Fixation theta effect maps after errors had mean correlation -0.095 across 48 sessions in 13 SEEG patients with the source reference

We tested whether spatial hippocampal fixation theta effects after a memory error agree across separate SEEG sessions. Public OpenNeuro ds004752 version 1.0.1 supplied 13 patients with epilepsy, 95 stable contacts, 48 sessions, and 839 selected trials at current load four. Each session model adjusted for prior load and trial position. We removed the contact mean effect and calculated correlations between sessions. The equal patient mean was -0.095. The 95 percent patient bootstrap interval was -0.271 to 0.058, under independent, exchangeable patient sampling. These estimated maps did not show clear positive agreement. This result does not prove absence of a stable underlying effect. All five fixed secondary checks are reported. A common average reference gave mean -0.243; its unadjusted interval excluded zero. Few errors per session, clinical contact coverage, and reference choice limit interpretation. The sign-flip p value of 0.322 needs a separate null sign symmetry assumption. A score bound explains why zero mean does not guarantee that assumption. A local plan preceded the new maps. Independent checks covered every fitted model and pair score. Fresh checks of 196 raw windows matched 3380 power values. An offline repeat gave 12 identical numerical files. The package preserves exact inputs, source versions, retrieval code, raw examples, maps, results, figures, checks, and instructions. This is a new task history test in a previously studied cohort.

hippocampus · theta power · memory errors · spatial effects · intracranial EEG · SEEG · session agreement · working memory