Fig. 3: Associations to cross-task functional diversity.
From: Multimodal gradients unify local and global cortical organization
A The FC matrices were generated using time series data obtained from for nine tasks fMRI sessions in the same subjects. This was followed by the estimation of cosine distances across different tasks. This process resulted in a cross-task diversity matrix of dimensions. To quantify the cross-task diversity for the first vertex of all tasks, the average of each column in this matrix was computed. B By repeating the procedures outlined in (A) for each vertex and subsequently mapping the results to areas, we created an area-wise cross-task diversity map. Two-sided Spearman’s correlation coefficient was computed, and p values were corrected using spin permutation tests. C By taking the average of each row of the vertex-wise FC within area and calculating the standard deviation (SD) across tasks, we generated the vertex-wise cross-task SD. Taking the average of all SDs within an area and repeating this procedure for all areas, we generated the intra-areal cross-task diversity map. Two-sided Spearman’s correlation coefficient was computed, and p values were corrected using spin permutation tests. Source data are provided as a Source Data file. Abbreviation: FC functional connectivity.
