Mean simultaneous 0.5 to 20 Hz SEEG excursion rates exceeded controls with fixed cross spectra in nine selected adults

We tested whether a reference with fixed pairwise spectra reproduces simultaneous large SEEG excursions. The test used 18 public rest and film records from nine selected adults with epilepsy. The source was OpenNeuro ds003688 version 1.0.7. We used a 0.5 to 20 Hz band and two 50 second blocks per record. Each channel was an adjacent contact difference. No scored pair shared a source contact. The event rule selected the largest 5 percent of absolute samples in each channel. We drew 499 common phase controls per block. They preserved every Fourier cross spectrum. The observed patient mean was 6.158 simultaneous excursions per 1,000 pair-samples. The reference mean was 4.511. The mean excess was 1.647, with a descriptive 95 percent patient bootstrap interval of 0.633 to 2.843. Eight of nine patient effects were positive. The exact two-sided sign-flip p value was 0.01953. This test requires independent patient effects and joint sign symmetry. Independent channel phases gave about 2.22 times the primary excess. Those controls test a different model. The primary result shows a limit of this conditional spectral reference in these selected records. It does not establish nonlinear brain coupling or clinical value. The package contains the compact data, source code, all reference scores, numerical checks, source hashes, figures, and reproduction instructions. Both full analysis runs gave identical saved numerical results. All imported file hashes passed. One fresh native range reproduced its compact array exactly. Short retrieval attempts for all 18 ranges timed out. Their errors are retained. This version fixes the package file check after AutoSci retrieval.

stereoelectroencephalography · SEEG · simultaneous excursions · common phase surrogate · Fourier cross spectra · epilepsy · signal processing