Academic journal article Memory & Cognition

Idealness and Similarity in Goal-Derived Categories: A Computational Examination

Academic journal article Memory & Cognition

Idealness and Similarity in Goal-Derived Categories: A Computational Examination

Article excerpt

Published online: 13 September 2012

© Psychonomic Society, Inc. 2012

Abstract The finding that the typicality gradient in goal-derived categories is mainly driven by ideals rather than by exemplar similarity has stood uncontested for nearly three decades. Due to the rather rigid earlier implementations of similarity, a key question has remained-that is, whether a more flexible approach to similarity would alter the conclusions. In the present study, we evaluated whether a similarity-based approach that allows for dimensional weighting could account for findings in goal-derived categories. To this end, we compared a computational model of exemplar similarity (the generalized context model; Nosofsky, Journal of Experimental Psychology. General 115:39-57, 1986) and a computational model of ideal representation (the ideal-dimension model; Voorspoels, Vanpaemel, & Storms, Psychonomic Bulletin & Review 18:1006-114, 2011) in their accounts of exemplar typicality in ten goal-derived categories. In terms of both goodness-of-fit and generalizability, we found strong evidence for an ideal approach in nearly all categories. We conclude that focusing on a limited set of features is necessary but not sufficient to account for the observed typicality gradient. A second aspect of ideal representations-that is, that extreme rather than common, central-tendency values drive typicality-seems to be crucial.

Keywords Concepts * Category representation * Computational models of representation * Typicality * Goal-derived categories * Semantic memory

(ProQuest: ... denotes formulae omitted.)

Introduction

A central topic in cognitive science is the study of the human conceptual system, containing the knowledge that people have of the world and of their culture. This knowledge is systematized in concepts-that is, the basic units of knowledge associated with a category or class of objects (e.g., Barsalou, Simmons, Barbey, & Wilson, 2003; Komatsu, 1992; Lakoff& Johnson, 1980; Medin, 1989; Murphy, 2002; E. E. Smith & Medin, 1981). Concepts are the mental representations that serve as the input to the wide variety of cognitive activities that people engage in (Barsalou et al., 2003), such as categorization and induction. Since concepts are often considered the building blocks of thought, understanding human cognition requires a solid theory on the nature of concepts (e.g., Pinker, 1997; Solomon, Medin, & Lynch, 1999).

Arguably the most dominant concept in concept research is similarity (e.g., Goldstone, 1994; Goldstone & Son, 2005; Komatsu, 1992; Markman & Gentner, 1993; Medin, Goldstone, & Gentner, 1993; Murphy, 2002; Rosch & Mervis, 1975; Sloman & Rips, 1998; L. B. Smith, 1989; Tversky, 1977; but see also Murphy & Medin, 1985). As Quine (1969, p. 114) noted, "Similarity is fundamental for learning, knowledge and thought, for only our sense of similarity allows us to order things in kinds so that these can function as stimulus meanings." Indeed, similarity is a crucial organizing principle for a large class of concepts that are world-derived. This type of concept originates from salient groupings of similar objects in the environment- that is to say, categories that derive from the correlational structure of the world (Rosch, Mervis, Gray, Johnson, & Boyes-Braem, 1976). For example, the concept raven systematizes our knowledge on ravens and is grounded on a very distinct category of objects, in that its members are roughly similar to each other, and dissimilar to other categories, such as aardvarks or arm chairs.

For some kinds of concepts, however, similarity does not seem the underlying principle. In particular, goal-derived concepts, which find their origin in the goals of a cognitive agent in a particular context, are-at least at first sight-far less dependent on correlations in the environment (Medin, Lynch, & Solomon, 2000). For example, when coming home from work, someone might construct the concept things to rescue from a burning house. …

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.