What were they thinking? An exploration of child sexual offenders’ beliefs using a lexical decision task

Abstract

Cognitive distortions have been afforded a key role in the offending behaviour of child sexual offenders. While the mechanisms underlying cognitive distortions are not fully understood, they are generally thought to reflect entrenched beliefs that distinguish child sexual offenders from other individuals. We investigated this hypothesis using a robust experimental technique called the lexical decision task. Child sexual offenders, offender controls, and non-offender controls completed a lexical decision task in which they responded to words that completed sentences in either an offence-supportive or nonoffence-supportive manner. Contrary to predictions, child sexual offenders did not respond faster to words that were consistent with offence-supportive beliefs, relative to controls. However, they did show accelerated recognition for word stems supporting external locus of control beliefs. These results highlight the need to use cognitive experimental methods to study child sexual offenders’ beliefs, and the importance of investigating potential alternative drivers of cognitive distortions.

Keywords:

• child sexual offenders
• child molesters
• beliefs
• cognition
• cognitive distortions
• lexical decision task

Notes

1. Because empathy deficits reflect an inability to recognise victim harm, in recent years they have been subsumed under the category of cognitive distortions (Marshall, Anderson, & Fernandez, 1999; Ward & Beech, 2005)

2. A total of 115 men completed the experiment (41 child sexual offenders, 43 offender controls, and 31 community controls). However, 15 participants were dropped from the study because they made more than six errors out of 45 trials during testing – an error rate of 13%.

3. Because we did not directly distribute fliers to prisoners we cannot say what proportion of men who were asked to take part in the study agreed to participate.

4. ANOVAs confirmed that participant groups in the three LDT versions did not significantly differ in terms of age, years of education, and reaction times to offence-supportive words, nonoffence-supportive words or non-words, while a chi-square test found no differences in ethnicity.

5. Incomplete sentence presentation duration was based on feedback from five independent community-based raters. On completion of the experiment all participants were asked to rate sentence presentation on a five-point Likert scale ranging from 0 (Much too quickly) to 4 (Much too slowly). Mean rating was 2.11, indicating that, on average, participants felt incomplete sentences were presented ‘at the right speed’.

6. Ninety-nine per cent of all responses were correct.