Local network-level integration mediates effects of transcranial Alternating Current Stimulation
Transcranial alternating current stimulation (tACS) has been proposed as a tool to draw causal inferences on the role of oscillatory activity in cognitive functioning and has the potential to induce long-term changes in cerebral networks. However, the mechanisms of action of tACS are not yet clear, though previous modeling works have suggested that variability may be mediated by local and network-level brain states. We used magnetoencephalography (MEG) to record brain activity from 17 healthy participants as they kept their eyes open (EO) or closed (EC) while being stimulated either with sham, weak, or strong alpha-tACS using a montage commonly assumed to target occipital areas. We reconstructed the activity of sources in all stimulation conditions by means of beamforming. The analysis of resting-state data revealed an interaction of the external stimulation with the endogenous alpha power difference between EO and EC in the posterior cingulate. This region is remote from occipital cortex, which showed strongest EC vs. EO alpha modulation, thus suggesting state-dependency long-range effects of tACS. In a follow-up analysis of this online-tACS effect, we find evidence that this dependency effect could be mediated by functional network changes: connection strength from the precuneus, a region adjusting for a measure of network integration in the two states (EC vs. EO during no-tACS), was significantly correlated with the state-dependency effect in the posterior cingulate (during tACS). No analogous correlation could be found for alpha power modulations in occipital cortex. Altogether, this is the first strong evidence to illustrate how functional network architectures can shape tACS effects.