Prior research has found testosterone related to individual differences in aggression, dominance, and criminal violence. This research will gather detailed information on testosterone and antisocial behavior in delinquent and normal populations. It will also test a biosocial model that predicts antisocial behavior as a joint function of high testosterone and low social control. Antisocial behavior ranges in severity from unfriendly facial expressions to unstable personal relationships to criminal violence. All these behaviors are correlated with testosterone, but the correlations are low, and large numbers of subjects are needed to have enough statistical power to determine their exact size. Other factors are needed to explain why high testosterone individuals are not always violent. The model proposed here treats social control forces, embodied in attachment to family and community, as restraining the more violent effects of testosterone. Most high testosterone individuals engage in mild antisocial behavior, especially when frustrated or challenged, but are violent only when social controls are weak. The testosterone-violence correlation should be strongest when social control is weakest, as when subjects do not identify with family and community or when they are young, mentally incompetent, or under the influence of drugs or alcohol. The research will employ salivary testosterone measurements and behavioral observations of prison inmates, college students, military veterans, children, and alcoholics. The findings will help to specify exactly what antisocial behaviors are related to testosterone, how strong the relationship is, and what other factors serve to increase or decrease the size of the relationship.
