List resets in catFR1 SEEG: HFA and theta replicate in 23 patients, but only theta replicates in 12 new patients

We tested whether the list-locked reset of serial-position effects in word-encoding high-frequency activity (HFA, 70–150 Hz) and theta power (3–8 Hz), previously reported in 51 FR1 SEEG patients, also occurs during categorized free recall (catFR1). We analyzed archived word-locked band-power features for 23 catFR1 patients (session 0; all bipolar pairs; 0–1.6 s after word onset), using the unchanged reset-test code from the earlier FR1 study. In the primary set, power from the end of one full 12-word list to the start of the next changed by +0.069 SD for HFA (95% CI +0.023 to +0.115; 18/23 with the FR1 sign; Holm p = 0.010) and −0.229 SD for theta (95% CI −0.329 to −0.132; 16/23; Holm p = 0.00075). After controlling for linear session time, position remained associated with HFA (−0.0052 SD per position; 95% CI −0.0082 to −0.0024; Holm p = 0.0043) and theta (+0.0203; 95% CI +0.0137 to +0.0274; Holm p = 0.0002). Reset plus within-list change was near zero, and the reset exceeded that predicted by linear session drift. The HFA reset was about half the FR1 estimate. In 12 patients absent from the FR1 sample, theta reset replicated (−0.272 SD; Holm p = 0.016), while HFA reset was not significant (+0.045 SD; 95% CI −0.010 to +0.100; Holm p = 0.21). Eleven of 23 patients overlapped with FR1, only 23 of at least 44 catFR1 SEEG patients had archived features, and the analysis averaged across all bipolar pairs and used session 0 only. The exact ds004809 release underlying the archived features is not recorded.

categorized free recall · catFR1 · SEEG · serial position · high-frequency activity · theta power · list reset · memory