Stable inter-individual differences in slow-wave sleep during nocturnal sleep and naps

Abstract

Stable interindividual differences in polysomnographically assessed nocturnal sleep have been reported, particularly in slow-wave sleep (SWS). This paper compares trait-like differences in nocturnal sleep periods and naps, after controlling for prior sleep history. Twelve healthy males (mean age, 27.3 years; SD, 4.9 years) participated in four weekend laboratory protocols with sleep monitoring (actigraphy, diaries) beginning one week prior to the first protocol. Each Friday night, participants went to bed from 22:00 until 06:00 (baseline sleep), and were then randomized to one of four conditions with extended wakefulness until 08:00 on Sunday morning. The conditions were: naps of 20 min, 40 min, or 60 min ending at 02:00 on Sunday morning, or no nap. All laboratory sleep was recorded polysomnographically. Across baseline sleep periods, intraclass correlation coefficients (ICC; measuring stable interindividual differences) were substantial for stage 1 sleep (min) (ICC = 0.74), % stage 1 sleep (ICC = 0.68), SWS (min) (ICC = 0.73), %SWS (ICC = 0.80), and sleep latency (ICC = 0.75). Adding average actigraphic sleep duration for the preceding three days as a covariate caused no substantive change in ICC values. For nap sleep (40-min nap and first 40-min of the 60-min nap), ICCs were substantial for minutes SWS (ICC = 0.71), %SWS (ICC = 0.78), and latency to SWS (ICC = 0.69). Baseline sleep was a significant predictor of nap sleep for SWS (min) and %SWS. The findings confirm trait-like differences between individuals in the amount of SWS they express in nocturnal sleep and naps, after controlling for prior sleep history. This raises important questions about the definition of “normal” sleep architecture in both clinical and experimental settings.