Encoding HFA showed no reliable primacy gradient in the analyzed pure-SEEG free-recall samples

We tested whether encoding high-frequency activity (HFA, 70–150 Hz) was higher for words at serial positions 1–2 in published per-word SEEG features from two partial free-recall cohorts. In the primary catFR1 sample (38 patients), HFA was 0.038 SD higher at positions 1–2 after control for recall and list (95% CI 0.009 to 0.068; p = 0.017; Holm p = 0.052), so the pre-specified criterion was not met. Theta was lower by 0.045 SD (Holm p = 0.052). The HFA effect did not replicate in FR1 (22 patients; +0.013 SD, p = 0.61); 11 patients overlapped between samples. Behavioural primacy was strong, but the early HFA effect did not correlate with behavioural primacy across patients (rho = −0.11, p = 0.51). A within-list permutation test in the primary sample gave p = 0.005 but used the same 38 patients. These whole-brain averages do not show a reliable encoding HFA primacy gradient that explains behavioural primacy; an effect smaller than 0.07 SD remains possible. Features cover a single 0–1.6 s window, include white-matter pairs, and have no artifact or interictal epileptiform discharge rejection. Samples are partial, and the FR1 cohort is not an independent replication.

SEEG · free recall · primacy effect · high-frequency activity · intracranial EEG · episodic memory · serial position