Light is the most important zeitgeber.1 You can potently shift your circadian clock by exposing yourself to light. Studies with continuous bright light have resulted in mean phase delays of 2.7h and advances of 2.3h in participants.2 Overall, it seems that delays can be induced more easily than advances.3 There are various factors that determine into what direction and to which extent light shifts your circadian phase:

  • Time of exposure: The greatest advancements can be achieved in the early subjective morning and delays in the subjective evening. See the following figure for a continuous curve as produced by study results from Khalsa et al.

    Exposure to bright light at different times of the day shifts your inner clock into different directions with varying magnitude. This circadian phase response curve, as estimated by Khalsa et al.2, represents the shift induced by 6.7h exposure to bright white light at different (biological) daytimes.


  • Brightness: The brighter the light is, the greater will be its impact on your circadian clock. Light intensity is measured in lux, which you can measure yourself with the help of “lightmeter” smartphone apps. Typical room illumination ranges from 50-500 lux.

    Light Intensity increases on a logarithmic scale from 0 (complete darkness) to around 100k lux (direct sunlight in summer) Source: [James Druzik]

    To effectively advance circadian phase, levels of illuminations are needed that by far exceed typical room lighting, while significant delays can be induced with no more than 15 lux brightness.3Comparison between circadian phase response curves to 6.7h of bright white light vs. low illumination background light (<15 lux). The latter is unlikely to advance circadian pahse, but can delay at evening exposure3.

  • Duration of exposure: The longer the exposure, the greater light influences your circadian clock – although there are natural limits, as can be seen in section “time of exposure” above. A study has shown that at same brightness and spectrum, 1h of bright white light has resulted in 40% shift magnitude compared to 6.7h bright white light exposure. At short exposure, the delays have been notably stronger than the advances.3Circadian phase response curve to 1 hour of bright white light, as estimated by Hilaire et al.3.

  • Wavelength: Light affects the circadian clock via photoreceptors in the eye, which are most sensitive to blue light with wavelengths around 480nm.1 In a study, 6.5h of approximately 90 lux blue light with 480nm had around 75% of the impact that 5k-10k lux bright light had at 6.7h exposure, which suggests that pure blue light is more efficient in shifting circadian phase than other forms of light.4 In another study (with fewer participants), just 20 minutes of 100 lux blue light (470 nm) after awakening advanced circadian phase by an average of 30min.

  • Angle: As most of the photoreceptors responsible for modulating circadian rhythm are located in the lower half of your retina, light that’s coming from above is more effective than light coming from below. The optimal angle for light therapy is believed to be 15° above eye level.1
    An illumination angle of 15 degrees above the eye is considered optimal for light therqapy.

[1] Blume, C., Garbazza, C., & Spitschan, M. (2019, September 1). Effects of light on human circadian rhythms, sleep and mood. Somnologie. Dr. Dietrich Steinkopff Verlag GmbH and Co. KG.

[2] Khalsa, S. B. S., Jewett, M. E., Cajochen, C., & Czeisler, C. A. (2003, June 15). A phase response curve to single bright light pulses in human subjects. Journal of Physiology. John Wiley & Sons, Ltd.

[3] St Hilaire, M. A., Gooley, J. J., Khalsa, S. B. S., Kronauer, R. E., Czeisler, C. A., & Lockley, S. W. (2012). Human phase response curve to a 1 h pulse of bright white light. Journal of Physiology, 590(13), 3035–3045.

[4] Rüger, M., St Hilaire, M. A., Brainard, G. C., Khalsa, S.-B. S., Kronauer, R. E., Czeisler, C. A., & Lockley, S. W. (2013). Human phase response curve to a single 6.5 h pulse of short-wavelength light. The Journal of Physiology, 591(1), 353–363.