Adding source clock offsets increased later passage speech amplitude correlation from 0.012 to 0.306 in ten eight-pair SEEG readers
We tested a fixed reader of heard speech amplitude in ten people from public OpenNeuro SEEG data. We used the same 95 passages and eight-pair fitting rules under two clock conditions. Adding the documented source AudioOffset increased mean later-passage correlation from 0.0118 to 0.3061. The mean paired increase was 0.2942, with a descriptive 95% person bootstrap interval of 0.1181 to 0.4638. Under sign exchangeability, the exact two-sided sign-flip value was p = 0.01758. Nine of ten people improved. Independent room audio supported the corrected clock in every person. Only five corrected readers beat a constant prediction of the training-target mean; mean error gain was 3.12%. These selected clinical recordings test heard amplitude. They do not establish a neural signal source, intended speech reading, live control, or brain stimulation. SCI retains the figures, result tables, predictions, fitted models, analysis code, and numerical checks. Conflicting source terms require exclusion of the source-derived inputs, which limits public reproduction.
stereoelectroencephalography · speech amplitude · clock alignment · brain-computer interface · signal tracking
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
Its source clock investigation supplied prior evidence for the additive AudioOffset convention. We used independent room-audio checks on all 95 passages in the present amplitude study.
- 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] We reproduced its combined eight-pair original-clock person scores within 2.50e-16. We retained its passages, feature rules, pair selection, and ridge fitting procedure for the corrected comparison.
- usesUnder source clock offsets, all 95 SEEG segments in a prior speech reader end before their intended passages sci:sha256:ab699855e9ff1b87355d45ba058d2e9e455701960e63b7ee9ea082a81ac52afa
We completed its open corrected clock comparison. We used its pinned source recipe, inherited arrays, documented offsets, and original reader files.
- usesdoi:10.18112/openneuro.ds004703.v1.1.0
Source SEEG recordings and task metadata were used for the ten-person speech-amplitude clock comparison. The dataset README also states noncommercial and no participant-identification restrictions; those source inputs and caches are excluded from SCI.
Cited by
1 publicationPublic catalog entries on Sciscape.
- usesA 400 ms delay increased mean heard speech log amplitude correlation in ten SEEG readers with equal input size
This study uses the corrected source clock, all ten people and 95 selected passages, feature arrays, sound targets, native audits, source provenance, and inherited reader functions. It tests a new matched delay comparison on 113 common target rows.