Short and long sleepers: a difference in sleep capacity or in the tolerance of sleep pressure?

Abstract

Objectives Sleep duration varies greatly among individuals. Whether this variation has a biological basis is largely unknown. Here we compared two extreme phenotypic groups, short sleepers and long sleepers. We tested 1) whether there is difference in the maximal sleep capacity between the two groups, or 2) whether there is a difference in the tolerance of homeostatic sleep pressure as measured on the basis of cognitive performance. Methods Healthy young (18-30 y) individuals who based on actigraphy were either short sleepers (n=7, habitual bedrest <6.5 h) or long sleepers (n=11, >9 h) underwent a 28-day inpatient protocol, including 4 days of habitual sleep (HS), 20 days of extended (12 h) sleep opportunities, a 36-h sleep deprivation (SD) period, and 2 days of recovery sleep. Total sleep time (TST) was quantified daily with polysomnography, and performance with the psychomotor vigilance task (PVT) several times every wake episode. Performance variables included number of lapses (reaction times, RT > 500 ms), median speed (1/RT), and interpercentile (IPR) range (difference between 90th and 10th percentile in 1/RT). Results In the HS condition, TST was 5.8 h in the short sleepers and 8.9 h in the long sleepers (p < 0.001). At the end of the sleep extension protocol (average of last three nights), daily TST was 8.5 h in the short sleepers and 8.8 h in the long sleepers (n.s.). None of the PVT measures differed between the two groups in the HS condition. When given extended sleep opportunities, PVT performance improved in the short sleepers (p < 0.001) but not in the long sleepers. Two-hourly PVTs during the SD revealed that short sleepers showed fewer lapses (p < 0.001) and a smaller IPR (p < 0.04) than the long sleepers, particularly in the latter part of the SD. Conclusion The maximal sleep capacity of young healthy adults is approximately 8.9 h. The disparity in habitual sleep duration between short and long sleepers appears to reflect a trait-like difference in the tolerance of homeostatic sleep pressure rather than in the capacity to sleep. Short sleepers seem to possess a ´cognitive reserve´ that becomes apparent at very low and very high levels of sleep pressure.