Purpose: Inspired by the finding in chickens that preferential stimulation of the ON retinal system suppresses myopia induced by negative spectacle lens wear and that stimulation of the OFF system suppresses the hyperopia induced by positive lens wear, we sought to determine whether stimulation of the ON-OFF retinal systems could drive directional accommodation responses in humans. If emmetropisation and accommodation use similar image processing algorithms, more accommodation would be expected with OFF stimulation.
Methods: Accommodation responses were measured while viewing a computer-generated pattern designed to stimulate the ON-OFF systems. The stimulus comprised a rectangular field (12 x 9.5 cm) on a black background filled with 196 discs (diameters: 0.4-1.0 cm). These were presented on an LCD monitor in a dark room at a viewing distance of 55 cm (1.8 D). Thirteen subjects aged 21-37 years took part. The individual discs had saw-tooth shaped temporal luminance profiles with the same time period but with random phases with respect to each other, so that the mean brightness of the stimulus was constant. To eliminate accommodation responses based on other cues (i.e. proximity) a 0.5 mm artificial pupil was used to open the accommodation loop. Refraction in the vertical pupil meridian was continuously recorded with an infrared photorefractor (the PowerRefractor). To verify that computer-based stimuli presented within our experimental design were effective in driving accommodation, previously studied stimuli were also tested: changes in size (looming) and incremental low pass filtering.
Results: Preferential stimulation of the ON or OFF subsystems produced a convincing depth illusion in all subjects (which was psychophysically confirmed in four subjects). Although the stimulus appeared to move in depth it did not produce accommodation responses that were consistent with that, i.e. the accommodation system did not appear to fluctuate in rhythm with the temporal oscillations of the stimulus. As the target appeared to loom it induced a greater accommodation response then when it appeared to recede. The looming target produced changes in the accommodation response in nine of 13 subjects that were consistent with its perceived change in proximity (although the target did not actually move in depth). Incremental low pass filtering produced non-directional drifts of accommodation in all subjects. Combinations of the stimuli (i.e. looming and low pass filtering, ON/OFF and looming) were not more effective stimuli to accommodation. After removal of the artificial pupil (closed loop conditions), accommodation was no longer induced with any of these stimuli.
Conclusions: Although the preferential ON or OFF stimulation produced a pronounced illusion of motion in depth despite constant average brightness, proximal accommodation was induced in only one subject. Therefore, the ON/OFF stimulation appeared to have only minor input into proximal accommodation. Potential inputs into reflex accommodation need to be defined in further studies.