Electroencephalographic correlates of sleepiness: A meta-analysis and narrative review

Excessive daytime sleepiness has become a significant problem in society. Despite evidence for the effect of sleep need on EEG and calls for better EEG-derived predictors and diagnostics, no study has conducted a review to determine the best practice for using EEG to measure sleepiness. We aimed to provide this. Correlation coefficients were calculated for beta, alpha, theta, and delta wake EEG power across normal and sleep restriction/deprivation conditions, in order to determine which measures are most predictive of sleep need, in what conditions and using which parameters. Our results demonstrate that increases in EEG theta power are associated with increases in subjective sleepiness during both normal and sleep loss conditions. A similar increase was observed for EEG delta power, but only in the conditions of sleep loss. EEG alpha power showed no correlation with subjective sleepiness in either condition. This review also identified novel EEG markers, such as microstates and event-related potentials, and a limited number of studies that examined EEG measures of sleepiness in clinical sleep-disorder cohorts. Our study is the first to thoroughly examine the current EEG markers of sleepiness and the influence of recording conditions on sleepiness assessment using the EEG.