values. Across a variety of stimulus materials, instructions, procedures, and category structures, family resemblance sorting was almost never observed. Despite procedures designed to prevent it, participants persisted in sorting on the basis of a single dimension. The second set of studies explored the idea that interproperty relationships rather than independent features serve to organize categories. We found that people will abandon unidimensional sorting in favor of sorting by correlated properties, especially when they can be causally connected. In addition, when conceptual knowledge is added which makes interproperty relationships salient, family resemblance sorting becomes fairly common. Implications of the results for the development of family resemblance categories and the practice of treating properties or features as additive and independent are discussed.

Context and structure in conceptual combination

Three experiments evaluated modifications of conceptual knowledge associated with judgments of adjective-noun conceptual combinations. Existing models, such as the Smith and Osherson modification model, assume that the changes associated with understanding an adjective noun combination are confined to the corresponding adjectival dimension. Our experiments indicate that this assumption is too strong. The first study found that naming one dimension affects correlated dimensions. For example, participants judge small spoons to be more typical spoons than large spoons, but for wooden spoons, large spoons are more typical than small spoons. The second study demonstrated that the similarity of adjectives is not independent of the noun context in which they appear. For example, white and gray are judged to be more similar than gray and black in the context of hair but this judgment reverses in the context of clouds. The third study showed that a property equally true (or false) for two concepts may be more central to one concept than the other (e.g., it is more important that boomerangs be curved than that bananas be curved). These results pose serious problems for current theories of how people combine concepts. We propose instead that we need richer views of both the conceptual structure and the modifications of it required by conceptual combination. We suggest that theoretical knowledge and the construct of centrality of meaning may play useful roles.

Categorization and Reasoning among Tree Experts: Do All Roads Lead to Rome?☆

To what degree do conceptual systems reflect universal patterns of featural covariation in the world (similarity) or universal organizing principles of mind, and to what degree do they reflect specific goals, theories, and beliefs of the categorizer? This question was addressed in experiments concerned with categorization and reasoning among different types of tree experts (e.g., taxonomists, landscape workers, parks maintenance personnel). The results show an intriguing pattern of similarities and differences. Differences in sorting between taxonomists and maintenance workers reflect differences in weighting of morphological features. Landscape workers, in contrast, sort trees into goal-derived categories based on utilitarian concerns. These sorting patterns carry over into category-based reasoning for the taxonomists and maintenance personnel but not the landscape workers. These generalizations interact with taxonomic rank and suggest that the genus (or folk generic) level is relatively and in some cases absolutely privileged. Implications of these findings for theories of categorization are discussed.

The Native Mind: Biological Categorization and Reasoning in Development and Across Cultures.

This article describes cross-cultural and developmental research on folk biology: that is, the study of how people conceptualize living kinds. The combination of a conceptual module for biology and cross-cultural comparison brings a new perspective to theories of categorization and reasoning. From the standpoint of cognitive psychology, the authors find that results gathered from standard populations in industrialized societies often fail to generalize to humanity at large. For example, similarity-driven typicality and diversity effects either are not found or pattern differently when one moves beyond undergraduates. From the perspective of folk biology, standard populations may yield misleading results because they represent examples of especially impoverished experience with nature. Certain phenomena are robust across populations, consistent with notions of a core module.