The role of aggression-related cognitions in aggressive behaviour

Violent offenders represent a substantial proportion of prisoner populations and the rehabilitation of this group is a necessary priority of correctional services. Despite the recognition that programs delivered in this area have some demonstrated efficacy, the principles of effective treatments are often inadequately demarcated and the extent to which interventions can reduce aggressive and violent propensities is modest relative to interventions targeting other forms of offending (Howells, 2010). Researchers have consequently called for greater intersection between psychological aggression theory and interventions for violent offenders (e.g., McGuire, 2008). In this thesis, the most comprehensive and contemporary aggression theory, the General Aggression Model (GAM; Anderson & Bushman, 2002), is drawn upon to examine how the conceptualisation of aggression in offender populations might be enhanced, enabling more sophisticated attempts to reduce aggressive and violent tendencies. Although existing research supports the GAM’s main contention that aggressive individuals regularly retrieve and employ aggression-related cognitions, the application of this framework to violent offending has been hindered by insufficient empirical validation of the model’s central tenets in individuals with clinically significant levels of aggression. In particular, studies examining the importance of the GAM’s characterisation of aggression-related cognitions to clinical formulations of personality pathology are non-existent, despite the recognition that individuals with certain personality disorders (PD) are over-represented in forensic populations and consistently demonstrate higher rates of aggression. A series of theoretical and empirical investigations of the importance of aggression-related cognitions to aggression were consequently undertaken in relation to the aggressive behaviour of offenders. A more specific aim of the research was to examine how the application of the GAM might assist in elucidating the relationship between PD and aggressiveness. The research commences with a comprehensive review of the literature regarding current psychological aggression theory and violent offender treatment, examining the evidence available to support the role of the constructs specified by the GAM in aggressive individuals. The findings emerging from this analyses were that: a) the GAM bears relevance to clinically significant levels of aggression and may assist in improving the efficacy of violent offender treatment programs, although further research is required to more fully explore the importance of the GAM constructs to aggression, and b) for offenders with PD, utilisation of the GAM can provide researchers and clinicians with a framework to more fully characterise the relationships between various PDs and aggression, and clinically, would enable systematic assessment and treatment of constructs that are conducive to aggression. For the empirical component of the research, 87 offenders referred for presentence evaluation were assessed on their histories of aggression, Axis I and II diagnostic status, psychopathy, aggression-related and other maladaptive cognitions (normative beliefs supportive of aggression, aggressive behavioural scripts, Early Maladaptive Schema [EMS]) and trait anger. The convergent validity of three instruments designed to measure past aggression was initially investigated (Life History of Aggression, Aggression Scale [LHA-A]; Violence Rating Scale; Cormier-Lang System for Quantifying Criminal History), and the finding that the LHA-A most reliably quantifies the density of past aggression (i.e., frequency and severity of acts) was used to inform the methodology of subsequent studies. The first empirical study examined the importance of three knowledge structures; namely, normative beliefs supportive of aggression, aggressive scripts, and EMS, along with trait anger, to the aggression histories of the sample. Regression analyses were used to investigate the unique contribution of each of these constructs to aggression, with the results suggesting that positive attitudes to violence, frequent script rehearsal and high trait anger concurrently increased the likelihood of past aggression. In contrast, EMS did not predict aggression. On the whole, these findings were consistent with the GAM’s understanding of aggression, and suggested that more systematic targeting of aggression-related cognitions in violent offenders is necessary. The second empirical study examined whether normative beliefs supportive of aggression, aggressive scripts and trait anger assisted in explaining differences in aggressiveness among offenders with Axis II PD symptoms and traits of psychopathy. Regression analyses were undertaken to examine the relative contributions of aggression-related constructs (i.e., normative beliefs, scripts and trait anger) and PD to aggression. The results confirmed a relationship between several PDs and aggression, and more importantly, suggested that for the majority of PDs related to aggression (Antisocial, Borderline, Paranoid and Schizotypal Personality Disorders, and psychopathy), the consideration of normative beliefs, scripts, and anger facilitated an improved understanding of aggressiveness. Overall, the findings suggest that the GAM offers valuable insight into the psychological features that characterise individuals with PD who are prone to aggression. Taken together, these findings suggest that contemporary aggression theory appears to have much to offer in enhancing understandings of aggression in offender populations. Furthermore, they indicate that broader adoption of the GAM framework is likely to assist in both characterising and differentiating among violent offenders and would provide a comprehensive conceptual framework upon which to target and ameliorate aggressive behaviour.