The purpose of the present study was to investigate the relationship of age trends in social cognitive skills and aggressive behaviors in normal children. It was predicted that sophistication in social cognitive skills would be negatively related to the rate of aggression and to the proportion of aggressive responses to accidental antecedent events. Age differences were also expected for the aggression measures (including physical and verbal, person-oriented and material-oriented aggression) and for the social cognitive measures Subjects were boys (from middle socioeconomic backgrounds) from kindergarten, first and third grades at two elementary schools. They were observed during free play situations at school over a 12 week period. Observations focused on aggressive events, using a time-sampling procedure and written specimen recording. Data on aggressive events was subsequently coded by the observer according to a classification system which categorized the aggressive behavior by function and the antecedent event by both function and intent (accidental, purposive, ambiguous or unknown). During the latter half of the observation phase, subjects were individually administered a battery of four tests to obtain measures of moral judgment, empathic role-taking, use of intention information, and verbal ability The major analyses of the data involved multivariate regression analyses to determine the relationship between the test measures and the aggression measures and multivariate analyses of variance to determine if grade accounted for differences in the aggression measures and the social cognitive measures separately. (Verbal ability was used as a covariate.) The results generally supported the predictions of age trends in the social cognitive test measures. However, no age trends were found in the aggression measures and no relationships were found between any aggression measure and any social cognitive measure. Various factors were suggested as possibly influencing the low rate of aggression and the lack of age trends, e.g. socioeconomic background, the age of the sample, and the observation procedure. It was proposed that a detailed theoretical framework is crucial for further investigation of developmental trends in cognitive/behavioral relationships and that more specific cognitive and behavioral measures are necessary
