Abstract
Purpose :
To examine how the transient and sustained components of the pupil light reflex (PLR) vary with light level and spectral composition. In particular, we examined whether the pupil size following exposure to rod-enhanced stimuli of equal photopic luminance is linked to interaction of rod and cone signals specific to the mesopic range.
Methods :
Pupil responses to increments in light flux were measured binocularly using the P-Scan system (Proc Natl Acad Sci U S A, 96, 1999: 11637-41). Six subjects with normal colour vision participated in this study. A 30o x 24o stimulus was centred on a uniform background field subtending 32o x 26o. Careful control of screen luminance in combination with spectrally-calibrated, neutral density filters made it possible to investigate 12 discrete luminances in the range 0.001 to 9 cd/m2. For each luminance, pairs of stimuli with scotopic to photopic ratios of 0.7 (cone-dominated response) and 8 (rod-enhanced) were employed. Retinal illuminance (RI) levels were calculated for every stimulus with appropriate pupil apodization. The 3s stimulus had twice the background luminance and each recording lasted for 10s. 32 traces were averaged for each condition.
Results :
The rod-enhanced stimulus results in smaller pupils, but the difference between the two stimuli remains small. The largest transient constriction amplitude for the dynamic PLR corresponds to a retinal illuminance of 0 to 1 log td. The sustained constriction of the pupil that follows the dynamic, transient response at flash onset increases linearly with log RI and peaks around 0.5 log td. Beyond this point, the effect is reversed with a linear, more rapid decrease. The magnitude of the post-stimulus constriction (measured 6s after stimulus offset) is also largest in amplitude for light levels within the low mesopic range.
Conclusions :
The sustained constriction of the pupil (2s after stimulus onset) is smallest for light levels at the centre of the mesopic range. Rod and cone signals add to drive the sustained PLR, but the efficiency of this addition varies with RI as a result of rod-cone interactions. The largest post-stimulus constriction (6s after stimulus offset) corresponds to ~ 0 log td, a value well within the mesopic range. This response is similar to the well-established PIPR that requires much higher RI levels. These findings highlight the potential usefulness of pupillometric techniques to study the interaction of rod and cone signals.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.