Why are Psychopaths Difficult to Treat? Testing the Two-Component Model for the Treatment of PCL Psychopaths

Psychopathic personality disorder as conceptualised by the family of scales referred to as the Psychopathy Checklist (PCL), is often cited as a specific responsivity characteristic that will interfere with an otherwise effective treatment programme. However, most research on the treatment of prisoners high on PCL psychopathy asks whether or not they are treatable as opposed to why they are difficult to treat. The Two-Component model (2-C; Wong & Olver, 2015) for the treatment of PCL psychopaths proposes that treatment difficulties observed for those high on PCL psychopathy are primarily caused by the interpersonal and affective personality features of psychopathy represented by PCL Factor 1 (F1). Thus the 2-C model suggests that therapists work around the emotional deficits and disruptive behaviours associated with PCL F1 to focus on changing risk-relevant behaviours that are associated with PCL Factor 2 (F2). In this thesis, we test the assumptions of the 2-C model with a group of high-risk violent men who attended an intensive violence treatment programme and were assessed with a PCL instrument. Specifically, we examined whether the personality features of PCL psychopathy led to more treatment difficulties than the behavioural features by exploring relationships between the factors/facets of the PCL and treatment completion, reconviction, change on dynamic risk, the therapeutic alliance and behaviour during treatment. In support of the 2-C model we found that PCL:SV Part 1 and its underlying facets were significantly associated with higher rates of removal from treatment, a poorer therapeutic alliance, and lower levels of emotional and performance based behaviours during treatment. Mediation analyses also revealed that the relationships between PCL:SV Part 1 variables and removal were partly explained by lower levels of emotional and performance based behaviours. Also in support of the 2-C model, we found that PCL:SV Part 2 and its underlying facets demonstrated stronger, significant associations with pre-treatment dynamic risk and post-treatment reconvictions when compared with PCL:SV Part 1 variables. Furthermore, poorer performance based behaviours during treatment mediated relationships between PCL:SV Part 2 variables and reconviction outcomes. Several of our findings however, also failed to support or contradicted assumptions of the 2-C model. All these findings are discussed in relation to their implications for the 2-C model, psychopathy treatment research, specific responsivity, the structure of PCL psychopathy, and the utility of the PCL in forensic and legal settings.