Context: In humans, sleep duration often determines the night (dark) length experienced because we close our eyes when we sleep and are exposed to artificial or natural light when we are awake. While it is recognized that there is an increasing trend in modern society toward shorter sleep, it is not known how short nights (long photoperiods) affect the human circadian system.

Objective: Here we investigated for the first time the effects of night length on circadian phase shifts to light in humans.

Design and Setting: Eight young healthy subjects experienced two weeks of 6 h sleep episodes in the dark (short nights) and two weeks of long 9 h sleep episodes (long nights) in counterbalanced order. After each series of nights, they were exposed to four 30 min pulses of morning bright light (5,000 lux) that advanced by 1 h per day for 3 consecutive days, while night (dark) length was maintained at 6 or 9 h. Circadian phase was determined from the circadian rhythm of melatonin in dim light before and after the 3 day bright light treatments.

Results: The phase advance in the melatonin rhythm during the short nights was less than half of that observed during the long nights (P < 0.05).

Conclusions: This result shows for the first time that people who curtail their sleep may unwittingly reduce their circadian responsiveness to morning light. This finding also demonstrates that sleep length can alter human circadian function and has important implications for enhancing the treatment of circadian rhythm sleep disorders.