Response and repeat difference covariance maps had positive mean agreement for fixed sounds in six SEEG patients and 24 fMRI adults

We tested whether sound response covariance agrees with repeat difference covariance on other sounds. Separate public studies supplied repeated human SEEG high-gamma activity and fMRI BOLD responses. Disjoint sound sets supplied site scales, cross-repeat response products, and repeat differences. We removed broad class means and additive site pair terms. Development used two patients and four adults. A local code lock preceded tests in six other patients, 16 B2021 adults, and eight NH2015 adults. Mean map cosines were 0.18129, 0.25313, and 0.28639. Pointwise 95% person t intervals were 0.12937–0.23320, 0.21166–0.29459, and 0.23068–0.34210. All 30 people were positive. The shared mean, prevalence, and site-reference rule passed, p = 0.015625. Repeat energy alignment and 16-site checks retained positive means. The short SEEG baseline was uncertain. Two MRI class-omission scores were negative. A fixed-scale algebra identity makes the score invariant to a common additive spatial reference. An exact shared-mixing model gives perfect population agreement with independent latent sources and errors. Thus, the result is a shared measured covariance relation, not proof of tuning-dependent neural noise or a common cause. The package preserves code, source versions and terms, local plans, all results, figures, and repeat instructions. Independent arithmetic and exact source-to-cache checks passed. Clinical sampling, source selection, separate cohorts, unequal sensing volumes, and measurement filters limit scope.

human neuroscience · SEEG · fMRI · auditory cortex · repeat variability · covariance · measurement mixing