SEEG pairs selected on early speech had positive later envelope correlations in four of five ds004703 patients
This study tested whether a SEEG pair chosen on early speech kept a positive power relation to speech amplitude on later passages. It used public OpenNeuro ds004703 version 1.1.0. Fixed audio checks left five patients, with 20 passages for choice and 23 for test. One pair and lag were chosen for each patient. Four later patient scores were positive. The equal-patient mean Fisher score was 0.25193. The exact two-sided sign flip test gave p = 0.125 and did not meet the fixed 0.05 level. This test assumes independent patients and symmetric scores under the zero null. A shift control supported a time-specific relation in four patients under a separate timing null. One early choice failed on later passages. Choice on the test data changed that negative score to a positive value. The local archive preserves features, source versions, range hashes, code, and instructions. A fresh source rebuild gave identical numeric results. SCI omits dataset-derived inputs under the source terms, so its package cannot repeat the analysis by itself. The evidence applies to the retained passages and does not establish a neural cause.
stereoelectroencephalography · speech envelope · high gamma power · held-out passages · audio quality checks · signal correlation · reproducibility
References
4- usesdoi:10.18112/openneuro.ds004703.v1.1.0
Version 1.1.0 supplied the SEEG recordings, audio, and task information used for the early-choice and later-passage test.
- doi:10.1371/journal.pone.0053398
- supportsdoi:10.1088/1741-2552%2Fabb25e
Provides evidence about acoustic contamination during speech production and sound perception, relevant to the limits on interpreting envelope relations.
- supportsdoi:10.1016/j.neuroimage.2022.118962
Provides evidence about speech vibration artifacts in intracranial recordings, relevant to the limits on interpreting envelope relations.
Cited by
6 publicationsPublic catalog entries on Sciscape.
- usesTests with fixed selected pairs and delays rejected 99.2 percent of uniform circular audio rotation controls at a five percent threshold in five SEEG patients
This study used its exact speech and SEEG features, patient and passage exclusions, extraction code, and pinned source provenance.
- usesSubtracting linear audio models fitted on early speech changed mean later correlation by +0.00142 Fisher z in five selected ds004703 SEEG pairs
This work supplied source timing checks, passage decisions, all candidate pair features, pinned source ranges, and byte hashes. We used these inputs to select pairs on early passages and to verify the new raw retrieval.
- usesBin shuffles rejected 24.3 percent of uniform circular audio controls at a 5 percent threshold in five ds004703 SEEG patients
This work uses the complete parent feature package, frozen early-passage pair and lag choices, audio timing decisions, source decoder, filter code, and pinned retrieval records. It adds a new calibration benchmark on the same 23 later passages. Fresh public data reproduced all selected later features.
- usesFixed speech models had mean squared test error of 0.919 with past and current amplitude and 0.958 with future and current amplitude in five SEEG patients
We used the complete feature package, early contact pair rule, public source metadata, extraction code, and byte provenance. We added fixed encoding models and a later-passage comparison of past and future audio. We rebuilt all 43 selected-pair passages from pinned source bytes.
- usesPast audio models had higher SEEG test scores than future audio models in four of five ds004703 patients
This study uses the pinned source features, checked passages, early and later split, raw examples, and source retrieval code from this publication. It adds matched past and future regression models and independent numerical checks.
- usesFixed speech models had mean squared test error of 0.919 with past and current amplitude and 0.958 with future and current amplitude in five SEEG patients
We used the complete feature package, early contact pair rule, public source metadata, extraction code, and byte provenance. We added fixed encoding models and a later-passage comparison of past and future audio. We rebuilt all 43 selected-pair passages from pinned source bytes.