While both social information processing and cortisol secretion in childhood aggression have generated a great deal of interest and research in the past few decades, these social-cognitive and physiological components of aggressive behavior have not been examined in the context of an integrative model. This lack of an integrative framework may underlie some of the inconsistencies that have plagued the literature in this area to date, especially with respect to hypothalamic-pituitary-adrenal (HPA) axis functioning in aggressive children. This investigation tested a mediational model of the relationship between social-information processing, cortisol secretion, and reactive and proactive aggression. Specifically, it was hypothesized that social-information processing variables would mediate the proposed relationship between reactive and proactive aggression and cortisol secretion. One hundred and twenty-six children between the ages of 13 and 18 were administered the Child Behavior Rating Form (CBR), the Home Interview with Child (HIC), the Response Decision and Social Goals Instrument (RDSGI), the Antisocial Processes Screening Device (APSD), the Buss-Durkee Hostility Inventory (BDHI), the Children's Depression Inventory (CDI), and the Revised Children's Manifest Anxiety Scale (RCMAS). Each child also contributed two samples of saliva for cortisol assay, and each child's teacher completed a teacher-version of the APSD and the CBR. Regression analyses revealed no significant associations between proactive or reactive aggression and cortisol secretion, or between any of the social-information processing variables and cortisol secretion. Predicted associations between proactive and reactive aggression and social-information processing variables were found. Overall, therefore, the mediational model was not supported. However, cortisol secretion was found to be associated with both anxiety and depression, and exploratory analyses revealed significant associations between cortisol secretion and Psychopathy as measured by the APSD. Taken together, the findings suggest that while the specific relationship proposed here among social-cognitive, psychophysiological, and behavioral variables was not found, an integrative model examining each of these components may be useful in further investigations of the complex phenomenon of childhood aggression.