This project represents phase one in a planned series of studies relating saliva testosterone concentrations to antisocial and violent behavior. To date, studies relating plasma testosterone to aggressive, impulsive, and antisocial behavior have produce conflicting results and suggest the effects of testosterone may be very specific. For example, different studies show serum testosterone related to provoked but not unprovoked aggression and to adolescent but not adult criminal history. There is little information from these studies on how much testosterone changes over time, what causes the changes, how they affect aggressive behavior, and how they might make it difficult to characterize a person as chronically high or low in testosterone. The present project explores the utility of saliva rather than serum measures of testosterone in future research on antisocial and violent behavior. Saliva provides a virtually pure measure of """"""""free"""""""" testosterone, while serum provides a measure of free plus bound. Saliva can be collected easily with no medical assistance, making it possible to study large numbers of subjects under diverse field conditions and to take repeated measures over periods of time ranging from minutes to years. Preliminary studies show saliva testosterone related to criminal violence in both male and female prison inmates and to personality measures in college students. This project will examine cyclicity (circadian, menstrual, circannual) in order to control for error that cycles can introduce into measures of individual differences in testosterone and to determine the best time of day to obtain measures. It will also examine changes in testosterone due to random variation, life events over a six month period, and experimental testing conditions. Preliminary correlative data on delinquent behavior will be collected, but the focus will be upon developing protocols for measuring individual differences in saliva testosterone.
