Abstract
Purpose:
Melanopsin-expressing retinal ganglion cells mediate various non-image-forming (NIF) visual functions such as the pupillary light reflex (PLR), circadian entrainment, and suppression of pineal melatonin release. Inadequate daytime NIF photostimulation can cause depression, sleep disorders and cognitive impairment, which can be treated with phototherapy. In an attempt to increase the efficacy of phototherapy, significant work has been done to identify wavelengths optimal for NIF vision. Here, we seek to enhance NIF visual stimulation using flickering light instead of steady light.
Methods:
Five subjects were studied. An infrared pupillometer was used to present 463 nm light to the right eye and image the consensual PLR from the left eye. Each subject was tested with a three-dimensional matrix of 63 flickering stimuli: 3 total photon counts (13.8, 14.8 and 15.8 log photons cm-2), 3 duty cycles (12%, 47% and 93%) and 7 flicker frequencies (0.1, 0.25, 0.5, 1, 2, 4 and 7 Hz). Each stimulus was presented for 4 min and steady-state pupil constriction measured during the final min. For comparison, steady lights with equivalent intensities or total photons were also tested.
Results:
Among stimuli delivering a total of 13.8 log photons cm-2, the one flickering at 2 Hz with a 12% duty cycle evoked the greatest PLR, constricting the pupil by 48%. This constriction was ~70% greater than that evoked by an equal-intensity (12.4 log photons cm-2 s-1) light presented continuously for 4 min. For stimuli containing 14.8 log photons cm-2, the most potent was again the 2 Hz light with a 12% duty cycle, inducing a 58% constriction. This was ~40% more than that caused by a 4-min continuous light of an equal intensity (13.4 log photons cm-2 s-1). For 15.8 log photons cm-2 stimuli, the 1Hz light with a 47% duty cycle was the best although the constriction was the same as that evoked by the best 14.8 log photons cm-2 light. While this constriction was ~20% greater than that caused by a 4-min continuous light totaling 15.8 log photons cm-2, it was statistically indistinguishable from that evoked by a continuous light of an equal intensity (13.8 log photons cm-2 s-1).
Conclusions:
This is the first demonstration that the magnitude of the PLR is affected by the duty cycle of the light. Under our recording conditions, a 13.4 log photons cm-2 s-1 stimulus flickering at 2 Hz with a 12% duty cycle stimulates the NIF visual system most efficiently.