Global power adjustment reduced beta power correlation excess in 19 public SEEG records of 64 seconds

We tested how global power adjustment changes beta power correlation in 19 public SEEG records. The data came from OpenNeuro ds003876, version 1.0.2. Each record used 64 seconds from one patient with epilepsy. The adjustment used mean standardized power on shafts other than both tested shafts. Mean excess above common Fourier phase controls fell from 0.0598 to 0.0106. The paired reduction was 0.0492. Its 95% patient bootstrap interval was 0.0255 to 0.0746. The remaining excess interval included zero. The patient sign-flip p value was 0.0816. The conditional phase p value was 0.024. These tests have different assumptions. A mean that included tested shafts gave negative correlations in independent phase controls. The mean correlation was -0.1019 with inclusion and -0.0010 with exclusion. This known algebraic risk was present in this SEEG power measure. Fixed checks showed dependence on bin size and the power measure. The package preserves code, coefficient data, source pins, controls, result tables, a population proof, and validation checks. This is a method test in the same data as a prior result. It does not establish causal coupling or clinical benefit.

stereoelectroencephalography · SEEG · beta power · global signal regression · phase controls · correlation · epilepsy · signal processing