0008 Wake Onset Variability Effect on Functional Connectivity in the Default Mode Network

Introduction Decoupling patterns of sleep and wake from underlying circadian oscillations is associated with poor sleep health. While this has been studied with respect to shift work, it is unclear if minor variations in normative sleep/wake patterns affect the restoration of cognitive function during sleep intervals while controlling for total sleep time. It is commonplace to adhere to a fixed sleep/wake schedule, however, this practice does not account for day-to-day variations in physiological need for sleep. We examined the relationship between sleep-interval variability and fMRI resting state functional connectivity while controlling for sleep duration. Methods Participants (n=21; 13 female; Age=23.3, SD=4.7) wore an actigraph for 7 days followed by a daytime functional magnetic resonance imaging (fMRI) session. Sleep/wake onset variability was measured from double-rated actigraphy data and cross referenced with sleep journals. Functional connectivity analysis and preprocessing of fMRI images were conducted in the CONN toolbox (SPM12). Seed-based functional connectivity maps were estimated using BOLD activity of the posterior parietal cortex node of the default mode network (DMN) as the seed region. Functional connectivity strength was represented by Fisher-transformed bivariate correlation coefficients from a weighted GLM, modeling the association between the seed and other brain regions. Results Wake onset variability (WOV) predicted functional connectivity of the DMN (controlling for age, gender, and sleep duration). WOV was significantly associated (Pthreshold <.005 ; Pfwe <.05) with a higher degree of anticorrelation between the posterior parietal node of the DMN and several regions of the frontoparietal network (FPN): Right superior frontal gyrus, Right mid-frontal gyrus, and Right frontal pole, as well as the occipital cortex and cerebellum. In contrast, sleep onset variability showed no significant changes in functional connectivity within these areas. Conclusion Greater WOV was associated with stronger inverse connectivity between the DMN and FPN. This suggests that increased WOV may improve attentional control through greater top-down suppression of the DMN. We speculate that WOV is representative of sleep satiety, resulting in voluntary arousal from sleep, rather than enforced arousal. These results suggest that sleep timing affects attentional control irrespective of duration. Support (if any)