[Risk-taking in adolescence: A neuroeconomics approach]

Risk-taking behaviors represent the main cause of morbi-mortality in adolescence. Here, we analyze their neural correlates, based on a neuroeconomics approach. This approach postulates that risk-taking behaviors result from multiple decision-making biases that impair the selection of the most appropriate action among alternatives based on their subjective evaluation. Specifically, we investigate three important domains in value-based decision-making: risk aversion, loss aversion and intertemporal choice.

First, when people have to make a decision between two rewarding options, they will usually prefer the more certain, even possibly lower, option - a phenomenon called "risk aversion". Yet adolescent people have been found to be less averse to risk than adults. This observation was linked to hypoactivation in (1) the anterior insula, involved in negative emotion such as fear and disgust and (2) the anterior cingular and the posterior ventromedial prefrontal cortices, involved in the monitoring of conflict and error detection. Second, people are generally described as being more sensitive to the possibility of losing objects than to that of gaining the same objects - "loss aversion". Here, we suggest that adolescents may be less averse to losses than adults when estimating the prospects of gaining and losing objects. Indeed, adolescent people have been found to be more affected by reward (e.g. euphoria or social integration consecutive to drug absorption) and less affected by punishment (e.g. malaise after drug consumption) than adults. Whereas the former process is subserved by hyperactivations in regions involved in reward evaluation such as the nucleus accumbens, the latter has been proposed to be subserved by hypoactivations in regions involved in negative emotions such as the amygdala or the insular cortex. This lower sensitivity to losses compared to gains in adolescents could be another important mechanism underlying risk-taking behaviors. A third dimension of adolescents' decision-making biases is temporality. It has been shown that adolescents favor immediate over delayed prospects, reflecting how future consequences of their decisions are heavily discounted. For example, adolescents can fail in projecting the future benefits of having safe sex - and thereby avoiding the risk of sexually transmitted disease or pregnancy - being more interested in the immediate reward of having romantic uninterrupted sexual intercourse. This impairment in inhibiting the choice of the early alternative could be related to the hypofunctionality of the lateral prefrontal cortex. Importantly, these three biases in the evaluation of decisions by adolescents may be related to the maturation of two neuronal systems. On the one hand, the early reorganization of dopaminergic neurons in the motivational system, due to the brutal secretion of sex hormones (mostly estrogens, testosterone and oxytocin) at the beginning of puberty, impels adolescents toward thrill seeking. On the other hand, the slow maturation of the cognitive control system, mostly exerted by the prefrontal cortex, implies that these impulses cannot be appropriately regulated.

Two important neurodevelopmental mechanisms are thought to play a key role in the genesis of risk-taking behaviors in adolescence: the brutal secretion of sex hormones at the beginning of puberty and the delayed maturation of cognitive control. As such, these behaviors can be considered as inevitable, even if other factors, like sex, heredity and precariousness, can enhance their frequency. The implications of these conclusions for the prevention of risk-taking behaviors in adolescence are discussed.