Imaginal Exposure and Response Prevention for Anger and Aggressive Behavior

Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Mitchell E. Berman

Advisor Department



The purpose of this project was to develop and test a theoretically-derived exposure and response prevention (ERP) intervention for anger and aggressive behavior. Theorists and researchers have hypothesized that anger is conceptually similar to anxiety, and both emotional experiences have many phenomenological and physiological correlates in common. Although all treatments for anger, and its behavioral manifestations, including "impulsive" aggressive acts, involve some exposure to perceived provocative or threatening stimuli, few published studies have explored the efficacy of exposure for the treatment of anger and aggressive behavior. The present study was a prospective, randomized outcome study of an exposure-based intervention for anger and aggressive behavior. Specifically, participants were randomly assigned to either six sessions of exposure to anger-eliciting situations that was accompanied by response prevention (imaginal Exposure and Response Prevention; ERP) or a six-session control condition. Participants in both conditions completed self- and clinician-ratings pre-ERP, post-ERP, and at four week follow up. ERP participants provided within-sessions ratings of anger levels every five minutes (i.e., subjective units of distress ratings) to explore the relationship between the intensity of the exposure and ERP outcome. Participants underwent an interview of overt aggressive behavior as part of the assessment process to investigate whether the proposed ERP is helpful in decreasing impulsive aggressive behavior. Although it was hypothesized that both groups would improve on all outcome measures from pre-ERP to post-ERP, there were only two significant effects of condition. Participants in the ERP condition reported significantly less derogatory thoughts than participants in the control condition. In addition, participants in the ERP condition increased their tendency to suppress their anger. In addition, the hypothesis that short-term habituation (i.e., decreases in within- and between-session SUDS ratings) would be correlated with outcome on measures of anger and aggressive behavior was not supported. However, participants', regardless of condition, improved over time on outcome measures of negative emotions (depression and anxiety), trait anger, anger expression, self-reported overt aggressive behavior, and hostile cognitions.