Eye-specific pattern-motion signals support the perception of three-dimensional motion.

Area of Science:

• Visual neuroscience
• Computational vision
• Perception psychology

Background:

• Three-dimensional (3D) motion perception is crucial for navigating complex environments.
• Interocular velocity differences are a key cue for 3D motion, but their computation requires specific processing of monocular motion signals.
• Previous models suggested 3D motion relies on global pattern motion, necessitating large-scale pooling of 2D motion information.

Purpose of the Study:

• To investigate whether 3D motion processing involves global spatial pooling of eye-specific 2D motion signals.
• To determine if 3D motion computations retain information about the eye of origin.
• To test the role of global pattern motion in 3D motion perception using motion aftereffects (MAEs).

Main Methods:

• Utilized 3D motion aftereffects (MAEs) with stimuli composed of numerous small, drifting Gabor elements.
• Compared local, frontoparallel (2D) MAEs with 3D MAEs under varying conditions of element location and orientation.
• Manipulated element retinal locations and interocular matching during adaptation to probe spatial pooling and eye-specificity.

Main Results:

• Conventional 2D MAEs were local and sensitive to element position.
• 3D MAEs demonstrated robustness to changes in element location, indicating global spatial pooling of motion signals.
• Strong 3D MAEs were observed even with unmatched element locations across eyes and random orientations, supporting global motion specification via intersection of constraints.

Conclusions:

• 3D motion processing involves eye-specific computation of 2D pattern motion, utilizing global pooling of local, eye-specific motion signals.
• The findings support the idea that visual computations traditionally considered "late" can preserve eye-of-origin information.
• This research advances our understanding of how the brain integrates visual information from both eyes to perceive 3D motion in dynamic environments.