NYU Shanghai Professor of Neural Science and Psychology Li Li has published a review article, “Visual Perception of Self-Motion,” in the 2025 volume of the Annual Review of Vision Science (Vol. 11, pp. 447–474). Synthesizing decades of research into how the brain interprets visual information to perceive self-motion, Professor Li proposes new models for understanding the visual system and identifies key areas for applying this research to advance technology in virtual reality, autonomous navigation, and sensory rehabilitation.
From simple tasks like walking through a city park, to complicated high-speed navigation like driving or flying, our visual perception of self-motion—how we know we are moving and where we are moving t—is essential. At the core of this process is optic flow, the shifting visual pattern produced as we move through our environment, that helps to convey information about direction and speed of our self-motion as well as the three-dimensional layout of the world.
Professor Li’s review demonstrates that in visual perception of self-motion, the brain flexibly combines multiple sources of visual information, (such as optic flow, target drift, and perspective change) and integrates them with non-visual cues generated during self-motion (such as vestibular and proprioceptive information about body and limb movements). Flow parsing, the disentanglement of retinal motion caused by self-motion for the perception of independent object motion during self-motion, also highlights the integration of both visual and nonvisual cues.
Bridging interdisciplinary evidence from behavioral, computational, and neurophysiological studies, this review article provides a comprehensive roadmap for refining future research, including identifying the brain areas responsive to real rather than fake optic flow, investigating the neural basis of path perception, and incorporating additional cortical areas into artificial neural networks modeling self-motion, all of which have far-reaching potential for real-world application in emergent fields.