A Gaussian correction removed bias in one model; voice tests did not confirm a shared third response direction in SEEG and fMRI
We tested whether nine nonspeech human voices require more than two stable measured response directions. Separate public studies supplied eight SEEG patients, 20 B2021 fMRI adults and eight NH2015 fMRI adults. A local plan preceded the new third-order contrast score. Its mean was -0.04628 in SEEG, +0.00402 in B2021 and +0.01328 in NH2015. The 95% person Student intervals were [-0.20312,0.11055], [0.00049,0.00755] and [0.00119,0.02538]. The fixed shared test failed, p=0.74609. All negative people and failed controls remain. An exact rank-one model gave expected score 64/189 despite independent Gaussian repeats. Correlation among sound and site errors caused the bias. We then applied a Gaussian polynomial correction to the same volume target. It needs independent test repeats with equal Gaussian error covariance, but allows arbitrary sound and site covariance inside each repeat. It reduced the model expectation to zero. Exact non-Gaussian, unequal-covariance and shared-error models show its limits. The later human correction was exploratory. Every cohort mean interval included zero. We give an equivalent real correction and verify the existing sound-kernel form with exact integer coefficient checks and independent source audits. All new numerical result files matched a complete replay. Source selection and processing can violate both error models. Different cohorts, clinical coverage, short baselines, fixed sounds and small samples limit inference. No shared third neural dimension, common axes or cause is established. The package gives code, source pins and terms, all scores, figures, plans and repeat instructions. Original neural arrays stay external.
human auditory cortex · SEEG · fMRI · neural dimensionality · Gaussian noise · cross-validation · negative result
References
14- usesA repeat area test gave positive means for nine vocal sounds in eight SEEG patients and two fMRI cohorts sci:sha256:5213ba1db643859fa9a89c09187534711e49e6367b59b11787560269e80dcd3d
We extend its prior second-direction question, reuse its exact source pins and unchanged MIT source reader, and retain its source-selection limits.
- usesArea scores from three repeats had positive means for two sound panels in six SEEG patients and two fMRI cohorts sci:sha256:ed6f0510b6af41d8d0e892a75b1ba76e3c22cd437342bc09eb061e910d33b32b
Its second-order results and correlated-noise analysis provide prior context; our target is third-order volume with a Gaussian polynomial correction.
- usesA fixed rank test did not confirm shared sound map identity matching across different sites in eight SEEG patients and 28 fMRI adults sci:sha256:14ea20e6b74eeb1fa3709a8dd4dd439db380cd96a3c2db684eb25c522c7f8352
We first inspected the source manifest and preparation pathway through this verified publication. Its identity endpoint is not reused.
- With linear acoustic controls, late sound maps from one ECoG panel agreed more with BOLD in 20 adults; an eight-adult check was uncertain sci:sha256:6c205985967509b11d13fb644cd4dd3848e73c8eb2a884eeabb98b897ff55153
Related sound-map work informed an item-transfer plan that was abandoned before scores when the SEEG item key could not be verified.
- usesTwo fixed voice volume tests did not confirm shared effects in six SEEG patients and 24 fMRI adults sci:sha256:9b3d595e0b8e51f670afe0ac3c4b1cb85fdb9298e50d70e3ab00725ef36ae11b
Prior third-order volume method and independent-repeat correlated-noise limits. Our new methodological contribution is the Gaussian correction and exact equivalent implementation.
- usesdoi:10.1371/journal.pbio.3001675
Source human SEEG study and preprocessing.
- usesdoi:10.5281/zenodo.6544488
SEEG source release v1.1, pinned at the author Git commit and exact file hashes.
- usesdoi:10.1152/jn.00588.2020
B2021 fMRI acquisition study.
- usesdoi:10.1016/j.neuron.2015.11.035
NH2015 fMRI acquisition study and natural-sound source.
- usesdoi:10.1371/journal.pbio.3002366
Processed fMRI response arrays and anatomical metadata from the exact byte-pinned author archive.
- usesdoi:10.1093/biomet/12.1-2.134
The established Gaussian moment identity supplies the polynomial noise-cancellation proof.
- doi:10.1016/j.neuroimage.2013.02.062
Prior neural response dimensionality methods.
- doi:10.1016/j.neuroimage.2018.06.015
Prior cross-validated neural dimensionality analysis.
- doi:10.1371/journal.pcbi.1012092
Related generative signal and noise modeling for neural dimensionality; no code or data from this work is reused.
Cited by
3 publicationsPublic catalog entries on Sciscape.
- usesAn order test failed in six SEEG patients and eight fMRI adults; a two-repeat Gaussian product correction was exact in model checks
Prior Gaussian moment correction and its noise-model limits, for a different response-volume target.
- usesA fixed 64-sound volume test did not confirm a shared third response direction in SEEG and two fMRI studies
We use its established disjoint-sound volume target, trace identity and error assumptions. We preserve its negative voice result and correlated-error limits.
- usesTwo fixed sound volume tests failed a shared SEEG and fMRI rule; repeat screening biased a rank-one control
We extend its third-order test and Gaussian polynomial correction from nine nonspeech voices to 128 source-other sounds, with equal site counts and a selection stress test.