Purpose: : Eyes that are emmetropic in daylight are thought to become myopic at low luminance conditions. This phenomenon, commonly referred to as night myopia was extensively studied during the mid XX century. However, as it is pointed out in a classical study (Koomen et al., JOSA, 1951): “night myopia has been rediscovered from time to time and the literature is full of conflicting results”. The purpose of this work is to revisit this phenomenon using an adaptive optics instrument operating in invisible infrared illumination to elucidate its actual magnitude and main causes.
Methods: : Best focus was determined subjectively for a Maltese cross stimulus illuminated with a broadband Halogen lamp in 9 subjects with normal vision. The luminance of the target was modified with neutral density filters from the baseline condition of 20 cd/m2 to the lower luminance case of 0.00002 cd/m2. Density filters of 1, 4, 5.5, 6, 6.5 and 7 were used. The low luminance measurements were performed after subjects dark adapted for at least 30 minutes. Defocus and aberrations were continuously measured with the 1050-nm Hartmann-Shack sensor incorporated in the adaptive optics instrument. Measurements were performed with natural pupils with polychromatic stimuli, either with natural aberrations or with the aberrations corrected.
Results: : Differences in the best focus position for the highest and the lowest luminance conditions were in average around -0.8 D, although with a large individual variability (SD = 0.8 D). The impact of the chromatic aberration combined with the Purkinje shift was estimated to be less than 0.2 D. We measured an accommodation shift at low luminance that accounted for most of the average myopic shift for low luminance. The correction of high order aberrations nearly doubled visual acuity at the low luminance conditions, but did not modify the average amount of myopic defocus.
Conclusions: : We have carried out innovative experiments using an adaptive optics instrument with invisible measuring light to re-evaluate the phenomenon of night myopia. We found large individual variability and an average modest (less than 1 D) myopic shift for low light conditions (0.00002 cd/m2). The main underlying reason for night myopia is the accommodation shift occurring for low light levels. Chromatic and high order monochromatic aberrations have a minor effect in this mechanism.