State dependent motor cortex stimulation reveals distinct mechanisms for corticospinal excitability and cortical responses

Transcranial magnetic stimulation (TMS) is a non-invasive brain stimulation method which can modulate brain activity by inducing electric fields in the brain. It is a popular tool to study causal brain-behavior relationships. However, brain states vary over time and affect the response of TMS. Neural oscillations can track the current brain state and are a promising marker to guide stimulation timing. Real-time, state-dependent brain stimulation has shown that neural oscillation phase modulates corticospinal excitability reflecting the connection from the primary motor cortex to a target muscle. However, such motor-evoked potentials (MEPs) only indirectly reflect motor cortex activation and are unavailable at other brain regions of interest. The direct and secondary cortical effects of phase-dependent brain stimulation remain an open question. In this study, we recorded the cortical responses during single-pulse transcranial magnetic stimulation (TMS) using electroencephalography (EEG) concurrently with the MEP measurements. TMS was delivered at peak, rising, trough, and falling phases of mu (8-13 Hz) and beta (14-30 Hz) oscillations in the motor cortex. The cortical responses were quantified through TMS-evoked potential components N15, P50, and N100 as peak-to-peak amplitudes (P50-N15 and P50-N100). We further analyzed whether the pre-stimulus frequency band power was predictive of the motor cortical responses. We found a significant main effect of neural oscillation phase on early evoked component (P50-N15). Furthermore, we found an interaction effect of oscillation phase and frequency on both early and late (P50-N100) components. Next, we compared the direct EEG response to the corticospinal excitability reflected by MEP amplitude. Interestingly, the preferred phase of the mu rhythm showed a 900 phase shift between the early TEP components and MEPs. The late component showed the same phase preference between EEG and MEPs. However, such a well-defined relationship did not exist for either of the components during beta phase specific stimulation. In addition, pre-TMS mu oscillatory power and phase significantly predicted both early and late cortical EEG responses when mu rhythm was targeted, indicating the independent causal effects of phase and power. However, only pre-TMS beta power significantly predicted the early and late TEP components when beta rhythm was targeted. Further analysis indicated that both pre-TMS mu and beta power jointly affect early cortical responses. In contrast, the late cortical responses were only influenced by pre-TMS mu power. These findings provide insight to mechanistic understanding of neural oscillation states in cortical and corticospinal activation in humans.