concept of Granger causality (Geweke, 1982; Kaminski and Blinowska, 1991; Sameshima and Baccalá, 1999; Marinazzo et al., 2008). In addition, methods to identify brain states can provide further insight into brain networks and especially into their dynamics. In this framework, microstate analysis identifies sensor level topographies that remain stable for a certain period of time before transitioning to a different topography. Changes in the topography are assumed to indicate a reorganization in the global coordination of neuronal activity over time. This approach has a long lasting tradition in EEG, but it has seen a new flourishing in the recent years due to close link to brain network analysis. For timely reviews see Michel and König (2018) and He et al. (201. Similarly, Hidden Markov Model (HMM) approaches reveal the transition of brain networks between states that recur at different points in time. This approach has been used in MEG source space data in Baker et al. (2014) for network identification through amplitude based correlation, and in Vidaurre et al. (2018) to identify phase coherence based networks. A more detailed description of all these methods for assessing brain connectivity and brain networks would require a dedicated review paper and it is
