Impact of high-frequency sampling rate and stimulation intensity on early TMS artifacts: considerations for immediate TMS-EEG responses

In studies combining transcranial magnetic stimulation and electroencephalography (TMS-EEG), two artifacts appear instantly after the TMS pulse, i.e., the TMS-pulse Artifact and the Decay Artifact, and limit the possibility to measure immediate cortical excitability responses. High-frequency sampling rates in EEG recordings have shown promise in reducing artifact duration, allowing more rapid signal recovery. However, the features of early TMS-induced artifacts for sampling rates above 5000 Hz are still not fully characterized. Here, we explored the duration of TMS artifacts to understand how they can be further reduced to allow the measure of immediate TMS-EEG responses (i.e., 2-3 ms after the TMS pulse). We recorded from two phantom head models, namely a melon and a simple electrical circuit, with a sampling rate of 4800 Hz, 9600 Hz, and 19,200 Hz and at three TMS intensities (40 %, 70 %, 100 % of maximum stimulator output) in two commercial stimulators. A non-parametric Aligned Rank Transform Analysis of Variance was applied with a 2 × 3 × 3 design, with Phantom, Sampling rate and TMS Intensity as factors. Results showed an initial sharp TMS-pulse Artifact lasting <1 ms and decreasing in duration at higher sampling rates (p < 0.001). However, the signal was back to baseline at about 2-3 ms due to the presence of a Decay Artifact that was evident even in optimal conditions of low impedance and mostly dependent on stimulation intensity (p < 0.001). These results highlight the need to develop efficient ways to eliminate the Decay Artifact in order to measure immediate TMS-EEG responses.