Cross-SessionNon-Stationarityin Motor-ImageryBCI:SurveyEvidence, GeometricStructure,andDecoder-Relevant Shift
This repository provides reproducible Python code and supporting results for the study “Cross-Session Non-Stationarity in Motor-Imagery BCI: Geometric Structure, Decoder-Relevant Shift, and a Past-Only Benchmark.” It investigates cross-session EEG variability using geometric decomposition, decoder-relevant shifts, and strictly past-only adaptation. The repository includes a complete Jupyter notebook, benchmark results, preprocessing sensitivity analyses, statistical evaluations, and longitudinal analyses across five publicly available EEG datasets. It supports independent verification and reproducibility of the manuscript's findings. Raw EEG recordings are not redistributed.
Authors
- Niraj Kumar Dewangan (ORCID: https://orcid.org/0000-0002-1001-4858)
- Ritesh Kumar Kushwaha (ORCID: https://orcid.org/0000-0002-5187-1514)
Institutions
- Manipal Academy of Higher Education (IN)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-19
- DOI
- https://doi.org/10.5281/zenodo.24372229
- Primary Topic
- EEG and Brain-Computer Interfaces
- Type
- article
- Field-Weighted Citation Impact
- 0.00