Three-sound order tests failed in six SEEG patients and 24 fMRI adults under fixed repeat rules

We tested whether human auditory response means can share one scalar order. Separate public studies supplied SEEG and BOLD fMRI. One repeat selected three-sound conflicts. Two others tested the smallest of four signed comparisons. Development used two patients and four MRI adults. A local code lock preceded tests in six other patients, 16 B2021 adults and eight adults from a separate MRI study. Individual-sound mean scores were -0.8663, -0.3401 and -0.4980 selection SD units. Their 95% person Student intervals were [-1.0162,-0.7164], [-0.4253,-0.2550] and [-0.6067,-0.3893]. All 30 scores were negative. A development-led class-mean test also failed. Both shared claims had Holm p=1. All sensitivity and sound-refit results remain. We add a sufficient central-symmetry condition for a whole-person sign reference. An exact scalar null with unbiased asymmetric errors gave positive scores with probability 0.748. This is a sign-reference limit, not a human error estimate or a false-positive rate for the complete rule. Another exact example shows that class averaging can create order conflicts from one monotone scalar. Source screens used all repeats, so the conditional error assumptions are not established in the human data. Clinical sampling, different cohorts, finite sounds, reference and measurement filters limit scope. Code, source pins and terms, all results, figures and repeat steps are supplied. Original neural arrays remain external. No positive cross-modal neural result or causal effect was confirmed.

human neuroscience · SEEG · fMRI · auditory responses · monotone models · order conflicts · negative results · sign test · selection bias · reproducibility