After camera cut controls, two boundary models did not confirm shared hippocampal prediction in 12 iEEG patients and ten fMRI adults

We tested whether film boundary labels improve hippocampal prediction beyond camera cuts, shot history and drift. Separate public studies supplied bipolar intracranial field power and BOLD fMRI. Four patients and five adults set the ridge value for two response models. A local plan and a model width amendment preceded all new endpoint scores. Fixed tests used twelve other patients and ten other adults. Mean gains for full time courses were -0.002375% and +0.005023% of residual energy. Five coefficient gains were -0.000448% and +0.004328%. All person t intervals and crossed person and block ranges included zero. Both shared positive claims failed Holm correction, with p values of 1. Full data BOLD boundary fits were positive, while prediction across excluded film blocks remained uncertain. The result limits these readers and processed signals. It does not show absent boundary processing. The package gives a compact benchmark, permitted electrical traces, code, source pins, all outcomes, figures and repeat instructions. A separate raw matrix solver checked all 62 primary person scores. Clinical coverage, unequal annotations, films, filters and cohorts limit scope. No positive cross-modal neural or causal claim is made.

hippocampus · event boundaries · intracranial EEG · fMRI · prediction · camera cuts · overlap · reproducibility