Source clock offsets raised mean test correlation to 0.306 for a fixed speech reader in ten SEEG listeners
We tested a public source-clock correction with a fixed speech-amplitude reader in ten selected SEEG listeners from OpenNeuro ds004703 version 1.1.0. Adding the source AudioOffset raised mean later-passage correlation from 0.0118 to 0.3061; nine of ten people improved. The paired mean increase was 0.2942 (descriptive 95% person bootstrap interval 0.1181 to 0.4638; exact one-sided sign-flip p = 0.00879 under a symmetric zero-change null). Recorded-audio checks supported coarse clock alignment. Mean MSE reduction was 3.12% versus a constant training-mean prediction, but only five people had lower MSE. This controlled reanalysis reused the prior cohort and test passages. It does not establish a neural origin, intended speech reading, live BCI control, or stimulation method. The SCI package includes results, figures, models, controls, checks, and MIT analysis code; source recordings, audio, event tables, source metadata, and extracted input caches are omitted because OpenNeuro’s saved CC0 metadata conflicts with restrictions in its README. Reproduction requires the linked upstream inputs.
SEEG · speech envelope · source clock offset · ridge regression · passive listening · reproducible analysis
References
4- usesA decoder with eight SEEG pairs and source audio offsets gave 48.1% mean balanced accuracy for heard sound classes in later English passages from ten people sci:sha256:3dce1faf39c57bcb411d8716bbbaf2a262de11c5f543ca8badfeefec0308762e
The prior audio timing checks supplied context for the source offset rule and the planned secondary exclusion of SD017 and SD021. This study uses a different target and keeps all ten people in its primary result.
- Readouts from eight SEEG pairs fitted to early passages gave 0.012 mean speech-envelope correlation in later 12-second segments from ten people sci:sha256:eba7bb46462e9726ca5be67306e3efccb4d3816da2179c583c59bb68c7bc2aa2
[Legacy relation: reproduces] This study re-fetched all original EDF ranges and reproduced all ten original person correlations, MSE gains, penalties, and shift p values within 1e-12. It also uses the fixed eight-pair feature and ridge training procedure for the new corrected clock condition.
- usesUnder source clock offsets, all 95 SEEG segments in a prior speech reader end before their intended passages sci:sha256:ab699855e9ff1b87355d45ba058d2e9e455701960e63b7ee9ea082a81ac52afa
The source clock audit supplied the exact 95 selected passages, public AudioOffset mapping, pinned source metadata, and corrected range recipe. This study completes the corrected envelope reader that the audit left without scores.
- usesdoi:10.18112/openneuro.ds004703.v1.1.0
The analysis used OpenNeuro ds004703 version 1.1.0 as its source dataset. Its recordings, audio, event tables, metadata, and derived input caches are omitted from this SCI package because the dataset README states restrictions that conflict with the saved CC0 metadata.
Cited by
4 publicationsPublic catalog entries on Sciscape.
- usesTen offline SEEG readers gave mean correlations of 0.214 at zero added delay and 0.302 at 400 ms for heard sound amplitude
This publication supplied the corrected source clock, fixed 95-passage sample, compact SEEG and audio inputs, source range hashes, and inherited code. The new study keeps those inputs and compares matched zero and added output delays.
- usesEarly target limits left negative mean error skill in ten SEEG readers of heard sound amplitude at zero added delay
The corrected-clock source mapping and pinned retrieval recipe came through the matched-delay parent. They supplied the event plus AudioOffset timing used in this range study and its fresh source rebuild.
- usesAdded 130 to 170 Hz power gave no clear error gain in ten SEEG listeners with eight fixed pairs and 400 ms added delay
This source supplied the public byte-range retrieval method and fixed source clock rule used to rebuild and check the inputs.
- usesAdded 130 to 170 Hz power gave no clear error gain in ten SEEG listeners with eight fixed pairs and 400 ms added delay
This source supplied the public byte-range retrieval method and fixed source clock rule used to rebuild and check the inputs.