The interindividual variability of sleep timing and circadian phase in humans is influenced by daytime and evening light conditions

Abstract

Human cognitive functioning shows circadian variations throughout the day. However, individuals largely differ in their timing during the day of when they are more capable of performing specific tasks and when they prefer to sleep. These interindividual differences in preferred temporal organization of sleep and daytime activities define the chronotype. Since a late chronotype is associated with adverse mental and physical consequences, it is of vital importance to study how lighting environments affect chronotype. Here, we use a mathematical model of the human circadian pacemaker to understand how light in the built environment changes the chronotype distribution in the population. In line with experimental findings, we show that when individuals spend their days in relatively dim light conditions, this not only results in a later phase of their biological clock but also increases interindividual differences in circadian phase angle of entrainment and preferred sleep timing. Increasing daytime illuminance results in a more narrow distribution of sleep timing and circadian phase, and this effect is more pronounced for longer photoperiods. The model results demonstrate that modern lifestyle changes the chronotype distribution towards more eveningness and more extreme differences in eveningness. Such model-based predictions can be used to design guidelines for workplace lighting that help limiting circadian phase differences, and craft new lighting strategies that support human performance, health and wellbeing.

Figure 1

Entrainment phase angle distribution for a population of 200 simulated individuals with a normally distributed intrinsic circadian period with means (± SD) of 24.15 (± 0.2) h when entrained to various corneal illuminances (i.e., at the eye position). (a) Results for the LD 16:8 schedule. (b) Results for the LD 10:14 schedule. The central marker indicates the median, and the bottom and top edges indicate the 25th and 75th percentiles, respectively. The whiskers indicate the most extreme data points (population minima and maxima).

Figure 2

Entrainment phase angle distribution for a population of 200 simulated individuals with a normally distributed intrinsic circadian period with means (± SD) of 24.15 (± 0.2) h. Illuminances refer to corneal light exposure (i.e., at the eye) for daytime (wake-19:00) illuminance (L₁) set to the values indicated on the x-axis. Evening light exposure L₂ was simulated at 30 lx from 19:00 until sleep onset. The central marker indicates the median, and the bottom and top edges indicate the 25th and 75th percentiles, respectively. The whiskers extend to the most extreme data points (population minima and maxima).

Figure 3

Entrainment phase angle distribution for a population of 200 simulated individuals with a normally distributed intrinsic circadian period with means (± SD) of 24.15 (± 0.2) h. Illuminances represent the corneal light exposure (i.e., at the eye position). Results are presented for three different daytime (wake-19:00) illuminances (L₁ = 200, 800 and 2000 lx at the eye), all with evening (19:00-sleep) illuminance L₂ of 30 lx.

Figure 4

Entrainment phase angle distribution for a population of 200 simulated individuals with a normally distributed intrinsic circadian period with means (± SD) of 24.15 (± 0.2) h. Illuminances represent the corneal light exposure (i.e., at the eye position). Results are presented for two different evening (19:00-sleep) illuminances of 10 and 35 lx, respectively, all with 200 lx daytime (wake-19:00) illuminance.

Figure 5

Entrainment phase angle distribution for a population of 200 simulated individuals with a normally distributed intrinsic circadian period with means (± SD) of 24.15 (± 0.2) h. Illuminances refer to corneal light exposure (i.e., at the eye). Results are presented for daytime (wake-19:00) illuminances (L₁) of 200, 800 and 2000 lx, and evening (19:00-sleep onset) illuminance (L₂) set to the values indicated on the x-axis. The central marker indicates the median, and the bottom and top edges indicate the 25th and 75th percentiles, respectively. The whiskers extend to the most extreme data points (population minima and maxima). We note that in the absence of sufficient differentiation between daytime and evening light the model fails to entrain as also noted and discussed in. Results are presented only for combinations of daytime-evening illuminances and intrinsic circadian periods for which the model entrains to 24-h rhythms.

Figure 6

Model-predicted chronotype distribution under a constant daytime (wake-19:00) illuminance (L₁) of 200 lx (at the eye) and various evening (19:00-sleep) illuminances (L₂), all at the eye position: (a) 0 lx, (b) 10 lx, (c) 20 lx, and (d) 35 lx. The distributions are based on hourly bins. The population is classified into seven chronotypes indicated in the legends according to their midsleep timepoints.

Figure 7

Example of daily shift in the predicted core body temperature minimum during entrainment. Results are presented for a 24 h light–dark cycle (LD 16:8 at 200 lx). Here, the model reaches a stable entrainment after 13 days. We consider that a stable entrainment has been reached if CBTmind − CBTmind − 1 < 0.01 ℎ. Initial state variable values (x and y) were set to − 0.9 and 0. 1, respectively, whereas variable n was set to 0.05.