Item

Abstract

We examined if the pupil response to melanopsin stimulation differs from that to cone stimulation, and if these pupil responses are stable in a healthy population. Following a pre-registered protocol (https://osf.io/76u9x/), we measured the pupil response to 3-second unipolar spectral pulses in each of 24 people during two separate sessions. These were designed to selectively stimulate either melanopsin (Mel) or the L, M, and S cones (LMS). The pulses produced 400% nominal contrast upon the targeted retinal mechanism and were presented against a rod-saturating background (~100 cd/m2 for Mel, ~40 cd/m2 for LMS). The group average shape of pupil constriction expressed as percentage change from baseline for each stimulus was highly consistent from the first to second session, suggesting that the pupil response is reproducible in a healthy population (Mel r = 0.995; LMS r = 0.999). We quantified the temporal dynamics of the pupil response. This revealed that the pupil response to Mel stimulation was prolonged relative to that of LMS stimulation. 21 subjects underwent a third pupillography session identical in structure to the first two except at higher light levels (background luminance for Mel and LMS pulses were increased to ~270 and ~100 cd/m2 respectively). The shape of the group average response (expressed in percentage change) to Mel and LMS stimulation was again highly overlapping with the responses from the first two sessions (Mel r = 0.994; LMS r = 0.999), suggesting that these pupil responses are independent of background light level. We find the pupil response to selective melanopsin stimulation is highly stable in a healthy population, and that the temporal dynamics of the Mel-driven response differ from those of the LMS-driven response. This high degree of reproducibility suggests that tests for clinical differences in response will be well powered with a reasonable number of subjects.