Altering Temporal Dynamics of Sleepiness and Mood During Sleep Deprivation: Evidence from Resting-State EEG Microstates

<b>Objective:</b> Sleep deprivation negatively affects mood and sleepiness across subjective, objective, behavioral, and neuroimaging measures. However, the relationship between abnormal brain dynamics after sleep deprivation and mood or sleepiness impairments, from a temporal perspective, remains underexplored. Electroencephalogram microstate analysis offers a valuable approach for investigating the large-scale dynamics of brain networks. <b>Methods:</b> We implemented a strict sleep deprivation protocol with 71 participants, collecting resting-state electroencephalogram data, subjective sleepiness, objective alertness, and mood assessments after normal sleep and sleep deprivation (SD) conditions. Microstate time parameters were compared between sleep control (SC) and SD conditions. Additionally, we investigated how changes in these parameters correlated with subjective or objective measures of sleepiness and mood changes between SC and SD. <b>Results:</b> SD significantly decreased the mean duration and occurrence of microstate B while increasing those of microstate C. A decrease in microstate B occurrence significantly correlated with a higher Karolinska Sleepiness Scale score, whereas a reduction in microstate B duration indicated an increased response time on the Psychomotor Vigilance Performance. No significant associations were observed between microstate parameters and positive mood decline between SC and SD. Pearson correlation analysis was performed on the positive mood scores in both conditions. The findings demonstrated a significant positive relationship between positive mood scores and the mean duration of microstate B under SD conditions. <b>Conclusions:</b> Using a large SD dataset, this study demonstrated that subjective or objective sleepiness and positive mood were associated with decreased microstate B. These findings suggest that SD disrupts neural dynamics within the visual network.