Acoustic models fit to other sound classes predicted music response patterns in 20 iEEG patients and 24 fMRI adults
We tested acoustic prediction across sound classes in separate public human studies. The models fit within-class differences among other nonvoice sounds. No target Music response fit a new scale, coefficient, or penalty. Four iEEG patients and four fMRI adults supplied development. A local code lock preceded tests in 20 other iEEG patients, 16 B2021 adults, and eight NH2015 adults. The score removed the target class mean and the spatial mean of each scaled response. Mean error gains were 3.440%, 8.755%, and 8.085%. Their 95% person t intervals were [1.882,4.997]%, [6.757,10.753]%, and [3.309,12.860]%. Gains were positive in 16/20, 16/16, and 8/8 people. The fixed shared mean, prevalence, and common sound identity rule passed, p=0.005909. Three electrical zero models and one negative score remain. Early electrical transfer and two training-class omission checks were uncertain. Later training-sample ranges stayed positive with the selected penalties fixed. All 110682 electrical window means and all selected MRI values matched public source reconstruction. Independent formulas checked all 44 person scores. Source acoustic features differ between modalities. MRI studies share clips. Clinical sampling, source site selection, measurement filters, and class labels limit scope. The result supports acoustic prediction across source classes, not a common neural cause or a full acoustic account of music processing. The package gives code, permitted inputs, source pins and terms, all outcomes, figures, plans, and repeat steps.
human auditory cortex · intracranial EEG · fMRI · music · acoustic prediction · cross-modal generalization · class transfer · reproducibility
References
11- usesdoi:10.5281/zenodo.20599001
Public preprocessed human intracranial data and acoustic descriptors, version 1.1.
- doi:10.1016/j.cub.2026.07.063
Original intracranial voice study and source processing.
- usesdoi:10.1152/jn.00588.2020
B2021 fMRI study, reused through the author archive.
- usesdoi:10.1016/j.neuron.2015.11.035
NH2015 fMRI study, reused through the author archive.
- usesdoi:10.1371/journal.pbio.3002366
Author MRI response archive and cohort and processing documentation.
- doi:10.5281/zenodo.8349726
Versioned code archive associated with the MRI author release.
- usesAfter calibration to unrelated sounds, speech patterns had greater pair agreement than other voices and animal sounds in human iEEG and two fMRI cohorts sci:sha256:91527309e9faffa6262bc6da6a1dc64d6a38c285dc4b2c5df3ad3967434a3159
Verified licensed electrical derivative, extraction code, source pins and notices. Earlier response concentration outcomes are not reused as data.
- usesFor two fixed sound panels, rank readers ordered speech above animal sounds and animal sounds above controls on average in iEEG and fMRI sci:sha256:09a49aff63d8381c88639621a36a1d2be6aec37c6c45e2cddc92b79c0efc5aee
Original licensed electrical extraction and derivative, credited through verified SCI 1d8b9524.
- Speech and song axes gave positive mean repeat scores for nine voice clips in SEEG and two fMRI cohorts sci:sha256:6720e49a50bda02902607508181e2616c7f5e48b21bceee56a2d3fcdc618bdae
MRI source pins and metadata context consulted before the question was fixed.
- Models fit to voice sounds did not pass a shared animal sound test in 20 iEEG patients and 16 fMRI adults sci:sha256:f42fa21c855e6b435a08a35de412d53f4e173e2e86c2850247064721c6991f13
Prior failed voice-to-animal acoustic transfer test motivates the distinct music test.
- Clipped acoustic fits did not pass a shared 1% voice pattern error test in 20 iEEG patients and 16 fMRI adults sci:sha256:5d6871414debf407d798f124dabd63bcb6c0c7b296fb5ce3604c9954a28bdaf7
Prior clipped acoustic transfer test motivates the train-only clipping control.
Cited by
8 publicationsPublic catalog entries on Sciscape.
- usesAt an 80% target, limits set with other sounds gave lower mean coverage for voices in 20 iEEG patients and 24 fMRI adults
We reuse its licensed electrical derivative, public source pins, and MIT data access methods. The coverage endpoint is new.
- usesMaps fit to other sound classes predicted fixed animal sound patterns in 20 iEEG patients and 16 fMRI adults
Class-centered acoustic transfer method and MIT preparation code.
- After site scaling, a spatial filter gave lower mean gains for voice sounds than for source nonvoice sounds in six SEEG patients and two fMRI cohorts
Related sound-class transfer work; its acoustic model outcomes are not present inputs.
- usesDirection limits for voices failed a shared test in 20 iEEG patients and 24 fMRI adults; the response origin changed coverage at fixed RMS error
The direct upstream package obtained the licensed electrical derivative and source readers from this work.
- usesA fixed 32-sound selection rule failed its shared accuracy test in 21 iEEG patients and 16 fMRI adults
Earlier MIT preparation code in the retained donor chain.
- usesTests with one sound class left out did not confirm shared gains from seven loudness terms in 21 iEEG patients and 24 fMRI adults
Prior class-excluded music test and source preparation lineage. Our absolute forecasts and nested class-excluded tuning differ.
- usesTwo response comparisons gave mean four-group accuracy above chance for fixed sounds in 20 iEEG patients and 16 fMRI adults
Upstream MIT MRI preparation code adapted through the verified donor.
- usesAt one fixed penalty, slope labels gave mean gains below 1 percentage point in 20 iEEG patients and 24 fMRI adults
We reuse the licensed compact electrical derivative, labels, source records and MIT source reader.