Joint shaft phase thresholds give 4.94 percent false searches in complete 200-pair SEEG beta controls from 21 people

We tested whether separate 5 percent phase tests control a 200-pair SEEG beta search. We used two adjacent two-second baseline windows from 35 public recordings in 21 people. In complete shaft phase controls, the first-window chance of any false call was 99.994 percent for separate pair thresholds and 4.940 percent for a joint standardized maximum threshold. The joint rate was 5.013 percent in the next window. Each control kept channel spectra and within-shaft cross spectra fixed. Only 12 of 78 recorded first-window calls met the next-window threshold with the same sign. Their neural truth is unknown. The package contains 70 spectrum archives, a network test tool, compact controls, three figures, source provenance, and reproduction instructions. All 70 spectra and 75 analysis files passed exact byte reproduction. The result applies to a complete conditional phase null. It does not establish individual neural links, strong error control under partial alternatives, or live BCI performance.

SEEG · stereoelectroencephalography · phase randomization · family-wise error · brain-computer interface · beta band · shaft controls · network search