To the Editor.—
Mannuzza et al1 showed that probands with histories of childhood hyperactivity were more often antisocial and criminal at follow-up compared with men who were normal children. They reasoned on that basis that early interventions may prevent later criminality among this higher risk population. Their data raise the following three related questions: (1) Since arrest rates were not significantly different when comparisons were made between antisocial subjects with and without histories of childhood hyperactivity, and between antisocial men who did and did not exhibit persisting hyperactivity at follow-up, it seems clear that antisocial personality and not hyperactivity (either transient or lasting) predicts criminality. If, as the authors propose, the relationship between childhood hyperactivity and later criminality is "almost entirely mediated by antisocial disorders in young adulthood," what are these intervening processes and how are they connected to childhood hyperactivity? (2) Among antisocial men, those who were