Bands of 5 Hz set from session 1 showed no clear alpha reactivity ICC gain over fixed 8-13 Hz bands over one month in 60 ds004148 adults

We tested whether a frequency band set for each person improved repeat EEG measurements in 60 adults from OpenNeuro ds004148 version 1.0.1. For each person, the largest closed-eye power density from 8 to 13 Hz in session 1 set the center of a 5 Hz band. We then tested sessions 2 and 3, about one month apart. Neither test session supplied data for band selection. The outcome was the closed-eye to open-eye power ratio in decibels. The comparison used a fixed 8-13 Hz band. The primary absolute-agreement ICC was 0.4347 for the person band and 0.4335 for the fixed band. The paired ICC change was 0.0012, with a 95% person-bootstrap interval of -0.0277 to 0.0360 from 10,000 draws. Six planned method checks also gave no clear gain. This does not prove equal performance. The evidence applies to the tested band maximum rule, source processing, and young adult sample. The package preserves all 360 input spectra, source hashes, source and analysis code, tables, random draws, and reproduction instructions. Independent numerical checks passed. Two fresh full EEG record checks rebuilt the archived arrays exactly.

electroencephalography · alpha reactivity · test-retest reliability · intraclass correlation · individualized frequency bands · bootstrap uncertainty