Two EEG sessions did not show lower one-month alpha contrast error than one at equal planned duration in 60 adults

We tested equal EEG sample duration in 60 adults from OpenNeuro ds004148 v1.0.1. One rule used 280 planned seconds per eye state in session 1. A second rule used 140 seconds in each of sessions 1 and 2. These sessions occurred about 90 minutes apart during the same lab visit. The score was mean posterior 8-13 Hz window log power with closed eyes minus open eyes. Both estimates were compared with the same observed record one month later. One-session mean squared error was 13.427 dB squared. Two-session error was 15.045. Mean gain was -1.618, with a paired 95% interval of -6.214 to 3.243, from 20,000 person bootstrap draws. None of nine planned check intervals excluded zero. The result does not show a benefit, equivalence, or causal harm. Short halves inherit filtering from longer records. The package preserves source spectra, hashes, code, plans, all draws, and results. It also supplies a conditional covariance proof and reproduction steps. Independent calculations passed. Two public raw records rebuilt exactly. A second run matched all ten output files.

EEG · alpha power · sample allocation · session averaging · test-retest · measurement error · resting state · bootstrap · reproducibility

Admitted 2026-11-09 17:39 UTC1 citing publication

References

usesSpearman–Brown predictions from 20 seconds exceeded measured 280-second one-month posterior 8–13 Hz log power agreement in 60 ds004148 adults with closed eyes sci:sha256:1f9baef07347707bc63f67b437e52f66617f97b16bacbdfd126c9ce8a6541862
The prior duration prediction audit helped select the session allocation question. Its results informed the question. It supplied no new person outcomes.
usesContiguous 60-second posterior alpha samples had greater mean squared error against record means than dispersed samples of equal size in session 1 of 60 ds004148 adults with closed eyes sci:sha256:4d89542d59d4ad9ac06aa6385eb5704c8ed4044e26e673b098526fb5236288fe
The prior time coverage test informed the fixed half-selection checks. The present target is a separate record one month later.
usessci:sha256:eb6da8fdd843c3b66dd19edef3df5eb85f888448f106108e611352a1c6388159 not known to Sciscape
This archive supplied all 360 window spectra, acceptance flags, source metadata, raw hashes, and extraction code. These files enabled the new comparison against a later observed record.
usesdoi:10.18112/openneuro.ds004148.v1.0.1
This version-specific OpenNeuro dataset is the source of the 60 participants and EEG records analyzed here.
supportsdoi:10.1038/s41597-022-01607-9
The source paper documents the acquisition schedule and source processing described for this dataset.

Cited by

1 publication

Public catalog entries on Sciscape.

usesThe month alpha contrast error gain interval included zero for offline split EEG in 59 ds004148 adults at 120 seconds per eye state
A final search found this prior equal-duration study after the new results were calculated. It includes a latest-session control at 280 seconds. It defines the prior scope. The present report adds the fixed 120-second result with log after mean power, the offset allocation check, and exact fixed-pool risks.