Spectral dependency of the human pupillary light reflex. Influences of pre-adaptation and chronotype

Abstract

Non-visual photoreceptors (ipRGCs) and rods both exert a strong influence on the human pupil, yet pupil models regularly use cone-derived sensitivity as their basis. This inconsistency is further exacerbated by the fact that circadian effects can modulate the wavelength sensitivity. We assessed the pupillary reaction to narrowband light stimuli in the mesopic range. Pupil size for eighty-three healthy participants with normal color vision was measured in nine experimental protocols with varying series of continuous or discontinuous light stimuli under Ganzfeld conditions, presented after 90 seconds of dark adaptation. One hundred and fifty series of stimulation were conducted across three experiments, and were analyzed for wavelength-dependency on the normalized pupillary constriction (nPC), conditional on experimental settings and individual traits. Traits were surveyed by questionnaire; color vision was tested by Ishihara plates or the Lanthony D15 test. Data were analyzed with generalized additive mixed models (GAMM). The normalized pupillary constriction response is consistent with L+M-cone derived sensitivity when the series of light stimuli is continuous, i.e., is not interrupted by periods of darkness, but not otherwise. The results also show that a mesopic illuminance weighing led to an overall best prediction of pupillary constriction compared to other types of illuminance measures. IpRGC influence on nPC is not readily apparent from the results. When we explored the interaction of chronotype and time of day on the wavelength dependency, differences consistent with ipRGC influence became apparent. The models indicate that subjects of differing chronotype show a heightened or lowered sensitivity to short wavelengths, depending on their time of preference. IpRGC influence is also seen in the post-illumination pupil reflex if the prior light-stimulus duration is one second. However, shorter wavelengths than expected become more important if the light-stimulus duration is fifteen or thirty seconds. The influence of sex on nPC was present, but showed no interaction with wavelength. Our results help to define the conditions, under which the different wavelength sensitivities in the literature hold up for narrowband light settings. The chronotype effect might signify a mechanism for strengthening the individual´s chronotype. It could also be the result of the participant’s prior exposure to light (light history). Our explorative findings for this effect demand replication in a controlled study.