From Prototypes to Caricatures: Geometrical Models for Concept Typicality

Autor/inn/en	Ameel, Eef; Storms, Gert
Titel	From Prototypes to Caricatures: Geometrical Models for Concept Typicality
Quelle	In: Journal of Memory and Language, 55 (2006) 3, S.402-421 (20 Seiten) PDF als Volltext Verfügbarkeit
Sprache	englisch
Dokumenttyp	gedruckt; online; Zeitschriftenaufsatz
ISSN	0749-596X
DOI	10.1016/j.jml.2006.05.005
Schlagwörter	Comparative Analysis; Predictor Variables; Concept Formation; Mathematical Models; Geometric Concepts; Psychological Patterns; Spatial Ability; Dimensional Preference; Transformations (Mathematics) + Suchen Sie Ihr Suchwort? Prädiktor; Concept learning; Begriffsbildung; Mathematical model; Mathematisches Modell; Elementare Geometrie; Räumliches Vorstellungsvermögen; Transformationstheorie
Abstract	In three studies, we investigated to what extent a geometrical representation in a psychological space succeeds in predicting typicality in animal, natural food and artifact concepts and whether contrast categories contribute to the prediction. In Study 1, we compared the predictive value of a family resemblance-based prototype model with a geometrical centroid model in which similarity is defined in terms of distance to the centroid. The geometrical centroids outperformed family resemblance, with increasingly better predictions as the number of underlying dimensions increased. In Study 2, we investigated possible contrast effects by moving the reference point to an optimal prediction location. For most of the categories, the optimal typicality prediction was obtained by moving the prediction point away from the centroid in the opposite direction of the contrast categories, comparable to a caricature. This result strongly suggests that contrast categories play an important role in typicality. Finally, Study 3 showed that the contrast effects on typicality can be situated in the underlying similarity representation. (Author).
Anmerkungen	Elsevier. 6277 Sea Harbor Drive, Orlando, FL 32887-4800. Tel: 877-839-7126; Tel: 407-345-4020; Fax: 407-363-1354; e-mail: usjcs@elsevier.com; Web site: http://www.elsevier.com.
Erfasst von	ERIC (Education Resources Information Center), Washington, DC
Update	2017/4/10