The 95% interval for confidence prediction gain from posterior alpha power included zero in 34 ds004368 participants under fixed ridge models for later trials

This study tested whether posterior 8–13 Hz EEG power improved confidence estimates for later trials in OpenNeuro ds004368 version 1.0.2. Fixed ridge models used early trials for training and later trials for testing. The behavior model used correctness, response times, recorded difficulty, direction, and trial sequence. Alpha power had a mean gain of 0.001 squared confidence percentage points across 34 eligible participants. The 95% participant bootstrap interval was −3.632 to 3.825 points squared. Thus, this test did not establish a prediction gain. The interval permits both gain and loss under the tested models and participant sampling assumptions. The package contains checked signal windows from all 39 public participants, behavior tables, source hashes, exclusion records, model fits, test estimates, and reproduction code. The EEG window precedes the stored movement event. It is not an independently checked measure of the time before physical movement. This is a conditional test on preprocessed data. It does not establish a causal effect or a disease difference.

confidence · EEG · alpha power · prediction · ridge regression · OpenNeuro · ds004368