On the variability of functional connectivity and network measures in source-reconstructed EEG time-series.

The idea to estimate the statistical interdependence among (interacting) EEG signals has motivated numerous researchers to investigate how the resulting networks may reorganize themselves under any conceivable scenario. Even though this idea is not at initial stages, its application is still far to be widespread. One concurrent cause may be related to the proliferation of different approaches that promise to catch the underlying correlation among the (interacting) units. This issue has probably contributed to hinder the comparison among different studies. Not only all these approaches go under the same name (functional connectivity) but they have been often tested and validated using different methods, therefore, making it difficult to understand to what extent they are similar or not. In this study, we aim to compare a set of different approaches commonly used to estimate the functional connectivity on a public EEG dataset representing a possible realistic scenario. Our results show that source-level EEG functional connectivity estimates and the derived network measures display a substantial dependency on the arbitrary choice of the selected connectivity metric. The observed variability reflects ambiguity and concern that should be always discussed when reporting findings based on any connectivity metric.