Whereas brain networks often show a preserved small-world property even in neuropsychiatric conditions (Bassett and Bullmore, 2009), some network characteristics are affected. The most typical finding from different diseases and modalities has been an increase in path length, which indicates that patients with brain disorders often have an inefficiently constructed network (Lo et al., 2010; Stam et al., 2007a, 2009; van den Heuvel et al., 2010). Similarly, this study found several altered network parameters (C, L, and Eg) in BD patients (Fig. 5), despite intact small-world organization. Specifically, the clustering coefficient C and global efficiency Eg were significantly decreased in the alpha-band over a wide range of thresholds, while characteristic path length L increased. These findings are consistent with a recent tractography-based network study in BD, in which the decreased C, increased L, and decreased Eg were also reported (Leow et al., 2013). In contrast, recent studies using resting-state fMRI and sleep EEG reported decreased path length and increased network efficiency in major depressive disorder (Leistedt et al., 2009; Zhang et al., 2011). These differences in brain network between mood
