FR1 and catFR1 pre-recall hippocampal HFA differences were not significant after Holm correction

We compared archived per-subject results from 95 patients in RAM free recall (FR1) and 46 different patients in categorized free recall (catFR1) to test whether the pre-recall hippocampal 70–150 Hz power increase was smaller in catFR1. In the 0.5 s before correct recall, the FR1 mean was +0.128 dB and the catFR1 mean was −0.012 dB. The difference was +0.141 dB (95% CI −0.011 to 0.296; permutation p = 0.037; Holm-adjusted p = 0.074), so it did not meet the pre-specified significance rule. With a time-matched baseline, the difference was +0.115 dB (p = 0.19). Correct-minus-intrusion effects were similar across tasks (+0.118 and +0.132 dB; p = 0.82). A secondary covariate model found a +0.177 dB task coefficient (p = 0.008, uncorrected). A timing check in three catFR1 participants found onset differences within recall periods differed from sample/fs by at most 0.028 s (238 events); the full-session shift reached 3075 s in one participant. This check supports timing accuracy to about 30 ms in the checked periods but covered only three participants. The results provide weak evidence that categorized lists remove the pre-recall increase. Different patient samples allow person-level confounding, and a larger catFR1 sample is needed.

hippocampus · high-frequency activity · free recall · categorized free recall · intracranial EEG · permutation test · RAM dataset