The Rhythm of Normality: A Comprehensive Normative Database for TMS-EEG Metrics with Reliability Characterization
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background
Transcranial Magnetic Stimulation combined with electroencephalography (TMS-EEG) offers unique insights into cortical excitability and connectivity, yet current analyses are primarily limited to group-level inferences with little validation of individual reliability and feature redundancy.
Objective
To construct a comprehensive, open-access, and reliable normative dataset of TMS-EEG features that enables individual-level comparison
Methods
We aggregated TMS-EEG data recorded over the primary motor cortex (M1) from 164 healthy adults (30.8 ± 9.8 years; 88 female) across nine studies using harmonized acquisition and preprocessing pipelines. Reliability analysis was conducted on a test-retest subset (N=57) for 968 extracted features, evaluating systematic bias, absolute error, and relative reliability (Intraclass Correlation Coefficient categorized by the lower bound of the 95% confidence interval). Additionally, feature clustering was performed to quantify redundancy and correlations across the high-dimensional feature space. We then established normative distributions and developed an online benchmarking platform.
Results
Reliability analyses (N=57) of the high-dimensional feature set revealed that 525 out of 968 features (54.3%) met at least moderate reliability standards (ICC lower bound > 0.5). Cluster analysis indicated substantial redundancy among metrics, with three distinct clusters having a moderate-to-high internal correlation (| r | = 0.64, 0.48, 0.39, respectively). Finally, normative data from the database identified abnormal results in a test patient, supporting the feasibility of individual-level classification in an open-science framework.
Conclusions
Harmonization of data acquisition and analysis pipelines led to the development of a reliable normative M1 TMS-EEG reference. This publicly available resource provides a validated tool for future individual-level classifications and an open platform for ongoing community contributions.
