Abstract
Purpose :
Several perceptual learning approaches have been used to train the visual system for stereopsis: mostly in the lab, using videogames and high-tech equipment such as head-mounted displays. Here, we wanted to develop a very simple system that participants could use at home on a tablet: Participants would just repeat over, at home, a same stereo task with which they would have been previously tested in the lab: the 4-wedges random dots test.
Methods :
5 amblyopes and 5 controls subject with normal stereopsis or a mild stereovision deficiency (≥ 70 arcsec, randot) participated in the study. Their stereo-sensitivity at baseline was precisely measured using 2 repetitions of the 4-wedges random dots test without feedback. One training session consisted of repeating the same task for 10 repetitions with feedback and lasted 30-45 min per day. Participants underwent 7 training sessions: the first one in the lab and then 6 at-home. Finally, their stereo-sensitivity was measured back in the lab, after the training program. All tests and training were run on the same tablet using anaglyph glasses. The equipment was then lent to participants to train at-home.
Results :
We observed an improvement in stereopsis in most participants (average 30 arcsec). None of them reported any difficulty in using or understanding the test and none of them reported any fatigue due to use of the test or the glasses. Compliance was obtained from the log files and was very good for all participants. All of them achieved the training in 7 or 8 days. Only two skipped one day and only one split the training session in 2 halves.
Conclusions :
Amblyopes as well as people with mild stereovision deficiencies can be trained with a simple and convenient at-home protocol and show improvement after only 7 training sessions of 30 to 45 minutes. The stimuli used in the test, consisting of filtered random dots, will allow us to assess the specificity of the training in different spatial frequency domains in the future.
This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.