Adaptive intermittent control: A computational model explaining motor intermittency observed in human behavior.

Area of Science:

• Neuroscience
• Computational Neuroscience
• Control Theory

Background:

• The brain must perform real-time motor tasks despite slow sensorimotor systems.
• Feed-forward control models are effective for ballistic movements but less so for continuous tasks.
• Existing intermittent control models lack speculative segmentation based on prediction and uncertainty.

Purpose of the Study:

• To propose and evaluate a novel computational model for continuous motor tasks.
• To explain how feed-forward control can operate in real-time for tasks like target tracking.
• To investigate the role of speculative segmentation and uncertainty in motor control.

Main Methods:

• Development of an "adaptive intermittent control" or "segmented control" model.
• Computer simulations to test the model's performance in visuo-manual tracking.
• Analysis of computational costs and replication of human motor characteristics.

Main Results:

• The proposed model achieved faithful visuo-manual tracking with realistic sensorimotor delays.
• The model demonstrated reduced computational costs by using fewer segments.
• The simulation replicated "motor intermittency," a common feature of human movement.

Conclusions:

• Temporally segmented control is a likely brain strategy for efficient task execution in uncertain environments.
• Adaptive intermittent control offers a viable mechanism for real-time motor control with limited computational resources.
• Motor intermittency may be an inherent side-effect of this efficient control strategy.