Neural scores had positive heat relations after linear report control in 115 EEG adults and 20 fMRI adults

We tested whether fixed neural scores retained heat-level relations after control for current pain reports. Public EEG and BOLD fMRI studies supplied separate human cohorts. Scores had been trained on heat labels in 39 EEG and eight MRI people. A local code and input lock preceded the new test in 115 other EEG people and 20 other MRI people. Mean partial correlations were 0.2263 and 0.1600. Their 95% person Student intervals were [0.1861,0.2666] and [0.0886,0.2314]. The shared positive test passed, p=0.0001024. Cubic report controls, ranks and fixed reader checks kept positive means. A later common-report-range restriction retained 111 EEG and ten MRI people. Its MRI mean was 0.1010, with interval [-0.0958,0.2978]. Thus a shared effect at matched reports was not established. A single latent process with noisy reports and no direct heat-to-score term matched each group average correlation matrix. The required report reliability values were 0.458 and 0.799. These are compatibility values, not measured reliabilities. Positive-correlation constructions met their conditions in 74/115 EEG and 12/20 MRI people; all failures remain. Under stated independent-error assumptions, varying report reliability changed the sign of a direct heat coefficient while keeping the same observed moments. Artificial-data checks and independent arithmetic passed. The finding supports a measured regression relation in both modalities, not distinct neural processes or a causal path. Source selection, report overlap, different cohorts and readouts, source processing and fixed readers limit scope. The package gives code, permitted inputs, all results, source versions and terms, figures and repeat steps.

pain · EEG · fMRI · measurement error · partial correlation · report reliability · cross-modal generalization · public data