Do individual characteristics explain variation in sanction risk updating among serious juvenile offenders? Advancing the logic of differential deterrence.

Abstract

Recent theoretical and empirical research in deterrence has detected evidence of differential deterrability, or that the effect of sanctions is not uniform across persons. Important questions in this area remain to be explored, and this study considered a central question: Whether important across-individual variability in risk perceptions can be tied to important individual-level factors. This article extends the Bayesian risk updating model developed by Anwar and Loughran (2011) to determine whether the weight individuals place on new offending information differs across persons on the basis of individual characteristics. Using longitudinal data from a sample of serious youthful offenders, we found that the effect of an arrest on updated risk perceptions differs as a function of IQ and early behavioral problems. In particular, individuals low in verbal IQ and those who display early behavioral problems place greater weight on new information, suggesting that individuals who are traditionally viewed as being high in criminal propensity are more affected by arrest such that they are more likely to increase their perceived risk of getting caught when offending after experiencing an arrest. Findings support the idea that high-propensity offenders are more responsive to formal sanctions, and thus (potentially) more deterrable, than low-propensity individuals.

Heuristics and Biases, Rational Choice, and Sanction Perceptions*

“…Neither the literature on situational crime prevention nor Bayesian learning theory is well suited for explaining how individuals’ dispositions or personalities influence their sanction perceptions. Nevertheless, the results of studies have shown that several personality traits are strongly associated with perceived punishment risk (Pickett and Bushway, ; van Gelder and de Vries, , ) and influence sanction perception updating (Thomas, Loughran, and Piquero, ). In a seminal study, van Gelder and de Vries () found that honesty‐humility, agreeableness, extraversion, conscientiousness, and self‐control were all associated with perceived sanction risk.…”

“…Such decisions often occur when actors are intoxicated and/or viscerally aroused (Loewenstein, Nagin, and Paternoster, ; Pridemore, ), with incomplete information, and under conditions of chaos and/or time‐sensitivity. Moreover, there is wide heterogeneity in the capacity and inclination for abstract and otherwise effortful thinking (Pickett and Bushway, ; Thomas, Loughran, and Piquero, ). For example, individuals with low cognitive reflection are less likely to persevere past an impulsive but incorrect conclusion to obtain the correct one with added cognitive effort (Frederick, ).…”

Peers and offender decision‐making

“…Second, individual differences in dual‐system processing might moderate the association between socialization by peers and decision outcomes. Individuals who tend to rely on the fast system may overweigh information from the current situation at the expense of their experiences in the past, thereby making prior socialization less relevant (Piquero & Pogarsky, ; Thomas & McGloin, ; Wikström, , see also Thomas, Loughran, & Piquero, on verbal IQ). It is also possible that those individuals find it difficult to translate vicarious observations to their own situation, which would imply that socialization has less impact on their developing perceptions.…”

Behavioral criminology and public policy

“…One example is a formal, rational model of dynamic Bayesian updating of perceptions of detection risk proposed and tested by Anwar and Loughran (). Systematic departures from this standard model have been subsequently studied by Thomas, Loughran, and Piquero () and by Pogarsky et al. ().…”