Computation of reduced energy input current stimuli for neuron phase models

Area of Science:

• Computational Neuroscience
• Mathematical Biology
• Neural Modeling

Background:

• Regularly spiking neurons are often modeled using phase models.
• Phase response curves (PRCs) describe how input stimuli affect a neuron's phase.
• Optimizing input stimulus currents is crucial for understanding and controlling neural activity.

Purpose of the Study:

• To adapt a technique for discovering optimal input stimulus currents for phase models.
• To minimize the 'energy' of the stimulus current while ensuring accurate phase response tracking.
• To investigate the characteristics of these optimal currents in different neuron phase models.

Main Methods:

• Computed neuron phase response θ(t) to an input stimulus current i(t) using a phase model.
• Defined the computed phase response as a reference phase r(t).
• Calculated an optimal input stimulus current i(*)(t) by minimizing a weighted sum of stimulus energy and tracking error between r(t) and the response to i(*)(t).

Main Results:

• The adapted technique successfully generated optimal currents i(*)(t) for two neuron phase models.
• The generated i(*)(t) produced phase responses similar to the reference phase r(t).
• Optimal currents i(*)(t) exhibited lower 'energy' compared to the original input currents i(t).
• Optimal currents i(*)(t) were not necessarily constant, being larger in regions of higher PRC sensitivity.

Conclusions:

• The developed method effectively identifies energy-efficient optimal stimulus currents for phase neuron models.
• The optimal currents dynamically adjust based on the neuron's phase sensitivity.
• This approach offers a valuable tool for precise control and analysis of neural dynamics in simplified models.