Shaft phase controls preserved bipolar beta power and gave 5.14 percent global rejection under independent uniform shaft phases in 19 SEEG records

Shaft phase controls kept measured local spectral products and bipolar beta power in phase models from 19 public SEEG records. We tested a graph maximum under independent uniform shaft phases. In 19,000 tests, shaft controls gave 5.14 percent global rejection at a 5 percent level. Independent contact controls gave 0.44 percent, and independent bipolar node controls gave 4.43 percent. The node control matched each edge distribution but changed dependence across edges. The proof and benchmark show why correct edge distributions do not ensure a correct graph maximum. A second minute gave the same direction of mismatch. All 4,222,800 Fourier coefficients were rebuilt exactly from ds003876 version 1.0.2. A full analysis repeat matched all 494 saved arrays and six result tables. The package contains code, fixed inputs, source records, results, figures, and repeat instructions. The evidence concerns a stated phase null, beta spectra, and fixed name groups. It does not test native neural coupling or clinical benefit.

SEEG · phase randomization · Fourier surrogate · bipolar signals · imaginary coherency · graph maximum · Monte Carlo calibration · electrophysiology