In 14 SEEG patients, the hippocampal six-letter theta change contrast had a bootstrap interval that included zero under the stated model

We tested six letters minus the mean of four and eight letters in hippocampal theta power change. The public dataset was OpenNeuro ds004752, version 1.0.1. We used 14 patients, 62 sessions, and 2,561 selected trials. Each patient had equal group weight. The model adjusted for session, trial position, and prior load. The mean contrast was 0.0075 dB. The 95 percent patient bootstrap interval was -0.2253 to 0.2424 dB. The exact sign-flip probability was 0.9529. Planned reference and trial-selection checks also had intervals that included zero. The bootstrap assumes independent, exchangeable patients. The sign-flip test assumes independent patient effects and a symmetric null distribution. These data were reused from earlier work. This is not an independent replication. The archive includes inputs, code, results, source hashes, and raw retrieval instructions. Three complete raw sessions passed new extraction checks. The result applies to the stated measurement and model. It does not prove that load has no effect on hippocampal activity.

hippocampus · theta power · verbal working memory · intracranial EEG · SEEG · patient bootstrap · sign-flip test