This study evaluated the effects of antecedent physical exercise (AE) and a mastery task on behaviorally disturbed children's self-concepts and rates of disruptive behaviors. In addition, we evaluated whether changes in self-concept mediated any exercise induced changes in rates of disruptive behavior. Fifty-eight children were randomly assigned to one of three conditions: (a) antecedent exercise (jog/walk), (b) “mastery” task (a successive improvement miniaturebasketball shooting task), and (c) no treatment control group. A week of baseline was followed by 4 weeks of treatment and, finally another week of return to baseline. Results indicated that: (a) AE produced significantly less disruptive behavior than no treatment, (b) the “Mastery” task did not produce significantly less disruptive behavior than no treatment, (c) neither treatment produced increases in self-concept relative to no treatment, and (d) changes in self-concept did not mediate AE induced reductions in disruptive behaviors.