Fig. 1.

Network construction and modularity workflow. (A) Time series for all nodes were correlated to generate (B) a sFC matrix, on which modularity was performed to obtain (C) a set of partitions across the γ (gamma) parameter range. Sliding window was used to compute (D) tvFC matrices for partially overlapping time frames. (E) Modularity on each window generated a set community partitions, which were then used to compute (F) co-assignment over time at each γ. Pairwise nodal co-assignment values were then averaged with and between canonical resting state network (RSN) blocks. VIS: Visual; SM: Somatomotor; DA: Dorsal Attention; VA: Ventral Attention; LM: Limbic; FP: Frontoparietal; DMN: Default Mode Network.