LAUSR

Individuals with post-traumatic stress disorder (PTSD) experience altered sleep and daytime function, including deficits in working memory (WM), the ability to store and manipulate information over short timeframes. As sleep contributes to WM, understanding how sleep parameters influence changes in WM may provide insight into potential intervention targets to improve or restore daytime function. Here, we investigated the relationship between sleep and WM improvement in Veterans with and without PTSD. Forty-eight post-911 Veterans without PTSD (Control) and 37 with PTSD (PTSD) completed a 48-hour lab stay during which WM was assessed using a n-back task. Nighttime polysomnography, using high-density (64-channels) electroencephalography, quantified time spent in non-REM and REM sleep and log-transformed spectral activity for non-REM sleep in delta (.5-4Hz), theta (4-8Hz), alpha (8-12Hz), sigma (12-16Hz), and beta (16-32Hz) bands. Pearson’s correlations were used to assess associations between baseline sleep and baseline WM on 1-back trials. Within control and PTSD groups, independent samples t-tests were used to compare changes in sleep across nights between improvers and non-improvers categorized by 1-back accuracy changes across days. Proportions of improvers and non-improvers were similar between groups (χ 2 = .023, p = .880). Within either group, baseline sleep did not relate to baseline WM and changes in time spent in each sleep stage did not differ between improvers and non-improvers. Within the control group, improvers (n = 15; 3.67 ± 3.56) had a greater increase (t = -2.826, p = .007) in delta activity than non-improvers (n = 33; .77 ± 3.20), but this relationship was not observed in the PTSD group (11 improvers, 26 non-improvers). Increased delta activity related to improved WM in the Control but not PTSD group. This suggests individuals with PTSD do not improve WM through non-REM sleep but that it may be a useful intervention target. USAMRMC MOMRP PT-130572 (PI: Reifman)