Objective: The primary goal of the present research was to determine whether retrospective report of childhood disciplinary experiences and perceptions correspond to actual events and whether the report was influenced by affective state. Method: Eighty-three adolescent and young adult males completed measures of physical child maltreatment and current adjustment. The participants had been observed naturalistically in their homes at an average of 10 years earlier. Results: Analyses were consistent with the hypothesis that prevailing mood and observation of parent-child interactions during childhood predict self-reported recollections of maltreatment. Further, veridicality appears to depend upon the objective specificity versus the perceptive nature of the questions. Conclusions: The findings suggest that instruments designed to measure earlier childhood victimization must utilize behaviorally-specific items. Thus, items that are either global or intimate normative comparison should be avoided.