Under source order, fixed neural heat scores had current report correlations above the neighbor mean in 100 EEG and 20 fMRI adults

We tested whether neural heat scores have a local peak in pain-report alignment after input and timing controls. Separate public human studies supplied scalp EEG features and fitted BOLD fMRI responses. Heat labels set the neural reader weights. Development used 40 EEG and eight MRI fitting people. A local code lock preceded new lag scores in reserved people. Fixed trial rules retained 100 EEG adults and 20 MRI adults. The current partial correlations were 0.21676 and 0.12184. Current minus mean adjacent-report correlations were 0.19494 and 0.12570. Their pointwise 95% person Student intervals were [0.14573,0.24414] and [0.01686,0.23454]. One fixed shared rule required positive current correlation and positive contrast in both studies, with person Student and sign-flip tests. It passed, p=0.01318645. Seventeen EEG and six MRI people had negative contrasts. The MRI contrast range also crossed zero when training people and test people were resampled. The separate MRI previous-report comparison and the 24 mm spatial check were uncertain. All negative results remain. The second EEG visit repeats the same cohort. Exact generic shared-state models also gave a positive peak without a pain-specific process. The contribution is an event-linked comparison tool and a restricted positive cross-modal result. Source order, source selection, few residual dimensions, full-record fits, separate cohorts and measurement filters limit scope. No common neural cause or clinical validity was established. Code, permitted inputs, source versions and terms, all scores, uncertainty, figures and repeat steps are supplied.

pain · EEG · fMRI · human · trial alignment · cross-modal test · temporal dependence