Mean response slopes were positive for previous target gaps and smaller for next gaps in 49 EEG and 16 fMRI adults during visual tasks

We tested the direction of target gap relations in two public visual target studies. Scalp EEG and BOLD fMRI came from separate cohorts. Development used 12 EEG IDs and one MRI person. A local code lock preceded new reserved scores. Source parser repairs preceded every reserved MRI score. The test retained 49 EEG and 16 MRI adults. Previous log gap slopes were +0.1063 [+0.0440, +0.1685] and +0.1476 [+0.0830, +0.2122]. Previous minus next gap slopes were +0.0947 [+0.0050, +0.1844] and +0.1638 [+0.0492, +0.2784]. Brackets give pointwise 95% Student limits across people. EEG slopes use target voltage SD. MRI slopes use a target contrast learned from other runs. These scales are not equal. One shared rule required positive previous slopes, positive direction contrasts, and positive assay responses in both studies. The fixed shared rule passed. Its p was 0.0198198. A source audit checked 51 MRI schedules and excluded one unresolved run. Constant simulated targets gave positive previous gap effects in simple MRI epoch averages. The event model removed them with the correct response shape. A changed shape left bias. All negative slopes and secondary checks remain. The earlier MRI sampling check and two EEG checks had direction limits across zero. The contribution is a fixed comparison tool, a source audit and a human benchmark. Different tasks, gap ranges, cohorts, sampling, missing RT values and response models limit scope. No common neural recovery cause was established. Code, compact inputs, source pins, terms, figures and repeat steps are supplied.

EEG · fMRI · target detection · event history · response overlap · reproducibility