ABSTRACT
Prior work has consistently revealed a relationship between antisocial behavior and reduced P300 amplitude. Fewer studies have directly evaluated behavioral indices of aggression and P300, and those that have generally do not account for potential mediating variables such as age, intelligence, and behavioral task performance. The current study assessed the relationship between the total number of convicted violent and non-violent offenses and P300 in a sample of inmates from a medium security state prison. Violent offenses evidenced a robust negative relationship with P300 amplitude, whereas non-violent offenses did not. Additional analyses evaluated age, intelligence, and behavioral task performance as potential mediating variables. Only reaction time significantly predicted P300 amplitude, and mediational analyses showed that this relationship did not account for the violent-offense/P300 relationship. Findings are discussed in terms of personality correlates and neurobiological process related to aggression. There is long-standing interest in the notion that antisocial behavior, and aggression in particular, involves neurobiologically-based deficits in information processing. Neuropsychological research has revealed that antisocial behavior is associated with impaired executive function (c.f. Morgan and Lilienfeld, 2000), and neuroimaging studies have consistently identified frontal lobe abnormalities among violent offenders (Goyer et al., 1994; Raine et al., 1997; Raine et al., 2000; Volkow et al., 1995). Furthermore, research using event-related brain potentials has indicated that antisocial behavior is associated with reduced P300 responses to task-relevant stimuli in target detection tasks (e.g., Bauer et al., 1994; Iacono et al., 2003). These deficits may reflect inefficient neural processing of salient environmental stimuli (Donchin and Coles, 1988), which could potentially contribute to risk for antisocial deviance. Notably, antisocial behavior encompasses both violent and nonviolent transgressions. One unresolved issue is whether P300 amplitude is associated with both violent and non-violent forms of antisocial behavior, or more predominantly with aggressive forms of acting out. Some research has examined P300 response in aggressive individuals specifically (e.g., Barratt et al., 1997); this work has revealed that reduced P300 amplitude is selectively related to impulsive, but not instrumental, aggression. However, this work has not compared associations for aggressive versus non-aggressive offending behavior. Thus, the aim of the present study was to replicate and extend prior research on antisocial behavior and brain response by examining relations between violent and non-violent offenses and P300 amplitude in a sample of adult male inmates. Our prediction was that reduced P300 response would be associated more predominantly with a history of violent offending, and that this effect would be independent of age, intelligence, and task performance effects.
