Social cognition as segmentation of the stream of behavior.

Abstract

This study examined the influence of familiarity with peers on social cognition in children. Secondand fourth-grade children were given information about an unfamiliar peer before viewing a videotape of that peer. Children were then instructed to segment the ongoing behavioral stream of the target peer into units by pressing a button. Extrapolating from research on the influence of expert knowledge on cognition, we predicted that children who were given prior information about the peer would break the behavioral stream into fewer, larger segments than would children in no-priorinformation control conditions. We also predicted that prior information would exert more influence on the segmentation of fourth graders than that of second graders. Results were consistent with these hypotheses. These and other results are discussed in terms of developmental and contextual influences on social cognition in children. In recent years, research on children's social cognition has been dominated by cognitive-developmental theory (Higgins, Ruble, & Hartup, 1983; Shantz, 1983). The emphasis in this approach has been on intellectual development as a determinant of age-related changes in understanding the social environment. \bunger children differ from older children in their pattern of social cognition because of cognitive-structural limitations in younger children. An alternative to the cognitive-developmental approach is a contextualist account (e.g., Higgins & Eccles Parsons, 1983; Laboratory of Comparative Human Cognition, 1983; Lerner & Lerner, 1986; McGuire, 1983). This approach does not deny that in fact there may be structural bases for some age-related differences in social cognition. This approach suggests, however, that some of the age and group differences reported in the literature may reflect not structural differences in basic cognitive capacities so much as differential familiarity with certain kinds of tasks and persons as well as differences in the contexts in which the behavior is typically assessed. There are data consistent with this view. Livesley and Bromley (1973) found that children provide more complex, differentiated, inferential statements in describing other children than in describing adults. One explanation for this is that children are more familiar with the behavior of peers than with that of adults (see Shantz, 1983, p. 506). Children have more diverse interaction with other children and more experience with the range of activities in which other children engage and thus perceive children differently. Similarly, children may have more differentiated concepts of friends than of disliked peers (e.g.,