from the larger group. Presented results represent averages across all between-hemisphere electrode pairs within each session, and across all sessions. As a control, this analysis was also performed on data reversed in time (Figure S4). This is expected to reverse the direction of true causality, but preserve the direction of any apparent causality due to confounding between-area differences in power or signal-to-noise (Bastos and Schoffelen, 2016; Haufe et al., 2013). Inference was based on the conjunction of significance in the standard analysis and significance in the opposite causal direction in the time-reversed analysis. This analysis was performed using the FieldTrip toolbox for MATLAB (Oostenveld et al., 2011).
