Labelled bipolar SEEG pairs with a shared contact have higher 1–40 Hz correlations before stimulation in 20 patients

We tested whether contact sharing predicts correlation in public SEEG data from OpenNeuro ds006519 version 1.0.0. The package contains epochs from 35 recordings in 21 patients. The primary comparison used 33 recordings in 20 patients. It compared pairs that share a labelled contact with nearby disjoint pairs on the same shaft. Before stimulation, the median patient contrast in signed 1–40 Hz correlation was 0.139. Its 95% patient bootstrap interval was 0.074 to 0.277. The exact two-sided sign test gave p = 0.000402. In an exploratory whole-network audit, shared pairs had a median selection-rate ratio of 6.05 among the strongest 10% of absolute correlations. A proof and simulation show that shared contacts can produce correlation and a change between time windows under independent-source assumptions. The observed association does not separate contact subtraction from neural or spatial effects. The package includes source hashes, all annotated epoch derivatives, edge flags, patient results, tests, figures, and reproduction code.

stereo-electroencephalography · bipolar channels · shared contacts · Pearson correlation · connectivity · negative motor stimulation · reproducibility