1 Department of Psychology, Concordia University, Montreal, QC, Canada.
2 EngAGE: Center for Research on Aging, Montreal, QC, Canada.
3 Montreal Neurological Institute, McGill University, Montreal, QC, Canada.

Introduction: Sleep quality is often thought to be a key determinant of cognitive performance, particularly in older adults who experience age-related changes in sleep architecture. However, the extent to which nightly variations in sleep quality impact next-day cognitive performance remains unclear-in part because it has only recently become practical to measure sleep over multiple nights.

Methods: In this study, we used an in-home wearable electroencephalography (EEG) device to monitor sleep patterns over ~10 nights in 17 healthy older adults, assessing metrics of sleep quality such as wake after sleep onset and the density of slow oscillations and sleep spindles. Next-day cognitive performance was evaluated using two computerized neuropsychological tasks measuring executive functions (inhibition and cognitive flexibility), and their relationships to sleep metrics were explored.

Results: Although participants placed the EEG device themselves, a high proportion of sleep data was usable (~71%), and clear nightly variations in sleep quality were captured. Sleep recordings showed considerable variability in sleep quality metrics across nights, with large inter-individual differences. However, we found no effects of either macro- or microarchitectural sleep metrics on executive task outcomes the following day.

Discussion: These results do not rule out the possibility that some aspects of cognitive performance may be affected by daily fluctuations in sleep quality; however, they suggest that inhibition and cognitive flexibility, which underlie reasoning and problem solving, may be relatively resilient to nightly sleep variability in older adults. The findings also demonstrate the feasibility of using emerging portable devices to extend sleep studies at home and over multiple nights in older adults, while providing variance estimates and effect sizes to guide power and sample size planning for future studies.