In this study, the relationship between psychopathy and the perpetration of sexual homicide was investigated. The official file descriptions of sexual homicides committed by 18 psychopathic and 20 nonpsychopathic Canadian offenders were coded (by coders unaware of Psychopathy Checklist-Revised [PCL-R] scores) for characteristics of the victim, victim/perpetrator relationship, and evidence of gratuitous and sadistic violent behavior. Results indicated that most (84.7%) of the sexual murderers scored in the moderate to high range on the PCL-R. The majority of victims (66.67%) were female strangers, with no apparent influence of psychopathy on victim choice. Homicides committed by psychopathic offenders (using a PCL-R cut-off of 30) contained a significantly higher level of both gratuitous and sadistic violence than nonpsychopathic offenders. Most (82.4%) of the psychopaths exhibited some degree of sadistic behavior in their homicides compared to 52.6% of the nonpsychopaths. Implications for homicide investigations are discussed.

Violence is a social problem that carries enormous costs; however, our understanding of its etiology is quite limited. A large body of research exists, which suggests a relationship between abnormalities of the frontal lobe and aggression; as a result, many researchers have implicated deficits in so-called "executive function" as an antecedent to aggressive behaviour. Another possibility is that violence may be related to problems interpreting facial expressions of emotion, a deficit associated with many forms of psychopathology, and an ability linked to the prefrontal cortex. The current study investigated performance on measures of executive function and on a facial-affect recognition task in 20 violent offenders, 20 non-violent offenders, and 20 controls. In support of our hypotheses, both offender groups performed significantly more poorly on measures of executive function relative to controls. In addition, violent offenders were significantly poorer on the facial-affect recognition task than either of the other two groups. Interestingly, scores on these measures were significantly correlated, with executive deficits associated with difficulties accurately interpreting facial affect. The implications of these results are discussed in terms of a broader understanding of violent behaviour.

Purpose. This study examined the verbal and non-verbal behaviours exhibited by criminal offender and non-offender participants while they related planned truthful and deceptive accounts about emotional autobiographical events.Methods. In a 2 £ 2 (participant group £ veracity) quasi-experimental design, offenders ðN ¼ 27Þ and university students ðN ¼ 38Þ provided videotaped accounts of four autobiographical emotional events: two honest and two fabricated (counterbalanced). Patterns of behaviour exhibited during the truthful and the deceptive accounts were then compared.Results. In general, offenders and non-offenders showed similar patterns of deceptive behaviour. Deceptive accounts by both groups contained fewer details than honest accounts. Deception was associated with an increase in illustrator usage and self-manipulations; however, univariate analyses indicated only that offenders exhibited significantly more self-manipulations when lying. A significant interaction emerged in which offenders showed a reduction in smiles when lying about the emotional events, while students showed no difference.Conclusions. Offenders and students showed similar patterns of lying on most cues. However, unlike non-offenders, offenders smiled less and showed an increase in self-manipulations when lying. We theorize that offenders may have been aware that smiling and laughing are negatively related to perceived credibility in the speaker and used self-manipulations to distract listeners from the content of their lies.