SEEG from five patients showed a mean 41 percent late absolute peak disagreement above 20 ms for separate smoothed five pulse means with Pearson r≥0.8

We tested whether waveform shape agreement gives late peak time agreement. The source was public SEEG from five patients in OpenNeuro ds004457 version 1.0.2. Each comparison used separate groups of five pulses. We tested all 126 divisions at 1,081 fixed stimulation site and receiving contact links. An 11-bin box mean preceded absolute peak selection between 50 and 250 ms. Pearson correlation used the same samples. Among comparisons with correlation at least 0.8, the mean patient fraction with a peak gap above 20 ms was 41.0 percent. The descriptive 95 percent patient bootstrap interval was 35.1 to 47.8 percent. The fraction was 25.3 percent when both peak signs agreed, and 33.7 percent at correlation at least 0.9. A proof shows that high correlation alone cannot bound the peak gap below the full time grid span. The result concerns this sample and estimator. It does not establish a biological or clinical error rate. All planned checks are reported. Fresh source decoding reproduced all 3,477,760 saved waveform values. Alternate calculations gave the same primary threshold decisions. A full repeat matched all 17 numerical output files. The package contains code, pulse windows, original metadata, source hashes, retrieval recipes, comparison scores, results, figures, a proof, checks, and instructions. The local plan had no external registration.

stereoelectroencephalography · evoked potentials · peak timing · waveform correlation · reproducibility