Legal and Criminological Psychology (2003), 8, 51–62 © 2003 The British Psychological Society www.bps.org.uk Personality traits, personality disorders and sensational interests in mentally disordered offenders Vincent Egan 1 *, Elizabeth Austin 2 , Debbie Elliot 3 , Darshana Patel 3 and Phillip Charlesworth 3 1 Department of Psychology, Glasgow Caledonian University, UK 2 Department of Psychology, Edinburgh University, UK 3 Department of Psychology, Arnold Lodge, Regional Secure Unit, Leicester, UK Purpose. Sensational interests (e.g. an interest in the occult or the methods of violence) in mentally disordered offenders are claimed to signify greater risk of psychopathology, but evidence to support this view is slight. Methods. The relationships between self-reported DSM-IV personality disorder (PD), general personality traits and sensational interests were examined in 155 of 167 consecutively referred offenders to a forensic psychology service. The subscales of the PD and personality trait measures were reduced to the four basic PD/trait dimensions (asocial, antisocial, anxious and anankastic) using conrmatory factor analysis. Results. Those high on the ‘antisocial’ factor (which was primarily dened by low Agreeableness, low Conscientiousness, and substantial elements of Paranoid, Anti- social and Borderline PD) were more interested in ‘violent-occult’ and militaristic topics. Conclusions. The aspects of the antisocial factor primarily associated with an interest in sensational and potentially violent topics cover a wide range of putative disorders. However, the factors reecting asocial, anxious or anankastic disorders do not show a reliable association with measures of sensational interests. These results suggest that the personality dimensions reecting an interest in ‘sensational’ topics in mentally disordered offenders are relatively specic. There is increasing recognition that personality disorders are on a dimensional con- tinuum and relate to general personality traits (Parker & Barrett, 2000). Co-morbidity is the norm, and the ostensibly discrete, diverse personality disorders appear to be based on a small set of overlapping and underlying constructs (Coid, 1992; Costa & Widiger, 1994; Widiger & Trull, 1992). Typically, four factors explain most of the variance shared between the 10 core DSM-IV personality disorders (Austin & Deary, 2000; Livesley, Jang, & Vernon, 1998). These factors can be summarized as ‘the 4 As’: *Requests for reprints should be addressed to Professor Vincent Egan, Department of Psychology, Glasgow Caledonian University, Cowcaddens Road, Glasgow G4 0BA, UK (e-mail: v.egan@gcal.ac.uk). 51