Risk Assessment of Criminal Offenders: An Integrated Neuropsychosocial Model


Student thesis: Doctoral Thesis

View graph of relations



Awarding Institution
Award date8 Sep 2017


Across the globe, criminal justice and forensic mental health professionals have shown great deference to risk assessment in their quest to predict individuals at heightened risk of offending and to institute intervention programs to mitigate their risk level. Risk assessment is predicated on the assumption that the factors that precede, associate with the onset, predict, or statistically correlate with offending are highly relevant in estimating the likelihood of future offending. Although studies have implicated both neurobiological and psychosocial factors in criminal trajectories, a major caveat in the risk assessment field is the gross neglect of neurobiological factors and their measures. Given that the predictive validity of the existing risk assessment instruments seems to have plateaued, an interesting area of empirical investigations is whether the infusion of neurobiological factors and their measures can enhance the prediction of offending. Consequently, the overarching goal of this dissertation is to reorient the focus of the risk assessment field by drawing researchers and practitioners’ attention to the likelihood of enhancing their professional endeavors by adopting a holistic approach involving the integration of neurocognitive and psychosocial factors of offending. To achieve this goal, first, the study proposes an integrated Neuropsychosocial Model (NM) incorporating both neurocognitive and psychosocial factors of crime to serve as a framework to guide risk assessment practice and research. Derived from the concepts and propositions of biosocial criminology, the proposed NM has two interrelated divisions. The risk evaluation division is based on the ethos of life-course or developmental criminology, tracing the origin of criminal behaviors from the factors at the prenatal through to the postnatal stages of human development, their influences on neurocognitive and psychosocial functioning, and the risk of offending. The risk management division, mainly informed by the risk evaluation division, also describes the offender management approaches to reduce the risk of offending. Second, the study investigates the utility of neurocognitive and psychosocial factors assessed with the Behavioral Assessment of Dysexecutive Syndrome (BADS) and Historical-Clinical-Risk Management-20 V3 (HCR-20), respectively, in discriminating between adult male recidivists and non-recidivists (N =120; recidivists = 60). A one-way multivariate analysis of variance was used to investigate whether recidivists and non-recidivists differ significantly on the linear combination of the HCR-20 and BADS. The area under curve (AUC) value of the receiver operating characteristic curve was used as a measure of discriminant validity, whereas incremental validity analysis was conducted using logistic regression. The results show that recidivists performed significantly poorer on the study measures than the non-recidivists. The independent and interaction effects of the HCR-20 and BADS significantly discriminate between recidivists and non-recidivists. The BADS similarly adds significant incremental validity to the HCR-20 to discriminate between the groups. Comparative analyses of the utility of the measures reveal no significant difference in the discriminant validity of the HCR-20 (i.e., clinical and risk factor and total score) and BADS. In contrast, the interaction between the measures significantly discriminates between the groups better than the BADS and HCR-20 clinical and risk management factors when used in isolation. These findings lend support to the proposed model and the utility of measures of both neurocognitive and psychosocial risk factors in risk assessment practice and research.

    Research areas

  • Risk assessment, risk evaluation, risk management, psychosocial factors, neurobiological factors, recidivism, Neuropsychosocial model