Decision-making in young athletes: how do sports children adapt to a non-sports uncertain environment?

Purpose: Physical and sports activities could enhance cognitive function in children and adults (1,2). However, the possibility of improving the high- order cognitive function, such as decision-mak- ing, is under debate. The study aims to fill this gap by testing decision-making processes featuring young open- and closed-skill sports athletes. In addition, fluid intelligence was monitored for its possible effect on these tasks. Methods: Thirty-five young high-level athletes (13 Track and Field Athletes [TFA] and 22 football players; Mage = 11.03, SD = 1.33y.o.) were recruited. We involved a perceptual decision task under uncertainty (i.e., low- medium- high uncertainty) where participants had to score as many points as possible. Their performance and decision confidence were analyzed. Two indexes (i.e., Spatial Error (SE) and Gain) evaluated performance. At the same time, implicit decision confidence was assessed through the bet on their decision and through the comparison with an Optimal User Model (SD-Error). Raven- APM(3) assessed fluid intelligence. Results: SE was better in low uncertainty compared to medium and high uncertainty. Again, it was better in medium uncertainty compared to high uncertainty. However, nonsignificant differences between the two groups in each level of uncertainty were found. The confounding factor ‘‘intelligence’’ was nonsignificant. Gain analysis revealed the factor ‘‘uncertainty’’ was significant (F=97.19, p\.001). The greater the uncertainty, the fewer points gained. Even if the interaction ‘‘group x uncertainty’’ was significant (F=7.77, p\.001), nonsignificant differences between the two groups in the three uncertainties emerged. The confounding factor ‘‘intelli- gence’’ was nonsignificant. Confidence analysis showed that the two groups behaved differ- ently among the three levels of uncertainties (F=8.42, p\.001). Football players modulated the bet according to uncertainty and bet more points than TFA. SD-E analysis confirmed that football athletes were closer to the optimal model than TFA. Conclusions: The SE and Gain performance was similar between the groups. However, implicit confidence and SD-Error showed that football players were better at managing the environment than TFA. This discrepancy could be due to the pay-out matrix adopted. Thus, in the subsequent studies, the involvement of different pay-out matrices will help to understand these results better. Furthermore, the recruitment of sedentary children will better understand the effect of sports practice on high-order cognitive processes. References: 1)Voss et al. (2010). 2)Gu et al. (2019). 3)Raven & Court (1998b).