Motion parallax links visual motion areas and scene regions

Area of Science:

• Neuroscience
• Computational Neuroscience
• Visual Perception

Background:

• Motion parallax is a crucial depth cue derived from relative object velocities during self-motion.
• The neural basis for integrating motion parallax with scene processing remains poorly understood.
• Existing research has not fully explored the brain networks supporting motion parallax perception.

Purpose of the Study:

• To investigate the neural substrates underlying motion parallax processing using functional magnetic resonance imaging (fMRI).
• To identify brain regions selectively responding to motion parallax stimuli.
• To explore functional connectivity between motion-sensitive and scene-selective brain areas during parallax perception.

Main Methods:

• fMRI was employed with human observers viewing a controlled, depth-layered scene stimulus simulating lateral self-motion.
• Parallax-selective responses were analyzed in predefined and anatomically defined regions of interest.
• Functional connectivity analysis was performed during the processing of motion parallax.

Main Results:

• Selective parallax responses were observed in parietal regions IPS3 and IPS4, and a novel occipital region adjacent to the Occipital Place Area (OPA).
• Traditionally defined scene-selective regions (OPA, Para-hippocampal Place Area (PPA), Retrosplenial Cortex (RSC)) did not show parallax-selective responses.
• The occipital parallax-selective region exhibited significant functional connectivity with IPS3 and the scene-selective PPA during parallax processing.

Conclusions:

• A distinct neural network linking dorsal motion-processing regions (IPS3/IPS4) and ventral scene-processing regions (PPA) is specifically engaged during motion parallax.
• This network likely plays a key role in the brain's ability to infer 3D scene structure from motion parallax.
• The findings reveal novel brain regions and connections crucial for deriving spatial information from visual motion cues.