Computational model of the motor program generator for pursuit.

Area of Science:

• Neuroscience
• Computational Neuroscience
• Motor Control

Background:

• Pursuit eye movements (PEM) are crucial for visual tracking.
• Understanding the neural basis of motor program generation is complex.
• Non-periodic motor programs require flexible control mechanisms.

Purpose of the Study:

• To develop a parallel processing neural network model for motor program generation (MPG) of PEM.
• To investigate the dynamics of neural activity peaks within the model.
• To explore how the model accounts for different movement trajectories.

Main Methods:

• Developed a parallel processing neural network model comprising two interconnected velocity maps (theta R and theta L).
• Modeled neurons arranged in a circular layer with local connectivity.
• Simulated a potential field analogous to a flexible membrane to represent neural activity.
• Analyzed the behavior of an activity peak (AP) traveling across the neural maps.

Main Results:

• The model successfully generated pursuit eye movements (PEM) using parallel processing.
• An activity peak (AP) demonstrated constant velocity (vT) movement across neural maps.
• The shape of the potential field determined the trajectory of the AP (circular, peripheral, or central).

Conclusions:

• The developed MPG model provides a novel computational framework for understanding PEM.
• The model's mechanism offers insights into the neural generation of non-periodic motor programs.
• This approach advances the study of complex eye movement control systems.