Variable E-field properties of dual-site tACS

Background
Dual-site transcranial alternating current stimulation (ds-tACS) enables the modulation of interregional functional connectivity by introducing a phase lag between the stimulating currents. However, overlapping electric fields (E-fields), particularly in closely spaced cortical targets like the primary motor cortices (M1s), may unintentionally alter E-field characteristics and confound the interpretation of functional connectivity modulation.

Objective
We aimed to systematically evaluate how different phase lags affect key E-field characteristics when using high-definition ds-tACS, particularly when targeting the M1s. We sought to determine which montage configuration best preserved stable E-field characteristics and investigated whether indi-vidualised montage selection could enhance control over E-field consistency.

Methods
We used individualised finite-element method simulations based on MRI-derived head models to quantify the effects of different phase lags on E-field characteristics. E-field magnitude, normal component, spatial distribution, directionality, and effective stimulation area were assessed for nine montages and eight phase lags.

Results
All E-field properties, including E-field peak magnitude, peak magnitude of the normal component, redistribution, difference to optimal directionality, and effective area of stimulation, were modulated significantly across differ-ent phase lags for all tested montages. Furthermore, we found substantial inter-individual variability in all E-field properties. Individual selection of montages improved critical properties, particularly the E-field directionality.

Conclusions
In contrast to common assumptions, variations in the phase lag can significantly affect key E-field properties of high-definition ds-tACS. Therefore, we recommend considering modulations of the E-field characteristics when comparing physiological or behavioural effects of ds-tACS at different phase lags. Moreover, given the high inter-individual variability, we suggest the individualisation of montages to the most relevant E-field property.