Abstract
Introduction
Studies have shown that sleep affects working memory (WM) improvement, but specific electrophysiological features are unclear (Sattari et al., 2019; MacDonald et al., 2018). In addition, sex differences have been found in both sleep and working memory (Mong, 2016; Harness, 2008). The goal of this study is to identify sex differences in EEG correlates of working memory over a night of sleep.
Methods
Sixty-three healthy (33 females), college-aged adults without sleep disorder were enrolled. A 32-channel electroencephalogram (EEG) cap was used to record brain activity during sleep. Operation span (OS) task was used to evaluate WM performance. Participants reported to the laboratory in the evening, performed OS before sleep (test1) and after sleep (test2). Trials were divided into easy and hard trials based on the number of letters subjects had to recall. A repeated-measure analysis of variance was conducted to examine the effects of time and trial difficulty. Paired-sample t-tests between test 1 and test 2 were conducted for males and females. Pearson’s correlations were examined between WM performance at test 1 and the difference score between test 1 and test 2 and EEG frequency bands. The Benjamini–Hochberg method was used to control for multiple comparisons.
Results
There was an increase in performance across the night in WM hard trials across all subjects (F(1,62)=4.86, p=0.03), no effect for easy trials (p>0.05). Females, but not males, showed a significant decrease in easy trials (t62= 2.40, p=0.02), while both males and females showed improvement in hard trials across the night. Females showed a positive correlation between test 1 hard trials and slow sigma, delta, slow oscillation in stage 3, this correlation is not seen in males. No correlations between overnight improvement and EEG bands were found.
Conclusion
Consistent with previous studies, participants showed better memory performance over a night of sleep, and the WM performance was associated with slow wave activity in females. Slow sigma also plays a role in the WM performance for females, indicating a possible role of sleep spindles. These associations were not shown in males, suggesting sex hormones mediate sleep’s impact on WM performance.
Support (if any):