A comparative approach to closed-loop computation.

Area of Science:

• Neuroscience
• Computational Neuroscience
• Systems Neuroscience

Background:

• Neural computation is inherently a closed-loop system, involving continuous interaction between sensory processing and motor output.
• Technological advancements allow for the manipulation of neural feedback loops in experimental settings.
• Understanding the flow of information (topology) between neural subsystems is key to deciphering behavior.

Purpose of the Study:

• To explore how manipulating neural feedback loop topology impacts behavior.
• To provide a computational framework for analyzing experiments across a spectrum of feedback loop conditions.
• To bridge the understanding between low-level neural mechanisms and high-level functions in closed-loop systems.

Main Methods:

• Utilizing control theory and system identification as a computational framework.
• Conducting experiments across a spectrum from open-loop (restrained) to fully closed-loop (free behavior) preparations.
• Translating experimental findings between different levels of the feedback loop spectrum.

Main Results:

• Demonstrated the feasibility of altering neural feedback loop topology in various preparations.
• Established a framework for quantitatively relating findings from different experimental approaches.
• Showcased how operating across the spectrum of feedback manipulation yields novel insights.

Conclusions:

• Manipulating neural feedback loop topology is a powerful tool for understanding neural computation and behavior.
• A unified computational framework is essential for integrating findings across different experimental paradigms.
• Bridging experimental approaches across the open- to closed-loop spectrum enhances understanding of neural mechanisms and function.