Abstract:

Juvenile crime, as an important factor of public security, has extensively concerned the whole society. Although juvenile offenders with antisocial tendency exhibit unique behavior patterns and development characteristics, few studies have investigated the potential mechanism of aggressive behaviors in juvenile offenders. Based on the dual-hormone hypothesis and social information processing theory, It is plausible to argue that biological hormones and the social information processing process related to hostile might play an fundamental role in shaping aggressive behaviors of juvenile offenders. Therefore, the present study proposes a moderated mediating model and examined the prediction of testosterone for aggressive behaviors in juvenile offenders that is mediated by hostile attention bias and moderated cortisol. Drawing on both biological hormone and social cognition perspectives, the present study attempts to answer the questions of “how” testosterone affects aggressive behavior and “when” the effect will be present, which provides implications for further developing interventions against aggressive behaviors in juvenile offenders with antisocial tendency.
This study included 84 juvenile offenders (average age: 17.55, standard deviations: 0.52) as participants. Their aggressive behavior was measured using the Chinese version of the Buss-Perry Attack Scale (AQCV). Enzyme-Linked Immunosorbent Assay (ELISA) was utilized to measure the salivary testosterone and cortisol levels of the participants. Moreover, the participants’ attention biases toward hostile aggressive phrases and hostile facial expressions were evaluated using an adopted eStroop and the dot-probe paradigm, respectively. Both traditional and trail-level attention bias scores were calculated to reflect the participants’ attention bias statically and dynamically.
Results showed that (1) attention bias toward hostile stimuli (TLBS-Variability, Peak TL-BS_negative) played a complete mediation role in the relationship between testosterone and aggressive behavior (i.e. variability played a complete mediation role in the prediction of AQCV total score, physical aggression subscale score, and anger subscale score. Peak TL-BS_negative played a complete mediation role in the prediction of verbal aggression subscale score by testosterone). (2) Cortisol moderated the relationship between testosterone and attention bias toward hostile stimuli. More specifically, at a high cortisol level, testosterone positively predicted individual attention avoidance (Peak TL-BS_negative) and variability (TLBS-Variability), thereby reducing the aggression level. However, the mediation effect of attention bias toward hostile stimuli was not significant at a low cortisol level.
Based on the dual-hormone hypothesis and social information processing theory, the present study examined a moderated mediating model in a sample of juvenile offenders with antisocial tendency using the perspectives of biological hormone and social information processing in aggressive behaviors. The present study revealed a potential development mechanism of aggressive behaviors in juvenile offenders and thus provides an empirical foundation for hormone-based interventions against aggressive behaviors in juvenile offenders. Finally, on the basis of the mediation and moderation effects of biological hormones on aggressive behavior, the present study indicates that increasing attention avoidance and the cortisol level for juvenile offenders with antisocial tendency might help reduce their aggressive violence.

Key words: aggressive behavior, testosterone, cortisol, attention bias, juvenile offender