In 36 ds004789 patients, unlabelled SEEG pairs showed a nominal theta effect that did not survive correction

We tested the word-encoding subsequent memory effect (SME) in archived sample/fs-timed bipolar band-power features from 36 patients in the delayed free-recall task (OpenNeuro ds004789, session 0). We compared pairs with no Desikan–Killiany atlas label (470 pairs; ‘n/a’) with labelled pairs (2,898 pairs). The label ‘n/a’ is not a verified white-matter classification. Across patients, the unlabelled-pair theta SME was −0.069 SD (two-sided sign-flip p = 0.015; Holm-adjusted p = 0.058). The unlabelled-pair high-frequency activity (HFA; 70–150 Hz) SME was +0.014 SD (p = 0.41; Holm p = 1.0). Labelled-minus-unlabelled differences were not significant for HFA (+0.009 SD; Holm p = 1.0) or theta (−0.014 SD; Holm p = 1.0). The unlabelled/labelled ratio for theta was 0.83 (bootstrap 95% CI 0.21–1.30); the HFA ratio was 0.61 (95% CI −1.07–2.97). No pre-specified test passed Holm correction. The results are consistent with unlabelled pairs carrying a theta SME, but do not establish that this signal is local to white matter or outside cortex. The study used one task and session in 36 patients; bipolar pairs combine two contacts, and some unlabelled pairs may be in deep gray matter or outside the brain.

subsequent memory effect · stereo-EEG · SEEG · theta · high-frequency activity · volume conduction · free recall · atlas labels · OpenNeuro ds004789