The Golden Age of Online Readout: EEG-Informed TMS from Manual Probing to Closed-Loop Neuromodulation

Over the past four decades, the integration of transcranial magnetic stimulation (TMS) withelectroencephalography (EEG) has substantially enhanced our ability to probe and characterize thebrain’s dynamic responses to external perturbations. This approach holds promise for both basicresearch and clinical applications, yet two challenges remain: (i) reliably eliciting TMS-evokedpotentials (TEPs) with a high signal-to-noise ratio (SNR), particularly early components, whileminimizing time-locked artifacts (e.g. auditory, muscular, and somatosensory responses); and (ii)improving the efficacy of TMS for both acute and long-term neuromodulation. EEG-informed TMSsetups have emerged to address the first problem, providing operators with electrophysiologicalfeedback online through specialized software and imaging readouts. This setup allows forcustomization of stimulation parameters based on individual anatomical and functional variability,to improve the TEP’s SNR and reduce both endogenous and exogenous artifacts. In parallel, systemsneuroscience has shown that oscillatory excitatory and inhibitory dynamics exert a significantinfluence on cortical plasticity. This has motivated the development of EEG-triggered TMSsystems, in which stimulation is synchronized with ongoing neural activity to improve efficacy.Despite progress, questions remain about how, when, and where to stimulate to maximize effectsand minimize variability. Closed-loop platforms are emerging to meet this need, using real-timeneuroimaging readouts to dynamically guide personalized stimulation. This review traces theevolution of TMS-EEG, comparing EEG-blind and EEG-informed setups, and highlighting the roleof EEG-triggered algorithms and closed-loop control in advancing the precision and reproducibilityof probing and neuromodulation strategies.