Understanding Genetic Risk for Aggression: Clues From the Brain’s Response to Social Exclusion Naomi I. Eisenberger, Baldwin M. Way, Shelley E. Taylor, William T. Welch, and Matthew D. Lieberman Background: Although research indicates a relationship between the monoamine oxidase-A (MAOA) gene and aggression, the interven- ing neural and psychological mechanisms are unknown. Individuals with the low expression allele (MAOA-L) of a functional polymorphism in the MAOA gene might be prone to aggression because they are socially or emotionally hyposensitive and thus care less about harming others or because they are socially or emotionally hypersensitive and thus respond to negative social experiences with defensively aggressive behavior. Methods: We investigated the relationships between the MAOA polymorphism, trait aggression, trait interpersonal hypersensitivity, and neural responses to social exclusion in 32 healthy men and women. Results: The MAOA-L individuals (men and women) reported higher trait aggression than individuals with the high expression allele (MAOA-H). The MAOA-L individuals reported higher trait interpersonal hypersensitivity and showed greater dorsal anterior cingulate cortex (dACC) activity (associated with rejection-related distress) to social exclusion compared with MAOA-H individuals, consistent with a social hypersensitivity hypothesis. Moreover, the MAOA–aggression relationship was mediated by greater dACC reactivity to social exclusion, suggesting that MAOA might relate to aggression through socioemotional hypersensitivity. Conclusions: These data suggest that the relationship between MAOA and aggression might be due to a heightened rather than a reduced sensitivity to negative socioemotional experiences like social rejection. Key Words: Aggression, dorsal anterior cingulate cortex, fMRI, in- terpersonal sensitivity, MAOA gene, MAOA-uVNTR, neuroimaging, social exclusion I n both animal and human populations, aggressive behavior has been linked to a genetic deficiency in monoamine oxidase-A (MAOA), an enzyme that degrades serotonin, dopamine, and norepinephrine (Shih et al. 1999). Monoamine oxidase-A– deficient male mice were found to be more aggres- sive as evidenced by a shorter latency to attack and a greater number of skin wounds in a resident-intruder paradigm (Cases et al. 1995). Monoamine oxidase-A– deficient men from a single Dutch kindred demonstrated elevated levels of impulsive aggres- sion, arson, and attempted rape (Brunner et al. 1993). In line with these findings, when exposed to early adversity, men with the low expression allele (MAOA-L) of the 30-base pair (bp) variable number tandem repeats polymorphism in the MAOA promoter (MAOA-uVNTR) were more likely to develop antisocial behavior than men with the high expression allele (MAOA-H; Caspi et al. 2002). Despite mounting evidence suggesting a relationship between the MAOA-uVNTR polymorphism and aggressive be- havior, it is unclear how this genetic polymorphism predisposes individuals to aggressive behavior. Aggression researchers have distinguished between two types of aggressive behavior, one resulting from a lack of emotional sensitivity and one resulting from excessive emotional sensitivity (Blair et al. 2006; Crick and Dodge 1996). Instrumental or proactive aggression is pre-meditated, goal-directed aggression that is used to obtain a desired goal. This type of aggression has been associated with psychopathy and often involves diminished emotional sensitivity, empathy, and remorse (Berkowitz 1993; Blair et al. 2006; Frick et al. 2003). Reactive aggression, in contrast, is triggered by negative experiences and involves exaggerated levels of negative emotion, such as anger or anxiety, in response. This type of aggression is thought to result from a more responsive threat detection system as well as a diminished capacity to regulate the heightened emotional responses (Blair 2004; Blair et al. 2006; Grafman et al. 1996). Despite the fact that aggressive behavior clearly relates to affective processes, few neuroimaging studies have investigated how the MAOA poly- morphism relates to neural activity associated with these affective processes. Instead, neuroimaging studies have focused primarily on how the MAOA polymorphism relates to executive attention or inhibitory control during cognitive tasks, typically observing that the MAOA polymorphism relates to altered activity in neural regions involved in triggering and instantiating cognitive control (Fan et al. 2003; Meyer-Lindenberg et al. 2006; Passamonti et al. 2006). To date, only one study has examined the relationship between the MAOA polymorphism and affect-related processing. This study examined how the MAOA polymorphism related to individual differences in the gray matter volume of limbic regions and in the responses of these regions to emotional stimuli, specifically negative emotional faces (Meyer-Lindenberg et al. 2006). Compared with MAOA-H, MAOA-L individuals showed reduced gray matter volumes in limbic regions such as the amygdala, dorsal anterior cingulate cortex (dACC), and sub- genual ACC and greater amygdala and subgenual ACC activity to negative emotional faces. Although this study represents an advance in understanding how the MAOA-uVNTR polymor- phism relates to affective processing, the study did not examine self-reports or behavioral assessments of aggression. Moreover, because the affective stimuli used in this study, namely pictures of negative emotional expressions, are not likely to elicit full- blown emotions, it is difficult to know how the MAOA polymor- From the Department of Psychiatry & Biobehavioral Sciences (NIE), Cous- ins Center for Psychoneuroimmunology; and the Department of Psy- chology (BMW, SET, WTW, MDL), University of California, Los Angeles, California. Address reprint requests to Naomi I. Eisenberger, M.D., UCLA, Cousins Cen- ter for Psychoneuroimmunology, 300 Medical Plaza, Room 3156, Los Angeles, CA 90095-7076; E-mail: neisenbe@ucla.edu. Received June 29, 2006; revised August 9, 2006; accepted August 10, 2006. BIOL PSYCHIATRY 2007;61:1100 –1108 0006-3223/07/$32.00 doi:10.1016/j.biopsych.2006.08.007 © 2007 Society of Biological Psychiatry