EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Daily, prospective associations between sleep architecture and affect: insights from Bayesian multilevel compositional data analysis

Flora Le iD, Yang Yap iD, Dorothea Dumuid iD, Joshua F Wiley iD

Annals of Behavioral Medicine · 2025

DOI: 10.1093/abm/kaaf050 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Annals of Behavioral Medicine, 2025. Verfasst von Flora Le und 3 weiteren Autor:innen. Beteiligte Einrichtung: School of Psychological Sciences, Monash University, Clayton, VIC 3800, Australia. Hauptthemen laut MEDLINE: Affect, Sleep Stages, Sleep. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 51 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Emerging statistical methods addressing the multilevel compositional nature of sleep architecture can offer insights into how daily time reallocations between sleep stages (total wake time in bed [TWT], light sleep [Non rapid eye movement stage 1 and 2], slow wave sleep [SWS], and rapid eye movement [REM] sleep) are associated with post-sleep affect.

Purpose

This study investigated the daily, prospective association between sleep architecture and affect.

Methods

In 96 healthy, young adults across 15 consecutive days, sleep architecture was measured at night using electroencephalography (Z-Machine Insight+) and affect was self-reported using the PANAS-X at awakening. Bayesian multilevel compositional data analysis examined how reallocating time between sleep stages was associated with affect.

Results

Various reallocations of sleep stages predicted affect, at both within- and between-person levels. Between-person reallocation of 30 min/night from light or REM sleep to SWS was associated with ≥0.38 points higher high and low arousal positive affect, and from SWS to any other sleep stages was associated with ≥0.21 points higher high arousal negative affect. Within-person reallocation of 30 min/night from REM to any other stages predicted ≥0.05 points higher high arousal negative affect, and 30 min/night from TWT to SWS or REM predicted ≤−0.07 lower low arousal negative affect.

Conclusions

Findings highlight the distinct constellations of sleep architecture associated with affect in everyday life. Extension of SWS and REM for improving affect, while considering other off-set sleep stages, should be confirmed in experimental research in daily settings, to inform diagnostic and intervention strategies for sleep and affective disorders.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

AffectSleep StagesSleepHumansBayes TheoremFemaleMaleYoung AdultAdultElectroencephalographyMultilevel AnalysisProspective StudiesAdolescent

Stichwörter der Autor:innen

Bayesian multilevel compositional data analysisREM sleepaffectsleepsleep architectureslow wave sleep

Autor:innen und Institutionen

  • School of Psychological Sciences, Monash University, Clayton, VIC 3800, Australia.
  • School of Health and Biomedical Sciences, RMIT University, Melbourne, VIC 3000, Australia.
  • Alliance for Research in Exercise, Nutrition and Activity, Allied Health & Human Performance, University of South Australia, Adelaide, SA 5001, Australia.

ORCID-Profile: Flora Le, Yang Yap, Dorothea Dumuid, Joshua F Wiley

Zitieren

APA

Le, F., Yap, Y., Dumuid, D., & Wiley, J. F. (2025). Daily, prospective associations between sleep architecture and affect: insights from Bayesian multilevel compositional data analysis. Annals of Behavioral Medicine. https://doi.org/10.1093/abm/kaaf050

Vancouver

Le F, Yap Y, Dumuid D, Wiley JF. Daily, prospective associations between sleep architecture and affect: insights from Bayesian multilevel compositional data analysis. Annals of Behavioral Medicine. 2025. doi:10.1093/abm/kaaf050

BibTeX
@article{Le2025,
  author = {Le, Flora and Yap, Yang and Dumuid, Dorothea and Wiley, Joshua F},
  title = {Daily, prospective associations between sleep architecture and affect: insights from Bayesian multilevel compositional data analysis},
  journal = {Annals of Behavioral Medicine},
  year = {2025},
  doi = {10.1093/abm/kaaf050}
}
Kostenlos starten →