Linear and categorical heat controls gave opposite mean history contrasts in 23 EEG adults; the shared test with 27 fMRI adults failed

We tested whether prior heat predicts a smaller current neural response than next heat in separate human studies. A local code lock preceded tests in 23 scalp EEG adults and 27 fMRI adults. The fixed model controlled current heat categories, run trends and MRI cue identity. Mean previous-minus-next coefficients were EEG +0.04439 [0.00415, 0.08464] and MRI -0.03155 [-0.12958, 0.06648]. Brackets give separate 95% person Student intervals. The previous coefficient also had to be negative in each study. The shared negative rule failed, p=0.983936. A fixed linear current-heat control gave an EEG contrast of -0.09450 [-0.13390, -0.05510]. Thus the two EEG controls gave opposite means. They were equivalent in MRI because that source has only two heat levels. This descriptive comparison does not confirm a positive neural history effect. Seven dose orders covered 29 EEG people; 17 shared one order. An added current-response model gave a negative synthetic EEG contrast under linear control without neural history. This does not establish the cause of the human result. We supply a new EEG table that keeps 3,775 original event links and 82 missing waveforms, all results, code, source pins, figures and replay steps. Original MRI scores remain external. Different cohorts, heat protocols, dose support, missingness, filters and fixed readouts limit scope. No shared neural mechanism or positive cross-modal history finding was confirmed.

human neuroscience · EEG · fMRI · pain · heat history · negative control · dose adjustment · reproducible data