Premium
The role of typical contours in object processing by children
Author(s) -
Derȩgowski J. B.,
Parker D. M.,
Dziurawiec S.
Publication year - 1996
Publication title -
british journal of developmental psychology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.062
H-Index - 75
eISSN - 2044-835X
pISSN - 0261-510X
DOI - 10.1111/j.2044-835x.1996.tb00716.x
Subject(s) - object (grammar) , depiction , psychology , cognitive neuroscience of visual object recognition , artificial intelligence , cognitive psychology , plane (geometry) , pre school , communication , pattern recognition (psychology) , computer vision , developmental psychology , computer science , geometry , visual arts , mathematics , art
Draughtsmen depicting objects often do so by means of outlines derived in a systematic manner from the objects' surfaces. These outlines represent those points on a solid's surface where the surface appears to change rapidly; a line joining these points of maximum change is here termed a typical contour. Previous research (Dziurawiec & Deregowski, 1992) has shown that children's distorted drawings of animal models may be explained by the child's tendency to depict typical contours for different regions of an object, e.g. head, trunk, tail, and bring these together in a single drawing. The present paper extends this research to the role of typical contours in the discrimination and recognition of unfamiliar solids. In Expt 1 results collected from 108 nursery school children indicate that learning to discriminate between unfamiliar models is significantly improved when the models' typical contours are presented in the participants' fronto‐parallel plane. In Expt 2 data collected from a further 107 nursery school children established that when all views of an unfamiliar object are available, nursery school children readily abstract the typical contours and use them as an aid to subsequent recognition. The utility of the typical contours approach, previously demonstrated for object depiction, is confirmed for object recognition.