Under patient independence and sign symmetry, the beta correlation persistence test did not detect a change after stimulation in 30 SEEG records from 17 patients

We tested signed beta correlation before and after stimulation that caused negative motor responses. The source was public OpenNeuro ds006519, version 1.0.0. The sample had 30 records from 17 patients. We removed the stimulated electrode shaft. We multiplied signed correlations from two separate two-second pieces within each marked period. This score measures persistence of signed zero-lag correlation. Each patient had equal weight. The mean after-minus-before change was +0.000469 correlation-product units. The 95 percent patient bootstrap interval was -0.000658 to +0.001631. The exact sign-flip p value was 0.446. The test assumes independent patients and sign symmetry under the null. All seven planned sensitivity intervals included zero. Phase controls gave persistence scores near zero, while squared correlation estimates stayed positive. Fresh public raw downloads reproduced all 30 epoch files exactly. A separate time-domain calculation and an offline repeat passed. The package preserves arrays, code, source hashes, controls, results, figures, and repeat instructions. There was no unstimulated trial control. The result does not establish equal responses or direct neural connections.

stereoelectroencephalography · beta correlation · negative motor response · phase persistence · patient bootstrap · sign-flip test · OpenNeuro ds006519