Individualized Prediction of Reading Comprehension Ability Using Gray Matter Volume

摘要: Reading comprehension is a crucial reading skill for learning and putatively contains 2 key components: decoding linguistic comprehension. Current understanding of the neural mechanism underlying these components lacking, whether how neuroanatomical features can be used to predict skills remain largely unexplored. In present study, we analyzed large sample from Human Connectome Project (HCP) dataset successfully built multivariate predictive models using whole-brain gray matter volume features. The results showed that effectively captured individual differences in were able significantly unseen individuals. strict cross-validation HCP cohort another independent children demonstrated model generalizability. identified regions contributing prediction consisted wide range covering putative reading, cerebellum, subcortical systems. Interestingly, there gender models, with female-specific overestimating males' abilities. Moreover, male-specific exhibited considerable differences, supporting gender-dependent substrate