Occurrence of 'microsleeps' during daytime sleep onset in normal subjects

The main aim of this study was to explore whether the multiple sleep latency test (MSLT) could be made more sensitive to low daytime sleepiness in normal, healthy subjects by adopting a shorter period of sleep (microsleep) as a sleep onset criterion. Subjects underwent MSLTs under two conditions: after normal (baseline) nighttime sleep, and after nighttime sleep extension (creating a 'floor effect' of minimal daytime sleepiness). MSLT sleep onset thresholds of 5 s (microsleeps), 30 a (the norm) and 90 s of sustained sleep gave 3 separate sleep latency scores for 240 MSLT trials derived from 10 subjects. With low daytime sleepiness, whether this be in the morning after baseline sleep or throughout the day after sleep extension, the 5 s sleep onset criterion was a more sensitive measure of sleepiness than the established 30 s criterion. This was the case both for sleep onset latency and for the frequency of sleep onsets. Spectral analyses of the EEG indicated that successive microsleep episodes generally became more substantial, and, depending on the level of sleepiness, culminated in more overt signs of sleep. There was little difference between the 30 s and 90 s criteria for sleep onset latency scores, although there was a small but significant difference between them in the frequency of sleep onsets. As daytime sleepiness increased, particularly in the afternoon and under baseline, the 5 s criterion reached a ceiling, with the 30 s criterion becoming more sensitive.