Optimization behavior of brainstem respiratory neurons. A cerebral neural network model

Area of Science:

• Physiology
• Neuroscience
• Computational Biology

Background:

• Optimal control laws may govern respiratory responses to CO2 and exercise.
• The biological feasibility of such complex neural optimization was uncertain.

Purpose of the Study:

• To test if a realistic neural network can achieve optimal respiratory control.
• To investigate the role of neural optimization in respiratory regulation during simulated CO2 inhalation and exercise.

Main Methods:

• Developed a hybrid computer-neural model simulating lung and tissue dynamics.
• Used a human subject as a controller, pedaling a bicycle to minimize breathing cost.
• Employed the visuomotor cortex as a proxy for brainstem respiratory neurons.

Main Results:

• Observed a linear steady-state ventilatory CO2 response to arterial PCO2.
• Found a nearly isocapnic steady-state response during simulated exercise.
• Demonstrated that neural optimization is a plausible mechanism for respiratory control.

Conclusions:

• Neural optimization can be achieved by neural networks with cognitive abilities.
• This mechanism can explain respiratory control during exercise without requiring an explicit exercise stimulus.