AUTHOR=Harlow Tylor J. , Jané Matthew B. , Read Heather L. , Chrobak James J. 

TITLE=Memory retention following acoustic stimulation in slow-wave sleep: a meta-analytic review of replicability and measurement quality

JOURNAL=Frontiers in Sleep

VOLUME=Volume 2 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/sleep/articles/10.3389/frsle.2023.1082253

DOI=10.3389/frsle.2023.1082253

ISSN=2813-2890

ABSTRACT=The role of slow oscillations during non-REM sleep on memory retention has become an area of great interest in the recent decade.  One response to this body of literature has been to examine the efficacy of acoustic stimulation during non-REM sleep to facilitate slow oscillations and associated memory retention.  In this meta-analysis, we focus on literature using audible noise-burst sound stimulation to modulate overnight retention of word pairs Ks = 12 studies, Kes = 14 effect sizes, n = 206 subjects). While recent reviews and meta-analyses of the literature have found moderate effects of acoustic stimulation during slow-wave sleep on memory retention (Wunderlin et al 2020, Stanyer et al 2021), we demonstrate a steady, yearly decline in effect size that accounts for 91.8% of the heterogeneity between studies. The predicted effect on memory retention in 2013 (year of seminal paper - Ngo et al 2013b) favored the acoustic stimulation condition at d = 0.99 (95% CI [0.49, 1.49]), while the predicted effect in 2021 (year of most recent study, Harrington et al 2021) declined to a moderate and significant effect favoring no acoustic stimulation at d = -0.39 (95% CI [-0.73, -0.05]). No coded study-level characteristics could account for the decline in effect sizes over time other than the publication date alone. We estimate that 34% of subjects are not actually blind to the acoustic stimulation condition due to hearing acoustic stimulation during sleep.  This suggests that future studies should include a placebo acoustic stimulus condition. Lastly, our meta-analysis finds that the test-retest difference score for memory task performance has near zero reliability (r = .01 95% CI [-.18, .21]). These results highlight the need for open-data, placebo phase-locked acoustic stimulation conditions, and improvements in the reliability of memory retention tasks.