A Neural Model of MST and MT Explains Perceived Object Motion during Self-Motion

Area of Science:

• Neuroscience
• Computational Neuroscience
• Visual Perception

Background:

• The human visual system must perceive object motion relative to the environment during self-motion.
• Self-motion significantly alters retinal motion, yet the visual system compensates to perceive world-relative object motion.

Purpose of the Study:

• To present a computational model explaining the neural mechanisms for recovering world-relative object motion during self-motion.
• To clarify the interaction between self-motion and object-motion perception pathways in the visual cortex.

Main Methods:

• Developed a computational model integrating self-motion (medial superior temporal area) and object-motion (middle temporal area) signals.
• Proposed two key mechanisms: MST-MT feedback and disinhibition of opponent motion signals in MT.

Main Results:

• The model explains how retinal motion is transformed into a world-relative reference frame.
• Demonstrates the interaction between global optic flow and local motion processing.
• Unifies existing 'flow parsing' hypotheses with neurophysiological evidence.

Conclusions:

• The proposed model offers explicit mechanisms for visual cortex processing of self-motion and object motion.
• Highlights the crucial role of feedback and disinhibition in accurate motion perception.
• Provides a framework for understanding how the brain achieves robust perception of object motion during self-motion.