Beta power increased from response levels at selected pairs in eight SEEG patients under independence and null sign symmetry

This study used public SEEG data from OpenNeuro ds004473 version 1.0.2. It tested beta recovery in eight clinical patients. Earlier trials selected 87 adjacent bipolar pairs with the strongest response suppression. Later trials supplied 942 test trials. The fixed post-response window was nominally 0.6 to 1.0 seconds after the recorded response. Mean beta power rose by 1.075 dB from response levels. The nominal 95 percent patient bootstrap interval was 0.516 to 1.727 dB. The exact two-sided sign-flip test had Holm p = 0.03125 across two tests. The post-response minus baseline mean was -0.036 dB, with interval -0.400 to 0.315 dB and Holm p = 0.867188. This contrast did not establish a baseline overshoot. The tests assume independent patients and effect sign symmetry under the zero null. This is a reuse analysis of the source cohort. It adds separate selection and estimation, source checks, and a reproducible package. The package has linear powers, time samples, code, source versions, results, figures, and instructions. Fixed windows and clinical contact coverage limit wider claims.

beta power · stereo-EEG · SEEG · local bipolar pairs · motor response · trial split · power rebound · sign-flip test