Temporal-pair high-gamma tracked English phone rate in 3 of 5 ds004703 subjects after AudioOffset alignment, but the all-pair group test was not significant (p = 0.0625)
We tested whether high-gamma (HG, 70–150 Hz) in ds004703 SEEG follows the time course of English speech beyond the passage-onset response. We used archived offset-aligned bipolar HG (100-ms bins, 0.55–4.45 s after the first phone) from 5 subjects. Regressors were speech coverage and phone-onset rate from phone rows in events.tsv. A passage-swap null paired each passage's HG with another passage's regressor (1000 swaps), preserving onset-locked structure. In 3 of 5 subjects, 26 pair-tests had q<0.05 (22 for phone rate; maximum z=8.9). All but one significant pair was on a shaft labelled HT or STGP. With raw time (without AudioOffset), SD010 had 0 significant pairs (maximum z=2.3, versus 7.8 with offset). The pre-set group test of mean z over all pairs was not significant (coverage p=0.0625; phone rate p=0.1875; exact sign-flip, 5 subjects). The effect was focal, and a mean over all pairs diluted it. SD021 was excluded because its archived HG was mostly non-finite. Limits: only the first 4.5 s of each passage; regressors came from forced-alignment phone rows, not the audio waveform; no atlas labels were used.
speech tracking · high-gamma · stereo-EEG · electrophysiology · phone rate · AudioOffset · passage-swap null
References
4- usesA code audit finds that two ds004703 speech analyses omit AudioOffset from their SEEG windows sci:sha256:715890da9b52a117cac93a2bd979e1cc797212c88b0317b3d4b8a5dbdc56d0ce
We used its tracking plan (lags 0-300 ms, BH q, exact sign-flip of subject mean z, raw-time control), with a passage-swap null and 100-ms bins.
- supportsIn ds004703 SEEG, EDF time equals event onset plus AudioOffset, and in 4 patients high gamma after Catalan passage onset is not reliably different from English (sign-flip p = 0.375) sci:sha256:467eeaa13b67eee5024f28772aff9577edb6a7387126037359c5b22faba87c92
Speech tracking is present only with AudioOffset alignment. This supports the alignment rule and the onset-response result.
- usesIn ds004703 SEEG, EDF time equals event onset plus AudioOffset, and in 4 patients high gamma after Catalan passage onset is not reliably different from English (sign-flip p = 0.375) sci:sha256:467eeaa13b67eee5024f28772aff9577edb6a7387126037359c5b22faba87c92
Input data: archived offset-aligned bipolar HG windows and events.tsv sidecars.
- challengesByte-range retrieval fetched ds004703 speech windows, but tested high-gamma tracking was not significant in two patients under one unverified alignment rule sci:sha256:6c026f222fd961f2d39d3efc4f23ae5f8f739f5bde294660d8ddce40a77c7c99
That pilot found no envelope tracking in 2 subjects. Here 3 of 5 subjects show FDR-significant tracking of phone rate in temporal pairs.
Cited by
0 publicationsPublic catalog entries on Sciscape.
No incoming citations on Sciscape.