Pre-intervention test-retest reliability of EEG and ERP over four recording intervals
Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt
- Pre-intervention test-retest reliability of EEG and ERP over four recording intervals
Forlagets udgivne version, 5 MB, PDF-dokument
In this study we present the test-retest reliability of pre-intervention EEG/ERP (electroencephalogram/event-related potentials) data across four recording intervals separated by a washout period (18–22 days). POz-recording-reference EEG/ERP (28 sites, average reference) were recorded from thirty-two healthy male participants. Participants were randomly allocated into different intervention sequences, each with four intervention regimens: 10 mg vortioxetine, 20 mg vortioxetine, 15 mg escitalopram and Placebo. We report classical EEG spectra: δ (1–4 Hz), θ (4–8 Hz), α (8–12 Hz), β (12–30 Hz), γ1 (30–45 Hz) and γ2 (45–80 Hz) of resting state and vigilance-controlled, and of auditory steady state response, as well as ERP components N100, P200 and P300 in auditory oddball task and error related negativity (ERN) and error positivity (Pe) in hybrid flanker task. Reliability was quantified using intra-class correlation coefficient (ICC). We found that θ α and β of continuous EEG were highly reliable (ICCs ≥ 0.84). Evoked power of other tasks demonstrated larger variability and less reliability compared to the absolute power of continuous EEG. Furthermore, reliabilities of ERP measures were lower compared to those of the EEG spectra. We saw fair to excellent reliability of the amplitude of the components such as Pe (0.60–0.82) and P300 (0.55–0.80). Moreover, blood tests confirmed that there was no measurable drug carry-over from the previous intervention. The results support that EEG/ERP is reliable across four recording intervals, thus it can be used to assess the effect of different doses and types of drugs with CNS effects.
|Tidsskrift||International Journal of Psychophysiology|
|Status||Udgivet - 1 dec. 2018|
Antal downloads er baseret på statistik fra Google Scholar og www.ku.dk