Theoretical and evolutionary parameter tuning of neural oscillators with a double-chain structure for generating...

Area of Science:

• Computational Neuroscience
• Biophysics
• Systems Biology

Background:

• Neural oscillators, particularly double-chain central pattern generators, model rhythmic movements in organisms.
• Tuning the numerous parameters of these oscillators to achieve specific rhythmic outputs is challenging.

Purpose of the Study:

• To develop an automated method for tuning neural oscillator parameters.
• To enable the accurate generation of desired rhythmic signals and wave patterns.

Main Methods:

• A two-part tuning strategy combining theoretical analysis and evolutionary algorithms.
• Part 1: Theoretical tuning rules for time constants and output amplitudes.
• Part 2: A two-step genetic algorithm (global and local GA) for connection weights.

Main Results:

• The proposed method successfully tuned all parameters to generate sinusoidal waves.
• Tuning performance remained robust despite variations in oscillator number or desired output.
• The method efficiently reproduced complex rhythmic signals in artificial and biological systems, including Caenorhabditis elegans.

Conclusions:

• The developed automatic tuning method effectively addresses the complexity of neural oscillator parameterization.
• This approach facilitates the simulation and understanding of rhythmic biological processes and wave generation.