Takacs, A. | Search bioRxiv

bioRxiv2017

Do adolescents take more risks? Not when facing a novel uncertain situation

In real-life decision-making, sub-optimal risk-taking seems characteristic of adolescents. Such behavior increases the chance of serious negative, and at times, irreversible outcomes for this population (e.g., road traffic accidents, addictions). We are still lacking conclusive evidence, however, for an inverted U-shaped developmental trajectory for risk-taking. This raises the question whether adolescents are really more risk-prone or when facing a novel risky situation, they behave just as children and adults do. To answer this question, we used the Balloon Analogue Risk Task (BART) to assess the risky decision making of 188 individuals ranging in age from 7 to 30. The BART provided useful data for characterizing multiple aspects of risk-taking. Surprisingly, we found that adolescents were not more inclined to take risks than children or young adults. Participants in all age groups were able to adapt their learning processes to the probabilistic environment and improve their performance during the sequential risky choice. There were no age-related differences in risk-taking at any stage of the task. Likewise, neither negative feedback reactivity nor overall task performance distinguished adolescents from the younger and older age groups. Our findings prompt 1) methodological considerations about the validity of the BART and 2) theoretical debate whether the amount of experience on its own may account for age-related changes in real-life risk-taking, since risk-taking in a novel and uncertain situation was invariant across developmental stages.

neuroscience↗

bioRxiv2017

Sensitivity to sequential regularities in risky decision making

Probabilistic sequence learning involves a set of robust mechanisms that enable the extraction of statistical patterns embedded in the environment. It contributes to different perceptual and cognitive processes as well as to effective behavior adaptation, which is a crucial aspect of decision making. Although previous research attempted to model reinforcement learning and reward sensitivity in different risky decision-making paradigms, the basic mechanism of the sensitivity to statistical regularities has not been anchored to external tasks. Therefore, the present study aimed to investigate the statistical learning mechanism underlying individual differences in risky decision making. To reach this goal, we tested whether implicit probabilistic sequence learning and risky decision making share common variance. To have a more complex characterization of individual differences in risky decision making, hierarchical cluster analysis was conducted on performance data obtained in the Balloon Analogue Risk Task (BART) in a large sample of healthy young adults. Implicit probabilistic sequence learning was measured by the Alternating Serial Reaction Time (ASRT) task. According to the results, a four-cluster structure was identified involving average risk-taking, slowly responding, risk-taker, and risk-averse groups of participants, respectively. While the entire sample showed significant learning on the ASRT task, we found greater sensitivity to statistical regularities in the risk-taker and risk-averse groups than in participants with average risk-taking. These findings revealed common mechanisms in risky decision making and implicit probabilistic sequence learning and an adaptive aspect of higher risk taking on the BART. Our results could help to clarify the neurocognitive complexity of decision making and its individual differences.

neuroscience↗