Sites with lower repeatability distinguished fixed sound pairs in SEEG and fMRI; SEEG controls were uncertain

We tested whether less repeatable auditory sites distinguish sounds selected for small response differences at more repeatable sites. Separate public human studies supplied SEEG high-gamma amplitude and two BOLD fMRI groups. A 64-sound panel ranked sites. One repeat of 64 other sounds selected 32 disjoint pairs in the upper site half. Two other repeats tested assignment in the lower half. Development used two patients and four MRI adults. Local plan and code locks preceded tests in six other patients, 16 B2021 adults and eight NH2015 adults. Mean correct assignment and 95% person Student intervals were 58.33% [52.98,63.69], 69.34% [65.48,73.19] and 78.91% [73.73,84.08]. Chance was 50%. The one fixed shared test passed, p=0.015625. It required positive means and person Student and sign-flip evidence in every group. All 30 primary person means exceeded chance. SEEG intervals included chance with 24 input sites and after removal of trial-order trends. Its response-minus-baseline contrast was also uncertain. All negative results remain. A common scalar signal with noisy site gains can pass the method. Thus success does not show independent neural sources or unique information beyond the upper half. Source full-panel screens, sparse clinical sampling, spatial smoothing, separate cohorts and unequal measurement filters limit scope. The contribution is a checked sound-pair challenge and a restricted positive result in reused measurements. No common generator, sound-evoked specificity or causal lesion effect was established. Code, source versions and terms, aggregate results, figures and repeat steps are supplied. Original neural arrays remain external.

auditory cortex · SEEG · fMRI · repeatability · sound discrimination · distributed representation · data reuse · negative controls