Exploratory tests found site pair map agreement across fixed sound panels in pre-onset iEEG from 17 patients and fMRI estimate differences from 16 adults

We tested whether site pair tuning relations persist between non-vocal sounds and human voices. Separate public studies supplied intracranial EEG and BOLD fMRI. Four people per study supplied development. A fixed test used 17 other patients and 16 other adults. Sound class means, atlas-pair means and additive site effects were removed. Mean residual map agreement was 0.84708 in iEEG and 0.56787 in fMRI. Their 95% person t intervals were 0.79015 to 0.90401 and 0.52240 to 0.61335. Every person had a positive score. The shared sign and within-atlas site-shuffle rule passed, conjunction p=0.0001. Sound and person resamples kept positive means. A label-width error first used only eight electrical speech sounds. We kept that run and restored the planned 71 voice sounds. The corrected execution was no longer fully blind. Removal of one fitted pattern kept high human scores. It also gave mean agreement 0.463 in independent-site models with electrical site counts. One shared scalar gave perfect population agreement after the primary pair controls. Exploratory pre-onset iEEG had mean agreement 0.86657. fMRI estimate differences gave 0.53791. Both primary-minus-background person intervals included zero. The result concerns measured tuning relations for two fixed panels. It does not establish the same map across modalities, several neural sources, direct coupling or a cause. The package preserves code, permitted inputs, source versions, all outcomes and repeat steps.

human neuroscience · intracranial EEG · fMRI · auditory cortex · tuning correlations · measurement model · sound classes