An Integrative Model of Aggression: The Role of Cognitions in Responses to Stressors in Forensic and Non-Forensic Populations

This PhD aimed to further the understanding of aggression through the integration of research findings with theoretical models. As a result, a Stratified Integrated Model of Behavioural Aggression (SIMBA) that specifies and stratifies the roles of stress, cognitive structures and information processing was proposed. This may help guide therapeutic interventions aimed at the reduction of aggressive behaviour and inform risk assessment. A systematic literature review of 77 papers was conducted to assess the relationship between stress systems’ activity and aggression. The results showed that this relationship is likely present and can be both positive and negative. Thematic analysis of these papers identified six themes: 1) the impact of testosterone on the relationship between activity of stress response systems and aggression is undetermined; 2) the presence of sex differences in the relationship between stress response and aggression depends on the stress system and type of aggression; 3) specific disorders do not influence the relationship between stress and aggression; 4) experience of victimisation does not have a clear influence on the relationship between stress systems’ activity and aggression; 5) the relationship between stress response markers and aggression differs among those with high stress exposure; and 6) history of aggression affects the relationship between stress response markers and anger-based aggression. These results highlighted that the stress-aggression relationship is present, but is likely to be indirect. However, the extraneous variables consistently affecting this relationship were not identified. Addressing this issue, study one recruited 20 male students and 11 patients from a high secure hospital to establish the effects of aggression supportive cognitions and stress on aggressive behaviour. To assess aggression after a stress-evoking task, the Taylor Aggression Paradigm was used. It was predicted that while the presence of aggressive Implicit Theories (ITs) would be positively associated with aggressive behaviour towards a stranger, the association of stress would differ between the samples. This was partially supported, as only one specific IT (“I am the law”) was associated with aggression. Furthermore, only elevated skin conductance, but not changes in the heart rate, during the stress task was positively associated with aggression, and only among patients. Study two involved 100 participants (49 men, 48 women, three not disclosed) with an average age of 29. It aimed to investigate the relationship between history of aggressive behaviour, affective states, and neutral and emotional information processing. Event Related Potentials (ERPs) during a Go/No-Go task were utilised to capture cognitive resources allocation, with a “supervisor – employee” laboratory paradigm used to assess aggression. Contrary to expectations, results showed that trait aggressiveness was only related to aggressive behaviour at higher levels of inhibitory processing. The hypothesis that artificially provoked changes in negative and positive affect would be related to aggressive behaviour was also not supported. However, as expected, feeling hostile was associated with short-lived aggressive behaviour, but only for those who had low response inhibition. Moreover, partially supporting expectations, a history of aggressive behaviour moderated the relationship between change in negative affect and aggressive behaviour. The last hypothesis, proposing emotional processing to be a mediator between response inhibition and aggressive behaviour, was also not confirmed. Study three included 462 participants, of whom 300 were adults aged 26 or older (151 men, 149 women), and 162 representing transitional aged youth, aged between 18 and 25 (21 men, 141 women). This study aimed to identify direct and indirect effects exerted by aggression supportive cognitive structures, working memory problems, and stress on aggression by building a Structural Equation Model. It was expected that a direct cognitive pathway from aggression supportive cognitions directly to aggression would be identified. This hypothesis was supported. Meanwhile, the second hypothesis proposing an indirect relationship between stress and aggression was only partially supported, with maladaptive coping style being the only mediator identified. The current research demonstrated that aggression-supportive cognitive structures are the primary facilitators of aggressive behaviour. Meanwhile, the effect exerted by situational demands is contingent on the preferred coping style. Furthermore, despite the indirect nature, the influence of information processing was present for multiple precursors of aggression. Consequently, all these elements were included in the SIMBA and are suggested as primary targets for therapeutic aggression interventions. The results are discussed with attention to this proposed model, capturing further directions for future research