A one-second window shift left 4.66 percent of offline Fourier theta crossings unmatched within 25 ms in 21 SEEG patients

We tested whether a one-second shift of a three-second data window changes possible stimulation times. We used offline 4 to 8 Hz Fourier phase from 35 SEEG records in 21 patients with epilepsy. At zero degrees, 4.66 percent of scored crossings had no one-to-one partner within 25 ms. The descriptive 95 percent person bootstrap interval was 4.06 to 5.30 percent. All 21 people had unmatched events. The matching limit, window weight, and separate amplitude gates changed the result. The package adds event-level crossing and matching code, three figures, full event tables, and a verified source retrieval recipe. All 105 raw files passed size and hash checks. Every rebuilt input array matched its archived source. A separate solver checked all 17,540 matching cases. Two independent root scans and a raw-data replay passed. This shared-cohort follow-up measures disagreement between offline reference schedules. It does not measure error against true brain phase or test a live controller or stimulation benefit.

SEEG · intracranial EEG · Fourier phase · theta · phase crossings · window context · event matching · brain-computer interface