Several types of evidence suggest that androgens exert clinically significant effects on central nervous system (CNS) function. In animal studies, androgens modulate brain serotonin function, and regulate sexual behavior, aggression, and other non-reproductive behaviors. These actions reflect both organizational and activational effects of androgens. Several studies in humans also support the central modulatory capacity of androgens. Correlative studies have described relationships between plasma androgen levels and sexual interest/behavior and cognitive task performance. Androgens are believed to underlie gender related differences in the prevalence of certain neuropsychiatric disorders, resulting in trials of anti-androgens in at least one of these disorders (Tourette's syndrome). Finally, androgens are believed to possess psychotropic effects in humans, evidenced by purported antidepressant effects and reports of psychotic reactions following administration of androgens. In a recent study, we demonstrated that androgenic/anabolic steroids precipitated mood and behavioral state disturbances when administered in a double-blind, placebo-controlled fashion in normal volunteers. There are remarkably few studies that attempt to identify the CNS effects of androgens or the central systems that may mediate these effects. In this study we propose to examine directly the effects of testosterone on CNS activity by first suppressing and then, in a double-blind, placebo-controlled fashion, replacing physiological levels of testosterone. We will evaluate mood, behavior, cognitive function, physiologic response to serotonergic agonists and cerebral blood flow (separate protocols) during both pharmacologically controlled hormonal conditions: hypogonadism and hypogonadism with testosterone replacement. On the basis of prior findings from our group and from others, we will be asking the following questions: 1) Does cognitive function differ as a function of gender or of hormonal condition; 2) Is the decreased r-CBF that we observed in the prefrontal cortex during the hypogonadal state in women also demonstrable in men; 3) Do measures of hypothalamic-pituitary-adrenal axis responsivity differ as a function of gender or of hormonal condition; and 4) Do CSF neurochemistry and neurosteroid levels differ as a consequence of changing hormonal state. This protocol will not only provide much needed information about the behavioral and physiological effects of androgens but will serve as a companion study for NIMH protocol #92-M-172, "The Central Nervous System Effects of Pharmacologically Induced Hypogonadotropic Hypogonadism with and without Estrogen and Progressive Replacement."