Individual and developmental differences in delinquency: Can they be explained by adolescent risk-taking models?

•Delinquency is often viewed as a prime manifestation of adolescent risk-taking.•Evidence for adolescent risk-taking models of delinquency is mostly indirect.•Better integration of risk-taking and developmental life-course criminology models could significantly accelerate progress. Delinquency shows an increase in adolescence and is hence often cited as a behaviour explainable in terms of adolescent risk-taking models. In this review we summarise contemporary developmental models of adolescent risk-taking as they apply to delinquency and evaluate the extent to which they are supported by empirical evidence. Dual Systems theory, Fuzzy Trace Theory, the Lifespan Wisdom Model (LWM), and the Developmental Neuro-Ecological Risk-taking Model (DNERM) are all discussed. We highlight that there have been very few direct empirical evaluations of developmental risk-taking models as applied to delinquency; however, indirect evidence supports the core Dual Systems theory claim that a developmental imbalance between sensation-seeking and self-regulation contributes to an adolescent peak in offending. However, this appears to apply particularly to a sub-group of vulnerable youth, as implied by the LWM. Further, risk-taking propensity likely interacts with age-related changes in exposure to risk-conducive situations, as implied by DNERM. There is little evidence to suggest that Fuzzy Trace Theory alone explains developmental changes in risk-taking, though it may help explain how young people learn about risk, as outlined in LWM, Better integration of risk-taking models with criminological perspectives as well as further longitudinal research using appropriate operationalisations of developmental imbalance, modelling individual differences in trajectories, and incorporating measures of exposure to risk-conducive situations will be essential for advancing knowledge of the drivers of engagement in delinquency in adolescence.