Objective: This study investigated whether improvements in working memory, reaction time, lapses of attention, interference control, academic motivation, and perceived competence mediated effects of methylphenidate on math performance. Method: Sixty-three children (ADHD diagnosis; methylphenidate treatment; age 8-13; IQ > 70) were randomly allocated to a 7-day methylphenidate or placebo treatment in this double-blind placebo-controlled crossover study and compared with 67 controls. Data were collected at schools and analyzed using mixed-model analysis. Methylphenidate was hypothesized to improve all measures; all measures were evaluated as potential mediators of methylphenidate-related math improvements. Results: Controls mostly outperformed the ADHD group. Methylphenidate did not affect measures of cognitive functioning ( p = .082-.641) or academic motivation ( p = .199-.865). Methylphenidate improved parent ratings of their child’s self-perceived competence ( p < .01), which mediated methylphenidate efficacy on math productivity. Conclusion: These results question the necessity of improvements in specific cognitive and motivational deficits associated with ADHD for medication-related academic improvement. They also stimulate further study of perceived competence as a mediator.