|
EEG slow-wave activity (SWA) is a marker of nonREM sleep intensity and serves as an indicator of sleep homeostasis. The homeostatic Process S reflects the prior history of sleep and wakefulness and is modeled by a saturating exponential function during waking and an exponential decline during sleep. There is considerable inter-individual variation in responses to sleep deprivation. The aim of the present study was to investigate whether the time constants of Process S are trait-like. We investigated this in healthy young adults who underwent a laboratory protocol with repeated sleep deprivation. Each subject's data consisted of EEG recordings from eight sleep periods (12 h time in bed, 22:00-10:00), which were interspersed with three 36 h periods of sustained wakefulness (Tucker et al., 2007). Empirical mean SWA per nonREM sleep episode at episode midpoint served as input for least-squares model parameter estimation. Time constants were constrained to a physiological range. Parameters were estimated for 6 different pairings of the recording nights. Random-effects ANOVA was used to investigate the intra- and inter-individual variation in the estimates of the time constants. The standard deviation across subjects for the time constant of buildup was 4.9h, and the inter-individual differences explained 59% of the variance. The standard deviation across subjects for the time constant of the decline was 0.3h, and inter-individual differences explained only 26% of the variance. Thus, there was substantial inter-individual variability in the dynamics of Process S during wakefulness but not during sleep. It remains to be determined to what extent the inter-individual differences during wakefulness are a byproduct of the model fitting procedure, or a result of systematic inter-individual differences in waking activities, or evidence of a trait-like feature.
Supported by SNSF grant 320000-112674 and NIH grants HL70154 and RR00040 |
|