Recent research suggests that video game playing is associated with many cognitive benefits. However, little is known about the neural mechanisms mediating such effects, especially with regard to probabilistic categorization learning, which is a widely unexplored area in gaming research. Therefore, the present study aimed to investigate the neural correlates of probabilistic classification learning in video garners in comparison to non garners. Subjects were scanned in a 3 T magnetic resonance imaging (MRI) scanner while performing a modified version of the weather prediction task. Behavioral data yielded evidence for better categorization performance of video garners, particularly under conditions characterized by stronger uncertainty. Furthermore, a post-experimental questionnaire showed that video garners had acquired higher declarative knowledge about the card combinations and the related weather outcomes. Functional imaging data revealed for video garners stronger activation clusters in the hippocampus, the precuneus, the cingulate gyrus and the middle temporal gyrus as well as in occipital visual areas and in areas related to attentional processes. All these areas are connected with each other and represent critical nodes for semantic memory, visual imagery and cognitive control. Apart from this, and in line with previous studies, both groups showed activation in brain areas that are related to attention and executive functions as well as in the basal ganglia and in memory-associated regions of the medial temporal lobe. These results suggest that playing video games might enhance the usage of declarative knowledge as well as hippocampal involvement and enhances overall learning performance during probabilistic learning. In contrast to non-garners, video garners showed better categorization performance, independently of the uncertainty of the condition.
