From Sunlight to Screens: Modeling When Light Exposure Matters Most for Sleep and Circadian Health

Understanding the effects of light on the body at different times of the 24 h solar day is a topic of increasing interest. In this paper, we use a mathematical model from the literature to simulate what would be expected of the human circadian clock on different light schedules. We first reproduce an influential experiment which found eBooks, when compared to a paper book, delayed sleep by roughly 10 min and melatonin onset by 1.5 h. The model is able to match the delay in sleep onset but struggles to reproduce the melatonin phase delay. However, certain initial conditions and parameters are capable of phase shifts consistent with the original study's magnitude, suggesting that the original study's finding may have been influenced by the pre-study entrainment or variability among the participants. We next simulate the same protocol under higher daytime light levels (increasing baseline illumination from 90 to 500 lux) and find that brighter daytime exposure reduces both sleep onset latency and the variability in phase delay attributable to evening eBook light. Finally, we explore how the timing of a bright light pulse during the day changes outcomes, such as sleep onset and circadian amplitude, and how these effects interact with light during the other hours of the 24 h day. Together, these modeling results suggest robust daytime light exposure confers resilience against the circadian-disruptive effects of evening light, generating testable predictions regarding the timing and intensity of beneficial light interventions for maintaining circadian alignment.