TLMM v4.9: Exploratory Multi-Channel EEG Twin Framework with Test–Retest, Multi-Dataset, and Uncertainty-Aware Inference

This Zenodo record contains TLMM v4.9, an exploratory extension of the Time-Lagged Mutual Modeling (TLMM) framework toward multi-channel, multi-dataset, test–retest, and uncertainty-aware EEG twin inference. The framework extends the previous public EEG contact stage (v4.8) by incorporating: • Multi-channel EEG envelope extraction• Cross-channel coupling analysis• Representative null-model comparison• Threshold-dependence and robustness analysis• Test–retest reproducibility assessment• Subject-specific Edge Perturbation Index (EPI)• Multi-dataset consistency analysis• Streaming adaptive twin simulation• Counterfactual scenario simulation• Full-pipeline uncertainty propagation The manuscript evaluates whether inferred potential landscapes, resilience trajectories, and digital twin states can be reproducibly estimated from public EEG data while explicitly tracking uncertainty, threshold dependence, and model assumptions. Public EEG datasets referenced include OpenNEURO, CHB-MIT, and TUAB-style datasets used as exploratory methodological resources. This release includes:• PDF manuscript• Figure set (Fig.1–Fig.10)• Lightweight demonstration code (Python)• README documentation Important scope statement:All analyses are exploratory and hypothesis-generating.No diagnostic, prognostic, therapeutic, medical-device, or clinical utility claims are made.The framework is intended for methodological and computational neuroscience research only. Author:Koji OkinoIndependent Researcher / SD Lab2026