Do Child Molesters Deliberately Fake Good on Cognitive Distortion Questionnaires? An Information Processing-Based Investigation

Abstract

Researchers and clinicians hypothesize that child molesters hold offence-supportive beliefs or cognitive distortions that require restructuring for successful rehabilitation. However, there is little empirical evidence to support this hypothesis. Current questionnaire measures of both untreated and treated child molesters' cognitive distortions show that these men typically disagree with cognitive distortions. Such findings, especially prior to treatment, are often interpreted to mean that child molesters are faking good. In this study we drew on personality-related research showing that when participants intentionally respond in a socially desirable way on questionnaires, they make faster item responses than when answering honestly. Untreated child molesters, treated child molesters, and two types of controls (nonsexual offenders and nonoffenders) were administered a computerized cognitive distortion questionnaire and their responses and response times were recorded. Consistent with previous research, all groups tended to disagree with the cognitive distortions. However, response time data were surprising; only the treated child molesters displayed the fake-good pattern of responding significantly faster in rejecting cognitive distortions. We speculate about the implications of these results for the cognitive distortion hypothesis, and for understanding how current treatment programs effect cognitive distortion change.

The Development and Validation of the Cognitions of Internet Sexual Offending (C-ISO) Scale

“…The tendency among men who have committed child sexual offenses to disagree less with statements indicative of cognitions that support sexual offending against children, rather than agreeing with them, was also observed in other studies (e.g., Arkowitz & Vess, 2003 ; Marshall, Marshall, Sachdev, & Kruger, 2003; Merdian, Curtis, Thakker, Wilson, & Boer, 2013). A potential explanation for this finding is that social desirability might influence respondents to hide their true beliefs, maybe to preserve their self-esteem (Gannon & Polaschek, 2005; Kolton, Boer, & Boer, 2001; Langevin, 1991). Given that the C-ISO did correlate somewhat with social desirability, it could be that this effect was at play here as well.…”

“…As the scale aims to assess the online sexual offending-offense supportive cognitions, it is expected that men who had committed Internet sexual offenses would obtain higher scores than those who did not. As men convicted of sexual offenses tend to answer auto-reveal questionnaires in a socially desirable way (Gannon & Polaschek, 2005; Langevin, 1991), both an analysis of variance and an analysis of covariance (ANCOVA) to control for social desirability were conducted to examine the discriminant validity of the C-ISO.…”