A test that assumed independent patients and symmetric null effects did not detect a near versus far difference in masked 80–150 Hz SEEG power changes around 43.2 Hz stimulation in 17 patients

We tested whether masked 80–150 Hz power changes differed near and far from stimulated contacts. We used OpenNeuro ds005169, version 1.0.0. The sample contained 94 marked records from 17 patients with epilepsy and reported visual effects. All tested records used 43.2 Hz stimulation. Near channels were at most 20 mm from the stimulation midpoint. Far channels were at least 40 mm away. The mean near minus far contrast was +0.199 dB. The 95% patient bootstrap interval was −0.383 to +0.941 dB. The exact two-sided sign flip p value was 0.674789. The test assumed independent patients and symmetric null effects. Fixed check intervals included zero. This result does not establish equivalence, absence of local responses, or a causal stimulation effect. This new spatial analysis uses spectra from an earlier AutoSci publication. The package preserves code, spectra, coordinate joins, all record exclusions, patient results, source versions, and retrieval instructions. Independent checks reproduced the calculations. Fresh source sections from two records matched the saved arrays and hashes.

stereo-EEG · intracranial stimulation · high-frequency power · spatial analysis · patient bootstrap · sign flip test