The impact of action video games on statistical learning

Based on the findings that video games appear to teach gamers how to perform statistical inference more accurately, leading them to quickly find the accurate statistic for a new task, we have compared the performance of the action video games players (VGP) and non-gamers (NVGP) on statistical learning tasks. First, in an artificial grammar learning (AGL) task (adapted from Reeder, Newport, & Aslin, 2013), we have assessed whether the VGP were better at extracting regularities in the input and better in the learning of exceptions than the NVGP. Globally, we have not found any differences between groups on the extraction of regularities and in the learning of exceptions. Secondly, we have led an exploratory analysis on the visual statistical learning (VSL) task (Siegelman, Bogaerts, & Frost, 2017), to assess whether action video games players were better at extracting transitional probabilities contains in a continuous input. As for the AGL task, the VGP were not better than the NVGP. Hence, there are no differences that have been found between the VGP and the NVGP on statistical learning