Mitchell Eric Berman - US grants

Violent behavior is of substantial social concern. Among potential biologic factors, serotonin (5-HT) neurotransmission has been widely implicated in aggressive behavior. A rich literature suggests that serotonergic (5-HT) activity is inversely associated with aggressive behavior in both human and non-human subjects. The human evidence for this relationship, however, consists almost exclusively of non-experimental research. Therefore, a causal relationship between serotonergic status and aggression in humans has not yet been clearly demonstrated. Theorists have proposed that serotonin influences aggressive behavior by altering the threshold at which an organism responds to provocative stimuli. According to this perspective, raising 5-HT levels should decrease aggressive responding to perceived provocation. At this time, however, no human experimental evidence for the interactive effects of provocation and 5-HT activity on aggression exists. The first aim of this study is to demonstrate a causal relationship between serotonin activity and aggressive behavior in humans. This aim will be addressed by determining if aggressive behavior is reduced after experimentally increasing brain levels of 5-HT. Specifically, 132 subjects (66 individuals with a history of impulsive aggression and 66 normal controls) will be randomly assigned to receive either an inactive placebo capsule or 0,5 mg/kg or 1.0 mg/kg d- fenfluramine (d-FEN). D-FEN has been shown to reliably increase indices of 5-HT functioning, decrease hostile ideation in humans, and some experimental evidence exists for the antiaggressive effects of this drug in animals. Approximately four hours after drug administration, aggressive responding will be observed using a laboratory paradign that has substantial empirical evidence supporting its validity. The second aim of the study is to determine if provocation moderates the relationship between serotonin and aggression. Therefore, provocation by an adversary during an aggressive encounter will be manipulated as a within-subjects factor. It is hypothesized that aggressive behavior will be attenuated by d-FEN compared with placebo, especially under conditions of high-provocation. This effect is expected to be greater in individuals with a history of impulsive aggressive behavior compared with normal controls.

DESCRIPTION (provided by applicant): Suicide and lesser forms of self-aggression are significant health concerns with devastating personal, social, and economic costs. Both acute alcohol intoxication and chronic alcohol use are significant risk factors for self-aggressive acts across the spectrum of lethality. Evidence for these relationships, however, is based primarily on non-experimental retrospective research. Although informative, these studies are limited in their ability to support causal inferences, and information about the respective roles of acute and chronic alcohol use are difficult to disentangle using non-experimental approaches. For these reasons, we experimentally examined the effects of acute alcohol intoxication on self-aggressive behavior in non-alcoholic individuals using a novel laboratory paradigm (McCIoskey & Berman, in press). Results indicated that consumption of a "high dose" of alcohol (mean 0.10 BAC)facilitates self-aggressive behavior in men compared to their counterparts who consumed a placebo drink. In this application, we propose two studies to examine this relationship more precisely. The primary aim of Study 1 is to extend our initial findings by conducting a dose-response determination in both women (n = 80) and men (n = 80). Participants will be randomly assigned to receive a control drink or ETOH doses designed to achieve either 0.05g, 0.75g, or 0.10g/100 ml BAL. Sixty minutes after drink consumption, participants will be provided the opportunity to self-administer electric stimulation, with self-aggression defined as the level of shock selected. We hypothesize that self-aggressive behavior will increase as a function of dose, and that women will behave similarly to men when intoxicated. Study 2 will address whether self-aggressive behavior differs as a function of ascending versus descending blood alcohol concentration (BAC) limb. Women (n = 80) and men (n = 80) will receive either a (a) control drink or (b) alcohol to achieve a 0.10 g/100 ml BAL. Self-aggressive behavior will be assessed either on the ascending or descending limb at BAC = 0.08, and compared to yoked controls. Because the ascending limb is associated with greater cognitive impairment, we expect greater levels of self-aggression at that point of the biphasic curve.