In catFR1, theta increased and HFA decreased across list positions, but the HFA effect was smaller and position 1 dependent

We tested whether serial-position changes in encoding power reported for uncategorized free recall (FR1) also occur in categorized free recall (catFR1). We applied the unchanged d3aeb47e analysis to word-locked bipolar band-power features from 23 SEEG patients in OpenNeuro ds004809 (session 0; sample/fs epochs). By the prespecified decision rules, both primary effects replicated: theta increased by 0.0217 SD per position (95% CI 0.0146 to 0.0296; 21/23 positive; Holm p = 0.00015), and HFA (70–150 Hz) decreased by 0.0054 SD per position (95% CI −0.0092 to −0.0021; 16/23 negative; Holm p = 0.0085). The HFA slope was 40% of the FR1 slope (−0.0136). Without position 1, the HFA slope was −0.0029 (p = 0.087); in 12 patients absent from the FR1 sample it was −0.0028 (8/12 negative; Holm p = 0.28), while theta still increased (+0.0225; 11/12 positive; Holm p = 0.0036). Word HFA predicted recall after controlling for serial position (t = +0.49; 95% CI +0.10 to +0.88; 17/23 positive; Holm p = 0.022), compared with t = +0.09 (Holm p = 0.58) in FR1. Exploratory comparisons, hippocampal findings, behavioral results, methods, and limitations are reported in the main text. The sample is small, with 12 new patients, and the catFR1 fetch covered 23 of 44 subjects. The theta increase was consistent across tasks and present in new patients; the HFA decrease was weaker in catFR1 and came mainly from position 1.

serial position · free recall · categorized free recall · intracranial EEG · SEEG · theta · high-frequency activity · encoding · hippocampus