The role of contrast sensitivity in the development of binocular vision: A computational study

摘要

The development of accurate binocular vision relies on the acquisition of disparity tuning and the calibration of vergence eye movements. Both processes are fundamentally limited by visual acuity, which increases only gradually during the first year of life. Next to limiting performance, however, early limitations of visual acuity may also aid rapid learning analogous to Newport's “less-is-more” hypothesis. Here we use computational modeling to assess the potential impact of early acuity limitations on the development of binocular vision. Our starting point is a previous model of the development of binocular vision, formulated in the Active Efficient Coding framework. We extend this model to incorporate the development of visual acuity between birth and 8 months. We find that the model does in fact learn faster if visual acuity starts out poor and increases with time, supporting the less-is-more hypothesis. Furthermore, we find that the speed of acuity improvement needs to be “just right”, i.e., neither too rapid nor too slow for fastest learning. Overall, our model suggests that early limitations of visual acuity may aid infants in acquiring binocular vision skills and it provides a good starting point for computational modeling of developmental disorders of binocular vision.