Assessment of whole-head magnetoencephalography during transcranial electric entrainment of brain oscillations

Abstract

Application of non-invasive brain stimulation for perturbing brain activity is well established. Various forms of brain stimulation protocols have been effectively demonstrated to modulate behavior associated with the perturbed brain activity. However, the interaction of brain stimulation with ongoing brain activity has been challenging to characterize as the stimulation artifacts in the recordings of brain activity impedes such characterization. The proposed amplitude modulated transcranial alternating current stimulation (tACSAM) attenuates possible stimulation artifacts at the frequency of interest. This is possible by modulating the amplitude of high frequency transcranial alternating current (tACS) signal at a lower physiological frequency of interest to generate the tACSAM signal. Furthermore, application of tACSAM allows localization of the perturbed brain activity with millimeter precision by applying spatial filters on magnetoencephalography (MEG) recordings. For characterization of the tACSAM-perturbed brain activity, conventional spectral analysis may not be sufficient. Thus, power and PLV were compared between tACS and tACSAM in a phantom model and MEG data recorded from healthy human volunteers. The synchronization estimate, phase lock value (PLV), is a measure of circular variance between two signals calculated as a function of instantaneous phase difference between the ongoing brain activity and the applied stimulation signal. Even though, systematic linear phase shifts due to the applied tES signal occur in MEG sensors, mathematically such systematic linear phase shifts nullify while calculating PLV. Systematic evaluation of the MEG data acquired during tACSAM showed increased PLV compared to tACS indicating increased demodulation in such paradigm. Upon observing tACSAM-related increased demodulation, it was still unclear whether such perturbations of brain activity could modulate behavior. To address this question, twenty volunteers while engaging in a working memory paradigm received tACSAM or no stimulation. Working memory is associated with transient storage and processing of information. Increasing the difficulty of working memory paradigm increases the amplitude of brain activity in the theta band (4 – 8 Hz), while encoding the temporal order of the transient information in the phase of the theta activity. Thus, by targeting individual’s theta peak frequency using tACSAM, it was possible to modulate the accuracy in the working memory paradigm. The accuracy on a working memory parading of volunteers receiving tACSAM deteriorated compared to the participants who did not receive brain stimulation. Therefore, targeting brain activity in theta band using tACSAM interferes with execution of normal working memory processes, probably by interfering with the maintenance of temporal order of the transient information. Furthermore, tACSAM but not sham stimulation inhibited the increase in amplitude of theta activity during the n-back task, which is essential for working memory processes. Even though, it is possible to assess the brain activity recorded during tACSAM, presence of stimulation artifacts in the assessed brain activity cannot be excluded. However, it was possible to gather evidence that tACSAM is associated with demodulation. TACSAM-induced phase synchrony at the modulation frequency was larger compared to tACS even though the power during tACS is larger compared to tACSAM. This observation is in favor of possible functional interaction of tACSAM signal with neurons in the brain. However, currently it is not possible to distinguish between the contribution towards demodulation of tACSAM signal by non-linearities of the stimulation setup and functional interactions with neurons in the brain. In conclusion, tACSAM can alter cognitive function, such as working memory performance, possibly through entrainment. The results obtained from such investigations must be interpreted with great care, as the extent by which possible stimulation artifacts impact the MEG recordings is not entirely clear. Further investigations are necessary to develop quantitative assessment techniques for characterizing artifacts of the stimulation and eventually develop brain state dependent stimulation paradigms in real time as a research tool and therapeutic intervention.