Two fixed tests did not confirm a shared rise in hippocampal change energy at film boundaries in 11 iEEG patients and ten fMRI adults

We tested hippocampal change energy at film boundaries after removal of the spatial mean. Separate public studies supplied bipolar iEEG power and BOLD fMRI. Development used four patients and five adults. A local lock preceded tests in 11 other patients and ten other adults. Against time references, mean scores were +0.02306 and +0.01955. Their 95% person Student intervals were [-0.00639,+0.05252] and [+0.00792,+0.03117]. The shared Holm p value was 0.111645. Against selected shot transitions, means were -0.01762 and +0.02304. The shared p value was 0.725586. Both fixed rules failed. Electrical matching left a large following-shot gap difference. A fixed short-window MRI check gave a mean time-reference score of +0.47124. A design audit gave +0.55454 under constant independent error because event and grid windows had different sample counts. A separate cut-linked variance model gave positive timing contrasts in both film schedules with zero conditional site means. These are interpretation limits, not fitted human causes. The package gives new site-wise power inputs, code, source pins and terms, all results, figures and repeat steps. Sparse clinical sampling, separate cohorts and films, unequal sensing, source processing, poor matching and missing nuisance traces limit scope. No positive cross-modal neural claim is made.

hippocampus · event boundaries · intracranial EEG · fMRI · measurement noise · pattern distance · negative result · reproducible data