Dynamic interaction between retinal and extraretinal signals in motion integration for smooth pursuit

Area of Science:

• Neuroscience
• Computational Neuroscience
• Vision Science

Background:

• The aperture problem causes initial tracking bias orthogonal to a translating bar's motion.
• Understanding retinal and extraretinal signal interactions is key to explaining action control.
• Smooth pursuit eye movements involve complex integration of visual and proprioceptive information.

Purpose of the Study:

• To investigate the dynamic interplay between retinal and extraretinal signals during smooth pursuit.
• To determine how target blanking affects visuomotor control at different pursuit phases.
• To develop and validate a computational model for motion integration in eye movements.

Main Methods:

• Two experiments involving brief target blanking (200-400 ms) during smooth pursuit of tilted bars.
• Experiment 1: Target blanking during steady-state pursuit.
• Experiment 2: Target blanking during the early phase of pursuit.
• Development of a closed-loop, two-stage recurrent Bayesian model.

Main Results:

• A marginal directional bias was observed in Experiment 1 (steady-state pursuit) under certain blanking conditions.
• No systematic directional bias was found in Experiment 2 (early pursuit phase) after target blanking.
• The proposed Bayesian model accurately reproduced observed smooth pursuit behaviors with a single parameter.

Conclusions:

• Retinal and extraretinal signal weighting is dynamically modulated throughout smooth pursuit.
• Human subjects rely more on prediction during early pursuit compared to steady-state pursuit.
• The model provides a framework for understanding adaptive action control via dynamic signal integration based on reliability.