Under patient independence and sign symmetry, the beta coherence pattern test did not detect excess change in 30 SEEG records from 17 patients

We tested whether beta coherence patterns changed more between marked before and after periods than within those periods. We used OpenNeuro ds006519 version 1.0.0. The test used 30 records from 17 patients. We removed the stimulated shaft. We compared centered maps from four matched two-second pieces. Each patient had equal weight. The mean excess squared map distance was 0.000253. The 95 percent patient bootstrap interval was [-0.00000489, 0.000531]. The exact two-sided sign-flip p value was 0.0928. This test needs independent patients and sign symmetry under the null. All eight planned check intervals also included zero. The result does not prove equal patterns. An additive map error model showed that correlated errors can give positive scores without a true map change. We rebuilt every input epoch from the public raw signals. Separate SciPy calculations checked all primary maps and resampling. The package preserves code, epochs, source versions, hashes, metadata, results, and reproduction instructions. The source has no sham condition or precise movement recovery times. The test cannot establish a causal motor mechanism.

stereoelectroencephalography · SEEG · beta coherence · coherence patterns · patient-level inference · sign-flip test · negative motor response