Taubert, MarcoZiegler, GabrielLehmann, Nico2023-03-142023-03-142023-03http://open-science.ub.ovgu.de/xmlui/handle/684882692/112https://doi.org/10.24352/UB.OVGU-2023-095behavioral_data contains anonymized subject specific information (e.g. age, gender, body height, TIV), motor performance (residualized learning rate, initial performance) and ROI-wise cortical folding values. SEM contains all anonymized variables to reproduce the four SEM's from the main manuscript. tmap contains unthresholded vertex-wise statistical maps of positive correlations between cortical folding and motor performance (learning rate [exploration and replication].Humans have successfully adapted to various challenging conditions in the past, circumstances that may have influenced the emergence of sophisticated learning mechanisms in the brain. Neuroanatomists have speculated about the impact of neocortical folding patterns on superior human performance, but their role remains unclear. Here we link neocortical folding in premotor regions to improved learning of a challenging new motor skill. Folding strongly predicted individual performance gains during practice. Gains mediated the effect of folding on superior performance. This finding was consistent across three sub-samples and reproducible across time. Higher cortical folding depended, in part, on larger cortical surface area, but not at the expense of lower cortical thickness or intra-cortical microstructure. Folding effects overlapped with practice-induced structural changes in premotor regions and with the prominence of tertiary sulci. Our results contradict the view that superior performances directly result from higher folding. Instead, superior performances emerge when high learning capabilities, reflected in part by larger and more convoluted cortices, can operate in fertile environments.encortical folding, human brain, motor learning, magnetic resonance imaging, stabilometerHigher cortical folding in the human frontal lobe predicts individual differences in long-term motor learningDataset