THALis: February 22 multimodal session bundle

Feb 22 bundle: EEG, cue times, video, overlay, landmarks. make a per-trial map before this turns into matching files by vibes

Archived acquisition (1f8e3eb): the session includes rhythmic and sustained/practice-style trial groups, each with EEG CSVs and associated hand/cue artifacts. The log names t1–t3, p1–p5, e1–e5, and f1–f2 in the initial dump; the next day reorganizes them by paradigm.

Per-trial bundle: EEG data, overlay video, hand-landmark CSV, pose sequence and periodic-trial sequence as applicable. A missing cue sequence is not interchangeable with a missing video: each affects a different part of interpretation.

Review questions: establish whether labels represent instructions or measured motion, align timestamps across modalities, and document tracking/EEG quality exclusions. Trial duration should be computed from timestamps rather than assumed from CSV line count.

Evidence boundary: a large data dump establishes available recordings, not independent participant count or effective sample size. Overlapping windows and repeated trials from the same session must not be mistaken for independent subjects.

Archived code references: 1f8e3eb.