After site calibration, sound matching without site labels failed a fixed shared test in six SEEG patients and 24 fMRI adults

We tested whether calibrated response magnitude distributions support repeated sound assignment without target site labels. Separate public human studies supplied SEEG high-gamma activity and fMRI BOLD estimates. Calibration used other sounds. Targets lost their spatial mean, vector length, global sign and site order. A reader assigned two sounds within one broad class to templates from another repeat. Development used two patients and four adults. Fixed tests used six other patients, 16 B2021 adults and eight NH2015 adults. Mean accuracies were 52.21%, 56.06% and 52.94%. Their 95% person t intervals were [49.50,54.92]%, [54.26,57.86]% and [50.59,55.30]%. The shared rule failed, p=0.090626. Both fMRI component rules passed. Joint person-and-sound resampling included chance for SEEG and NH2015. No positive cross-modal generalization was confirmed. All fixed controls and negative person scores remain. One concentration value gave positive pointwise MRI intervals. Noiseless one-direction signals gave ties; one direction plus independent private normal errors gave positive scores. A population moment calculation explains how signal-to-noise ratio can remain in normalized magnitude distributions. These examples were not fit to people. Source audits exactly rebuilt 784080 electrical window means and 9665865 selected MRI values. Separate cohorts, unequal sites, clinical sampling, shared schedules and source selection limit neural claims. The contribution is a new assay, a scoped negative shared result and a tested observation-model limit. The package gives code, source pins and terms, all statistics, figures, local plans and repeat instructions. Original neural arrays remain external.

auditory cortex · SEEG · fMRI · sound matching · magnitude distributions · normalization · private noise · negative result