The development of size scaling in children's figure drawings

Abstract

Size scaling in drawings made by children aged 3:6 years to 6:5 years, was investigated in three experiments. In Expt 1, children's drawings of a man were taller than their drawings of a dog (i.e. correct ordinal height scaling) and the order of making the drawings (and thus space remaining on the paper) had no effect on their height. In terms of the combined area of head and trunk, however, dog drawings were smaller than man drawings only when the man was drawn first. In Expt 2, even the youngest children on average produced correct ordinal height scaling in their drawings of topics which differed greatly in real size (house, man and dog), but the height of the dog was overestimated relative to the man and the height of the man was overestimated relative to the house. The relative overestimation of dog and man declined with increasing age. In Expt 3, topics which did not differ greatly in size (man and door) were generally not scaled correctly, but the scaling was more accurate when man and door were drawn close together in the picture. Thus, an effort towards visually correct size scaling could be detected even in very young children's drawings, but only on a size dimension which was salient and relevant for topic differentiation.