An adjusted model linked preceding exported errors to lower hippocampal fixation theta power in four-letter trials from 10 SEEG patients

We tested the association between a preceding exported error and hippocampal fixation theta power. Public SEEG came from OpenNeuro ds004752, version 1.0.1. The fixed primary analysis used 332 four-letter trials from 10 patients. Each patient model adjusted for previous load, session, and trial order. Each patient had equal group weight. The mean adjusted difference was -1.234 dB. Eight patient estimates were negative. The exact two-sided patient sign flip p value was 0.042969. The 95 percent t interval was -2.494 to 0.026 dB. The sign flip test requires independent patients and sign symmetry under the null. The t interval requires approximately normal patient coefficients. These procedures give different decisions at 0.05. The evidence is weak. All nine planned descriptive check means were negative. Original acquisition adjacency is unverified. The result concerns the preceding exported trial and does not identify a causal error response. The patients and spectra were reused from an earlier archive. Fresh public downloads verified all 555 primary current or previous fixation records. The package preserves the fixed plan, code, spectra, raw examples, metadata, source references, results, audit receipts, and repeat instructions. A clean offline run reproduced all 12 analysis output files byte for byte.

stereoelectroencephalography · hippocampus · theta power · fixation · trial history · error monitoring · verbal working memory · patient sign flip test