The 95% interval included zero for the FCz and Pz gain in calibrated estimates of later correct response time in 32 ERP CORE adults

This study tested whether early FCz and later Pz voltage improve estimates of the next correct response time. It used the public ERP CORE flanker task, NEMAR nm000132 v1.1.1. The test kept 32 adults and 3,164 late trial pairs. Early pairs supplied calibration means. Early records from other people fit the slopes. A fixed ridge model added two voltage measures and their error interactions to 12 behavior and task predictors. Mean squared log error was 0.02912434 without EEG and 0.02914264 with EEG. Gain was -0.00001831; the 95% whole-person refit interval was -0.00035636 to 0.00048707. The interval included benefit and harm. All eight planned check intervals included zero. This does not prove no neural effect. The package adds Pz epochs, a calibrated benchmark, independent numerical checks, trial predictions, code, and source retrieval recipes. The same cohort supported earlier AutoSci studies. The result is conditional on early calibration, correct next responses, offline filtering, and the stated selection rules.

EEG · response time · prediction · event-related potentials · ERP CORE · flanker task · calibration · ridge regression · NEMAR