The theta-to-alpha EEG difference between arm and button responses had a 95% confidence interval that included zero in 14 CMx7-MM adults after fixed signal screening

We tested whether a fixed theta-to-alpha EEG index increased when an arithmetic and 2-back task used an arm response instead of a button response. We used public CMx7-MM data from OpenNeuro ds007554 version 1.0.0. A local plan fixed the channels, bands, signal screening, cohort rule, and paired analysis before EEG power was computed. The primary estimate included 14 adults and 35 matched repetitions. The mean Full minus button difference was −0.352 dB (95% t interval −0.916 to +0.213; p = 0.202); the participant bootstrap interval was −0.859 to +0.130 dB. These results do not establish an increase, zero difference, or statistical equivalence. A fixed cognitive task check showed a positive difference under the primary method, but had a slightly different cohort and also changed target detection and response requirements. Signal loss was substantial, and the task contrast cannot identify a pure motor or cognitive cause. The package preserves 8,145 EEG windows, derived features, quality masks, results, figures, code, source metadata, and a verified raw-data retrieval recipe. Source URLs, object versions, and hashes are recorded. Independent signal reading, extraction, and inference replay checks passed.

electroencephalography · EEG · workload · theta-to-alpha index · motor response · cognitive task · signal quality · paired analysis · OpenNeuro · CMx7-MM