Pair-level HFA memory effects did not repeat; theta effects repeated weakly within 10 mm in 49 FR1 subjects

We tested the spatial repeatability of the subsequent memory effect (SME) across bipolar SEEG pairs in 49 subjects performing FR1. We used archived log band power from 0–1.6 s after word onset (HFA 70–150 Hz, theta 3–8 Hz, and HFA minus theta) and MNI contact coordinates. Cohen d for recalled versus unrecalled words was calculated separately for odd and even lists, then correlated across pairs in distance bins; the halves shared no trials. Pair-level HFA split-half reliability was near zero (Fisher z = −0.002, p = .94), no HFA bin survived Holm correction, and the planned near-versus-far contrast was not supported (difference = 0.011, one-sided p = .32). Theta showed a small positive correlation for pairs 0–10 mm apart (z = 0.068, Holm p = .031), approximately zero beyond 20 mm. Thus, this pipeline did not support estimating an HFA spatial scale; theta showed a weak local pattern at approximately 1 cm. Correlations were small (z < 0.07), each subject had about 150 words per half, and the analysis used MNI coordinates and one archived pipeline. The estimate reflects variation across pairs; Pearson correlation removes a shared global offset.

subsequent memory effect · SEEG · spatial correlation · high-frequency activity · theta · memory · split-half reliability