Thomas L. Spalding

Comparison-Based Learning: Effects of Comparing Instances During Category Learning

When learning about a category, people often compare new instances with similar old instances and notice features common to the compared instances. Five experiments demonstrate that such comparisons cause features common to compared instances to be considered more important for the category than equally frequent features that are not common to compared instances. Experiment 1 shows that what is learned depends on which instances are compared. Experiment 2 investigates the conditions under which comparison-based learning occurs. The next experiments find that these comparisons affect subjective feature frequency (Experiment 3) and sensitivity to feature correlations (Experiment 4). Experiment 5 shows that comparisons during early learning affect what is learned from later instances. The discussion focuses on the implications for models of category representation.

Effect of Relation Availability on the Interpretation and Access of Familiar Noun-Noun Compounds.

Two experiments investigate whether relations that link the constituents of compounds during compound formation (e.g., teapot is formed by combining tea and pot using the relation head noun FOR modifier) also influence the processing of familiar compounds. Although there is evidence for the use of such relations in forming compounds, whether such relations affect the processing of familiar compounds is unknown. The data show clear effects of repetition and relational priming for written words on both a sense-nonsense task and a lexical decision task. These results indicate that the relation linking the constituents of familiar compounds is important to their access and use.

What is learned in knowledge-related categories? Evidence from typicality and feature frequency judgments

When a category’s features are tied together by integrative knowledge, subjects learn the category faster than when the features are not directly related. What do subjects learn about the category in such circumstances? Some research has suggested that the subjects can use the knowledge itself in performing the category learning task and, thus, do not learn the details of the category’s features. Two experiments investigated this hypothesis by collecting feature frequency estimates after category learning. The results showed that integrative knowledge about a category did not decrease subjects’ sensitivity to feature frequency—if anything, knowledge improved it. A third experiment found that integrative knowledge did reduce sensitivity to feature frequency in typicality ratings. The results suggest that knowledge does not inhibit the learning of detailed category information, though it may replace its use in some tasks.

Concepts and Categories

Publisher Summary Research on concepts and categories is progressing rapidly in many directions. Given the great amount of work in the field, any review of concepts and categories has to select from many issues, areas, and results. This chapter explains the distinction between classification and other uses of concepts. A consideration of intelligent behavior requires a better understanding of classification and also of how the classification might be used. Although much of the work has focused on classification, there is a trend toward examining other functions of concepts. In addition, this distinction forces one to consider if and how the knowledge underlying classification might be related to the knowledge used for the other functions. In problem solving, categories play an increasingly important role. Many researchers believe that good problem solving requires people to develop problem categories and associated procedures for identifying problems of that type.

Infants Can Rapidly Form New Categorical Representations

Young infants learn common categorical distinctions, such asanimals versusvehicles. But can they, like adults, rapidly form new categories, such asblack-and-white animals? To answer this question, 6-, 10-, and 13-month-old infants were familiarized with four land animals that were black and white in coloring (e.g., a zebra and a black-and-white tiger) and then were tested with novel animals and a truck. The infants responded to an exclusive category that apparently included only black-and-white animals, suggesting that they formed a new categorical representation during familiarization. A comparison group of infants familiarized with a set of land animals that were more variable in coloring (e.g., a pale yellow horse and a yellow-and-brown tiger) formed a very general categorical representation that included many different kinds of animals, regardless of coloring. Therefore, like adults, infants rapidly form new categorical representations in response to the context.

Preferences for ascending and descending hierarchical organization in spatial communication.

People prefer to order spatial information in a hierarchy of decreasing size of spatial unit when giving directions for finding objects and in a hierarchy of increasing size of spatial unit when providingescriptions of object locations (Plumert, Carswell, DeVet, & Ihrig, 1995). In five experiments, we examined whther people have a preference for ascending or descending organization when the task does not involve conveying spatial information to others. In Experiments 1-3, people learned the locations of objects in a model house and then verified statements describing those locations. People verified statements faster when spatial units were organized in an ascending (i.e., small to large) than in a descending (i.e., large to small) or random order. In Experiment 4, people first performed a sentence verification task and afterward wrote down directions for finding the objects. People again exhibited a preference for ascending organization in the verification task but exhibited a preference for descending organization when giving directions for finding the same objects. Experiment 5 demonstrated that the ascending advantage was not due to the link between the object and small landmark. Discussion focuses on the role of pragmatics and memory retrieval in preferences for ascending versus descending hierarchical organization.

Concept learning and feature interpretation.

Models of categorization often assume that people classify new instances directly on the basis of the presented, observable features. Recent research, however, has suggested that the coherence of a category may depend in part on more abstract features that can link together observable features that might otherwise seem to have little similarity. Thus, category learning may also involve the determination of the appropriate abstract features that underlie a category and link together the observable features. We show in four experiments that observable features of a category member are often interpreted as congruent with abstract features that are suggested by observable features of other highly available category members. Our discussion focuses on the implications of these findings for future research.

Relation availability was not confounded with familiarity or plausibility in Gagné and Shoben (1997): Comment on Wisniewski and Murphy (2005)

C. L. Gagné and E. Shoben (1997) proposed that the conceptual system contains information about how concepts are used to modify other concepts and that this relational information influences the ease with which concepts combine. Recently, E. J. Wisniewski and G. L. Murphy suggested that C. L. Gagné and E. Shobens measure of relation availability was confounded with familiarity and plausibility and that the participants could simply retrieve the stored meanings of the phrases because the phrases were not novel. In this article, the authors demonstrate that E. J. Wisniewski and G. L. Murphys plausibility and familiarity judgments are dependent variables that (a) are themselves responsive to changes in relation availability, (b) modifier relation availability predicts response time even when the influence of phrase familiarity and plausibility is controlled, and (c) the materials consisted of mainly novel phrases.

Sentential Context and the Interpretation of Familiar Open-Compounds and Novel Modifier-Noun Phrases.

Two experiments investigated the influence of sentential context on the relative ease of deriving a particular meaning for novel and familiar compounds. Experiment 1 determined which of two possible meanings was preferred for a set of novel phrases. Experiment 2 used both novel (e.g., brain sponge) and familiar compounds (e.g., bug spray). The compounds appeared in a sentential context that supported either the dominant or subdominant meaning. Next, participants saw either the dominant or subdominant definition and indicated whether it was plausible. When the definition was consistent with the preceding sentence, the participants were more likely to consider the definition plausible regardless of whether the compound was novel or familiar, although this difference was more pronounced for novel phrases than for familiar phrases. In terms of response times, the effect of sentential context also depended on the degree of dominance. The data suggest that the interpretation of compounds is affected by at least two sources of information: sentential context and the relative dominance of the preferred meaning.

Conceptual Composition: The Role of Relational Competition in the Comprehension of Modifier-Noun Phrases and Noun–Noun Compounds

Abstract Compositionality and productivity, which are the abilities to combining existing concepts and words to create new concepts and phrases, words, and sentences, are hallmarks of the human conceptual and language systems. Combined concepts are formed within the conceptual system and can be expressed via modifier-noun phrases (e.g. purple beans ) and compound words (e.g. snowball ), which are the simplest forms of productivity. Modifier-noun phrases and compound words are often paraphrased using a relation to connect the constituents (e.g. beans that are purple, ball made of snow ). The phrase or compound does not explicitly contain the underlying relation, but the existence of the relation can be shown by manipulating the availability of the relation and observing the effect on the interpretation of the phrase or compound. This chapter describes how novel modifier-noun phrases and established compounds are interpreted. We present a theoretical account of relational interpretation of combined concepts and present the empirical evidence for the use of relational structures. We then present the empirical evidence supporting our theoretical account’s specific predictions about how relational interpretations are selected and evaluated and how the relational interpretation is elaborated to create a fully specified new concept.