Fixed tests did not confirm a shared video response repeatability change in six EEG and ten fMRI adults

We tested whether response-pattern repeatability changes across repeated video sessions in two separate human studies. Six EEG adults and ten fMRI adults viewed the same 102 three-second clips. EEG used eight posterior channels and 0.3 to less than 1.0 s. MRI used selected visual regions. One fixed EEG calibration used other videos. A local plan and code lock preceded the new endpoint. Mean late-minus-early correlations were +0.00810 and +0.00185. Their 95% person Student intervals were [-0.00690, +0.02309] and [-0.01167, +0.01538]. The shared increase rule failed after Holm correction, p=0.820312. The shared decrease rule failed with p=0.890625. All people, fixed panels and low-movement checks remain. Separate cohorts, small samples, EEG audio, silent MRI videos, unequal measurement filters and session-linked repetition limit scope. This is a test of processed response stability. It does not identify a shared neural cause or a causal repetition effect. The package gives code, permitted small inputs, exact source versions, source checks, all results, figures and repeat steps.

EEG · fMRI · natural videos · repeatability · repetition · session effects · human neuroscience