The beta phase test did not reject zero at selected SEEG pairs in eight patients under patient independence and sign symmetry

We tested beta phase consistency near a keypress in OpenNeuro ds004473 version 1.0.2. The sample had eight patients with epilepsy and 64 trials per patient. The first 32 trials selected adjacent SEEG pairs with lower beta power. The last 32 trials tested 167 selected pairs. Mean response minus baseline pairwise phase consistency was -0.002069. The descriptive 95 percent patient bootstrap interval was [-0.004901, 0.000653]. The exact two-sided patient sign-flip test did not reject zero. Its Holm p value was 0.21875. The test assumes independent patients and sign symmetry. Mean test beta power change was -0.537 dB, with Holm p = 0.10938. Neither planned test rejected zero at 0.05. This does not prove absence or equivalence. The study is a secondary analysis of a prior public trial sample. The package adds complex Fourier data for 985 contacts and preserves code, source versions, hashes, raw examples, results, figures, and retrieval instructions. Fresh source checks, direct numerical checks, and an offline rebuild passed.

SEEG · beta oscillations · pairwise phase consistency · phase · movement · electrophysiology · secondary analysis