Key Factors in the Cortical Response to Transcranial Electrical Stimulations—A Multi-Scale Modeling Study

Abstract

Transcranial electrode stimulation (tES), one of the techniques used to apply non-invasive brain stimulation (NIBS), modulates cortical activities by delivering weak electric currents through scalp-attached electrodes. This emerging technique has gained increasing attention recently; however, the results of tES vary greatly depending upon subjects and the stimulation paradigm, and its cellular mechanism remains uncertain. In particular, there is a controversy over the factors that determine the cortical response to tES. Some studies have reported that the electric field’s (EF) orientation is the determining factor, while others have demonstrated that the EF magnitude itself is the crucial factor. In this work, we conducted an in-depth investigation of cortical activity in two types of electrode montages used widely—the conventional (C)-tES and high-definition (HD)-tES—as well as two stimulation waveforms—direct current (DC) and alternating current (AC). To do so, we constructed a multi-scale model by coupling an anatomically realistic human head model and morphologically realistic multi-compartmental models of three types of cortical neurons (layer 2/3 pyramidal neuron, layer 4 basket cell, layer 5 pyramidal neuron). Then, we quantified the neuronal response to the C-/HD-tDCS/tACS and explored the relation between the electric field (EF) and the radial field’s (RF: radial component of EF) magnitude and the cortical neurons’ threshold. The EF tES induced depended upon the electrode montage, and the neuronal responses were correlated with the EF rather than the RF’s magnitude. The electrode montages and stimulation waveforms caused a small difference in threshold, but the higher correlation between the EF’s magnitude and the threshold was consistent. Further, we observed that the neurons’ morphological features affected the degree of the correlation highly. Thus, the EF magnitude was a key factor in the responses of neurons with arborized axons. Our results demonstrate that the crucial factor in neuronal excitability depends upon the neuron models’ morphological and biophysical properties. Hence, to predict the cellular targets of NIBS precisely, it is necessary to adopt more advanced neuron models that mimic realistic morphological and biophysical features of actual human cells.