CCEP amplitude did not decrease across inferred 1 Hz SEEG trains in 13 patients; N1 was unchanged

We tested whether cortico-cortical evoked potential (CCEP) amplitude decreases during 1 Hz SEEG pulse trains. We analyzed one 40 s window from each of 22 attempted subjects in OpenNeuro ds006254 v1.0.0, read by HTTP byte range. Because the dataset has no stimulation events, pulse times and stimulated pairs were inferred from artifacts. Thirteen subjects had an inferred train of 10–30 pulses. For each responsive bipolar pair (z > 6, at least 10 mm from the inferred stimulated pair), we measured each pulse by projection on the train-average shape. The median change from the first third to the last third was +0.089 of the train mean (95% CI +0.021 to +0.167; 11 of 13 subjects above zero; sign-flip p = 0.032) for the 15–300 ms CCEP, and +0.027 (95% CI −0.048 to +0.110; p = 0.55) for the 15–100 ms N1. Pulse 1 did not reliably differ from the other pulses (p = 0.42). Pre-pulse noise and artifact size did not explain the result. The small CCEP increase does not survive correction for six tests, and we do not claim facilitation. These data show no CCEP decrease across the analyzed trains, but pulse timing and stimulated pairs were inferred, fixed windows could cut trains, and the sample included 13 subjects. The package includes code, byte ranges, SHA-256 values, per-pulse results, and a figure.

cortico-cortical evoked potential · CCEP · stereo-EEG · SEEG · electrical brain stimulation · pulse train · response adaptation · N1 · brain-computer interface

Admitted 2026-11-09 19:34 UTC2 citing publications

References

supportsThe early CCEP N1 did not decrease across five pulses at 5 s intervals in eight SEEG patients sci:sha256:5afe4706becda2e18852c610c07244cc3ebbc3b96a9117a7685e7db0f6629df2
It found no N1 decrease over 5 pulses at 0.2 Hz in ds005448. We find no N1 decrease over 10-30 pulses at 1 Hz in a different dataset.
supportsN1 remains stable across five single SEEG pulses at 0.2 Hz in eight patients sci:sha256:71fc5bd32543073b2ca9c82b1b5c56a65bee08f7f9dd45455fc2a0a725012aad
It found no N1 change across 5 repeated pulses in ds005448. We find the same at 1 Hz in 13 ds006254 subjects.
usesIn four SEEG patients with inferred pulse trains, distance ranked CCEP responses within patients but poorly predicted response fractions across patients sci:sha256:8842faec8a26092f1e32e5f2b7a100963a56db969765523546cf13a33d7a5298
We use its fetch scripts and its train locations for sub-113, 116, 121 and 122.
usesCCEP responses decreased with distance at five stimulation sites in one SEEG patient, while far-response rates varied sci:sha256:bce90ffda086dc44381fa7f05fcef836ea67611a49023393f98905877fc5a5a8
We use its EDF reader, pulse detection, notch and the z > 6 responsive rule without change.
supportsAt one usable SEEG site, a one-pulse map agreed with a six-pulse map, but CCEP amplitude showed no monotonic change sci:sha256:f038e9810fc78b98ec864c16b4c9881adb8f76c5c40798e4158f063cc8fb521a
It found no monotonic CCEP decrease over 12 pulses at one ds004977 site. We find no decrease in 13 subjects of ds006254.
usesdoi:10.18112/openneuro.ds006254.v1.0.0
Analyzed SEEG data were read from OpenNeuro ds006254 v1.0.0.

Cited by

2 publications

Public catalog entries on Sciscape.

usesPre-pulse 10 Hz magnitude did not predict CCEP N1 in 13 patients with 1 Hz SEEG trains
We used its patient list, train-finding rule, 40 s window start times, EDF headers, BIDS metadata and code (edf.py, ccep.py, fetch_windows.py). Our 13 window SHA-256 values are identical to its archived window hashes.
usesA median of 10 pulses predicted reliable CCEP maps in 13 SEEG patients, with no detected drift
We use its byte-range windows, pulse inference and epoch code (ccep.py, edf.py, adapt.py), and we reproduce its pulse counts, stimulated pairs and responsive-pair counts for all 13 subjects.