Adolescent risk-taking is predicted by individual differences in cognitive control over emotional, but not non-emotional, response conflict

ABSTRACT

While much research on adolescent risk behaviour has focused on the development of prefrontal self-regulatory mechanisms, prior studies have elicited mixed evidence of a relationship between individual differences in the capacity for self-regulation and individual differences in risk taking. To explain these inconsistent findings, it has been suggested that the capacity for self-regulation may be, for most adolescents, adequately mature to produce adaptive behaviour in non-affective, “cold” circumstances, but that adolescents have a more difficult time exerting control in affective, “hot” contexts. To further explore this claim, the present study examined individual differences in self-control in the face of affective and non-affective response conflict, and examined whether differences in the functioning of cognitive control processes under these different conditions was related to risk taking. Participants completed a cognitive Stroop task, an emotional Stroop task, and a risky driving task known as the Stoplight game. Regression analyses showed that performance on the emotional Stroop task predicted laboratory risk-taking in the driving task, whereas performance on the cognitive Stroop task did not exhibit the same trend. This pattern of results is consistent with theories of adolescent risk-taking that emphasise the impacts of affective contextual influences on the ability to enact effective cognitive control.

KEYWORDS:

• Cognitive control
• risk taking
• executive function
• adolescence
• affect

Acknowledgment

The authors would like to thank Melissa Wilson for her efforts in recruiting and scheduling participants.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. The significant difference in RTs could have been due to the imposition of a response deadline in the cognitive, but not emotional, Stroop variant. To partially address this possibility, we duplicated all analyses after excluding trials from the emotional Stroop for which participants’ RTs exceeded 4000 ms (the response deadline in cognitive Stroop). The overall differences in reaction time remained significant, and eliminating these longer response trials did not qualitatively change any of the subsequently reported findings. Accordingly, we report the results based on inclusion of all trials.

Additional information

Funding

This work was supported by the National Institute on Drug Abuse [grant number R21 DA031436-02].