Controlled and efficient processing of psychological and spatial perspectives in children and adults

Eight experiments investigated psychological and spatial perspective-taking in children and adults. Experiments were designed to identify under what circumstances and at what stage of development computing the perspectives of others is cognitively effortful and when it is cognitively efficient. Level-1 perspective-taking requires judgements of whether or not another person can see a given object. Level-2 perspective-taking requires judgements of how another person sees a given object. Using an indirect measure of perspective-taking, Experiments 1-3 found Level-1 perspective-taking to be somewhat automatic, even in children, whilst we found no evidence of automatic Level-2 perspective-taking. Using a direct measure, all forms of perspective-taking required cognitive control and were susceptible to egocentric interference. Such interference is a feature of perspective-taking of both children and adults and in Experiment 4 it was shown to be related to individual differences in executive function. In Experiments 5A and 5B children and adults made spatial frame of reference judgements, in where they integrated information from their own spatial viewpoint and that produced by the frame of reference of an object or agent. Perspective-taking makes conflicting demands for efficiency and flexibility on cognitive systems. This thesis lends support to the idea that one way these demands are resolved is through distinct systems with distinct processing features