Independent phase controls kept positive median bipolar gamma power correlations between shafts in two SEEG rest records

We tested whether independent contact phases remove correlations between bipolar gamma power signals. The source was OpenNeuro ds005670 version 1.0.0. Two rest records supplied 24 contacts and 20 bipolar channels each. We used the first 240 seconds, two-second windows, and 70 to less than 110 Hz. Each model had 1,000 draws. With fixed contact magnitudes in their original windows, median correlations between shafts were 0.0669 and 0.1694. An added independent contact window permutation gave medians of -0.0014 and -0.0047. Recorded medians were 0.0083 and 0.0420. An exact finite-window proof shows why the phase model keeps the covariance of contact energy sums. The order control has zero expected covariance for channels with no common contact. A common contact can retain positive covariance. Complete phase grids and all planned checks supported the calculation. A second process reproduced all numerical arrays exactly. Fresh source checks covered all 23,040,000 archived digital values. The package keeps data, code, proof, figures, source pins, and replay instructions. The records reuse a prior sample. These conditional model results do not establish neural coupling, population effects, or better BCI performance.

stereoelectroencephalography · SEEG · gamma power · phase randomization · bipolar channels · power correlation · finite-window covariance · brain-computer interface