A single day of natural daylight exposure and physical activity does not conclusively change circadian phase: A Bayesian crossover experiment

Abstract

Among the environmental signals that shape human physiology, the daily light-dark cycle is the primary zeitgeber of the human circadian system. Physical activity may provide an additional non-photic time cue and frequently co-occurs with daylight in everyday life. However, their independent and combined effects on circadian phase have rarely been disentangled under naturalistic daylight conditions. In a counterbalanced mixed-design experiment, 41 healthy adults (24 women; mean age = 23.7 years, SD = 4.3) participated in up to three 40-h visits on fixed individual sleep-wake schedules. Each visit comprised an 8-h exposure beginning 1.5 h after wake to natural daylight, daylight attenuated by amber-tinted glasses, or indoor electric light (within subjects; mean photopic illuminances: 1196, 460, and 75 lx), with otherwise dim-light conditions. Participants hiked for approximately 4 h or remained sedentary (between subjects; activity indexed by median heart rate reserve [HRR]; group means: 32.7% and 2.7%). Bayesian hierarchical models used individual photopic illuminance (ActTrust 2) and median HRR to predict change in dim-light melatonin onset (DLMO; primary outcome) and exploratory outcomes (daytime and evening sleepiness, mood, and wellbeing, daytime perceived exertion, and evening psychomotor vigilance). Per standard-deviation increase in log-transformed illuminance and median HRR, estimated DLMO changes were -4.22 min (95% HDI [-10.93, 2.22]; BF = 0.70) and -6.48 min (95% HDI [-13.79, 1.27], BF = 1.56), respectively. Evidence remained inconclusive for both predictors and their interaction. Exploratory findings were largely inconclusive, practically negligible, or favoured the absence of effects. The circadian effects of a single day of natural daylight exposure, with or without physical activity, therefore remain unclear. Resolving this question may require larger samples and brighter, repeated, or habitual exposure timed to phases of greater circadian light sensitivity.