Go with the flow: use of a directed phase lag index (dPLI) to characterize patterns of phase relations in a large-scale model of brain dynamics.

We introduce a directed phase lag index to investigate the spatial and temporal pattern of phase relations of oscillatory activity in a model of macroscopic structural and functional brain networks. Direction of information flow was determined with the directed phase lag index (dPLI) defined as the probability that the instantaneous phase of X was smaller than the phase of Y (modulo π). X was said to phase-lead Y if 0.5<dPLI(XY)<=1. The dPLI was used to characterize the phase relations between simulated EEG time series. The model consisted of 78 brain regions, coupled according to DTI findings in human subjects (Gong et al., 2009). Activity of each brain region was simulated with a neural mass model. Phase patterns were investigated as a function of coupling strength without stimulation, and with stimulation of the primary visual areas. At rest a clear spatial pattern of phase relations emerged with regions belonging to the anterior part of the default mode network leading in phase and regions belonging to the posterior part of the default mode network and surrounding visual areas lagging in phase. Patterns of phase leading and lagging displayed characteristic patterns with time scales from a few hundred milliseconds to 1-2 s. Stimulation of the primary visual areas induced a reversal of the global phase pattern with the visual and basal temporal areas leading and the anterior and superior frontal areas lagging in phase. This study shows that the directed phase lag index (dPLI) is an effective measure to characterize spatial temporal patterns of phase relations at rest and during stimulation. Coupling strength and node degree were found to be critical determinants of the direction of information flow. At a timescale of milliseconds to seconds the phase dynamics revealed spontaneous structure that might correspond to previously described "microstates". Stimulation of two visual areas reversed the global pattern of phase relations.