The emergence of spatial perspective taking: an information-processing alternative to egocentrism.

120 children between ages 4 and 8 were administered a set of 4 visual perspective-taking tasks similar to ones employed in previous investigations. It was hypothesized that children's competence with the tasks would be a function of the number and type of spatial relationships embedded in the stimulus displays. 2 kinds of spatial relationships were described: (1) internal relationships--those between objects within a display; and (2) external relationships--those between the display and a frame of reference. Of the 4 task types, type A tasks necessitated awareness of possible variations in external relations, types B and C tasks required both awareness and computation of changes in those external relationships, and type D tasks required simultaneous coordination of internal and external spatial relationships. It was predicted that the tasks would sequence in terms of difficulty level as A less than B, B = C, and C less than D. Analysis of variance results on the obtained data revealed a good fit with the hypothesized predictions. The findings were discussed in terms of an information-processing analysis of children's performance on perspective-taking tasks.