In 71 pure-SEEG patients, left-hemisphere 70–150 Hz memory effects were null and pair effects matched chance (FR1, session 0)

We tested whether the previously reported null encoding subsequent memory effect (SME) for high-frequency activity (HFA; 70–150 Hz) in pure-SEEG free recall arose from averaging across hemispheres. We analyzed archived per-word power features from 83 patients in OpenNeuro ds004789 (FR1, session 0; 0–1.6 s after word onset) and MNI electrode tables whose source URLs and SHA-256 hashes are recorded. Pair SME was the Fisher z transform of the partial correlation between z-scored power and recall after removing list and serial-position effects. Among 71 patients eligible for the left-hemisphere test, mean HFA SME was +0.0026 (95% CI −0.0022 to +0.0074; Holm p = 0.83). Among 53 patients with both hemispheres, left-minus-right HFA SME was +0.0052 (95% CI −0.0056 to +0.0183; Holm p = 0.83). Left theta SME was positive (+0.0046), not negative (Holm p = 0.83). No band or region-by-hemisphere HFA cell was significant after correction. The share of pairs with |t| > 1.96 was 5.5%, compared with 5.1% under within-serial-position label permutation (p = 0.35). These archived features do not support a positive left HFA SME or a left advantage, and hemisphere averaging does not explain the earlier all-pair null. The result applies to this feature pipeline, pure-SEEG bipolar pairs, and the 0–1.6 s window; event timing was not checked against raw data, and language dominance was unknown.

SEEG · free recall · subsequent memory effect · high-frequency activity · hemisphere · intracranial EEG · OpenNeuro