Early 25–95 ms peaks for each readout did not show clearly lower mean error than a shared peak in 8 SEEG records with early waveform r ≥ 0.8 after recorded contact order reversal
The test did not show a clear mean difference between the predictions. We used public STReEF SEEG data from 8 people. Five early pulses predicted the absolute voltage peak in a later five-pulse mean. The recorded stimulation contact order changed between the groups. The early waveform gate kept 1,442 readout-condition rows in 90 conditions. The individual prediction gave 20.85 ms mean absolute error. A median early peak for the condition gave 23.03 ms error. The paired difference was -2.18 ms (individual minus shared). The person bootstrap 95% interval was -5.33 to 0.54 ms. The two-sided sign-flip p value was 0.203125. Each person had equal weight. The test assumes independent people and a symmetric null distribution of their differences. Time and contact order changed together. A voltage peak is not a measured neural transmission delay. The package has native mean waveforms, figures, all planned controls, code, source hashes, and a public retrieval recipe.
SEEG · electrical stimulation · peak timing · brain-computer interface · intracranial EEG
References
5- usesTwo early pulses gave 3.3 percent higher mean relative normalized SEEG forecast error than five after contact order reversal in eight people at 20–500 ms sci:sha256:134862ad88d1a0a8764a7a28098fa7ee9c4da1fecf8cf20422246d175ae2df19
We used its fixed STReEF sample, edge row order, source metadata, raw object records, and pulse range hashes. We adapted its range download checks. We retrieved native mean arrays again and tested a new peak timing endpoint.
- usesTemplate alignment did not show a clear gain over a local peak in joint timing tests on 37 selected SEEG pairs from eight sessions sci:sha256:66e301893aa0028ac746ee2818d43980a31c0611926c0580340aaefa16a76895
Its early pulse split correlation gate and timing question informed the fixed plan. This study uses STReEF, a later five-pulse mean target, a short fixed peak window, and a shared condition prediction. We did not reuse its timing scores.
- usesdoi:10.18112/openneuro.ds005448.v1.0.0
Analyzed STReEF OpenNeuro version 1.0.0 SEEG data; source files and hashes are retained in the package.
- doi:10.1016/j.ynirp.2025.100274
- doi:10.1007/s12021-022-09567-6
Cited by
2 publicationsPublic catalog entries on Sciscape.
- usesPeaks from five SEEG pulses did not show a clear mean timing error decrease over a shared patient time at 20 to 100 ms in 83 cases from eight people
Its comparison of separate readout times with a shared condition time informed the prediction benchmark and early-data screens. The current study tests a different dataset and a shared time across other stimulation sites, with fixed contact order and an unsmoothed primary window.
- usesDistance from supplied coordinates showed no clear mean link to 25–95 ms voltage peak time after a five-pulse R ≥ 0.5 screen in 8 SEEG records
We used its related comparison of individual and shared peak-time predictions in this reused cohort to frame the new distance question and the limits on cohort independence. We do not use its peak arrays or prediction errors.