Median paired 30 to 80 Hz exponent changes were positive from one to four contact steps in 19 SEEG patients

We tested whether wider bipolar contact spans change a fixed spectrum exponent. We used 19 public interictal SEEG records from OpenNeuro ds003876 version 1.0.2. Each patient had a positive median paired change from one contact step to four contact steps. The fit used 30 to 80 Hz and removed 58 to 62 Hz. The mean patient change was 0.276 units. The 95 percent patient bootstrap interval was 0.211 to 0.349 units. This interval assumes independent patient sampling and applies to the fixed records and contact paths. All patient medians were positive in both one-minute halves. An exact spectrum allocation assigned 0.292 of the 0.308 mean anchor change to a cancellation factor. That allocation depends on the recorded reference and does not identify a causal mechanism. The 30 to 100 Hz check had 17 positive patient values. This study supports the direction of an earlier two-patient result in a separate dataset. Source range hashes, independent calculations, and a complete replay passed. The package preserves code, provenance, spectra, results, a raw example, and repeat instructions.

SEEG · intracranial EEG · bipolar referencing · contact span · spectral exponent · power spectrum · epilepsy