The measured mean rate of consecutive large SEEG forecast errors increased by 0.82 percentage points after pulses in eight people

We tested the order of large errors in a fixed SEEG waveform forecast. We compared windows before and after stimulation. We used OpenNeuro ds004696 version 1.0.1. The study had eight people, 317 conditions, and 1,933 later pulses. A forecast used the first five pulses. We marked the largest third of later errors, rounded up, in each series. The main windows were 20 to 500 ms after each pulse and 500 to 20 ms before it. The mean rate difference was 0.82 percentage points. The 95% person bootstrap interval was -0.21 to 1.98 points. The joint pulse order reference gave a two sided p value of 0.0893. Five of eight people had a positive difference. The interval includes zero. These data do not show a clear rise above the before pulse comparison. They also do not establish independent errors. All seven planned secondary differences had intervals that included zero. The package gives an exact order reference for fixed flag counts. It includes three figures, compact data, code, source hashes, and retrieval steps. A fresh rebuild from 64 source channels matched all five inherited data files. A separate numerical run matched all eight result files.

SEEG · intracranial EEG · electrical stimulation · waveform forecasting · forecast error · error order · pulse sequences · exchangeability · brain computer interface · bootstrap