A fixed test did not confirm shared sound distinctiveness in six SEEG patients and 16 fMRI adults

We tested whether reference listeners predict which sounds have patterns far from other sounds in the same group. We also tested whether this agreement exceeds agreement on repeat-to-repeat differences. Separate human studies supplied auditory SEEG high-gamma amplitude and BOLD fMRI estimates for 165 sounds. Two patients and four MRI adults supplied reference profiles. A local code lock preceded new scores in six other patients and sixteen MRI adults. Mean repeated-pattern agreement was -0.0530 and 0.5472. Their 95% person Student limits were [-0.1650, 0.0590] and [0.4941, 0.6002]. Mean advantages over disagreement agreement were -0.0109 and 0.5111. The fixed shared gate failed, p=0.86103. A fixed MRI reference also gave positive scores in all eight adults from a second MRI study, which has descriptive status under the main-first gate. Planned SEEG checks gave positive mean agreement before a new short-baseline subtraction and uncertain agreement after it. All negative people and controls remain. An exact identity links the assay to average cross-repeat pair separation. Independent literal calculations, generated models and offline score replay check the method. The result does not confirm cross-modal generalization or establish a biological modality difference. Fixed libraries, source selection, shared order, sparse clinical sampling, baseline uncertainty, unequal spatial support and vascular effects limit scope. The package supplies code, source versions and terms, result vectors, all outcomes, figures and repeat instructions. Original site arrays remain external.

auditory cortex · SEEG · fMRI · natural sounds · crossvalidated distance · repeat disagreement · cross-modal test · baseline subtraction