Graham, S.A. (2002). Assessing emotional responses in sex offenders. (Doctoral dissertation, University of Taxes at Austin, 2002). Dissertation Abstracts International, 64 (2-B), 963.

Abstract:  Emotional responses were assessed for groups of child molesters, rapists, and controls to determine differences between the groups on a variety of factors. Subjects were shown film clips that were typically emotional in content, and which included a rape scenario, child molest scenario, and war scenario. The subjects self-reported their sense of involvement in what they were viewing. Self-report was correlated with physiological measures of heart rate, and electromyography, to compare physiological responses of involvement with self-report. There was no significant correlation found between self-report measures and physiological measures of involvement. In addition, subjects were tested on a global measure of empathy. This study corroborated prior research which found that no differences existed between sex offender subjects and controls in assessing empathy through a global empathy measure. However, when presented with specific subscales of this measure of empathy such as perspective taking and fantasy, significance was noted. Congruent emotional response was defined a s feeling emotion that was similar in nature to that exhibited in the film clip. Incongruency was defined as feeling opposite feelings, i.e. feeling positive feelings such as anticipation, and expectancy, when viewing a scene that portrays negative emotions, i.e. fear, anger, sadness. Congruency was measured by electromyography (EMG), specifically by an electrode attached to the zygotic muscle on the side of the face. This muscle is used solely in the smile response, and if activated, indicated an oppositional response to the seriousness of the film clip content. No significant difference was found in EMG response when tested in a film clip by group interaction, implying that congruency of emotional response was not related to type of offense. Factors such as aggression, social alienation, and affect intensity (comfortableness with emotion) were also studied. These factors yielded no significant correlation with the congruity measure (EMG). In related analyses, the offender groups significantly differed from the controls on levels of aggression, alienation, and affect intensity. In a comparison among means, the child molester group showed the highest levels of aggression and alienation, while the rapist group exhibited the highest level of affect intensity. (PsycINFO Database Record (c) 2004 APA, all rights reserved)