Abstract

The aim of the experiments described below was to elucidate aspects of children's mental representations of what they see, through studying their drawings of simple geometric forms when copying from standard models. Two specific questions were studied: (a) Do children produce symbolic representations based, as in machine perception, on decomposition of the visual object into features or properties, subsequently reaggregated to a greater or lesser degree, rather than attempts to copy the visual appearance of the model? The answer was affirmative, though the phenomenon is relatively rare, (b) Does a child's drawings of simple geometric drawings improve if he can watch an adult do the same drawings first? This has been asserted in the literature and bears on the question of static versus procedural representation. In a sample of 499 drawings made by children aged from 2½ to 5 years we showed that such an effect is absent or very small.