Fig. 3
Display of three parameters related to graph efficiency in the thresholded 60 ROI WU graph. (A) The upper subgraph shows global efficiency which is related to the average inverse shortest path. (B) The middle subgraph shows local efficiency, which is computed from the neighbourhoods of all nodes and related to the clustering coefficient. (C) The small world propensity indicates the extent to which a network is characterized by high local clustering and low average path length. Note (1) a threshold of 0.45 is the optimal trade-off between high global and local efficiencies and associated high small world propensities, of which the latter are superior to those of the 113-ROI graphs (for comparison see the corresponding figure in the repository), (2) the stable evolution of small world propensity until a threshold of 0.45 for all conditions, beyond which it becomes unstable (dotted lines).
