A comparative analysis of multi-conductance neuronal models in silico

Area of Science:

• Neuroscience
• Computational Neuroscience
• Materials Science

Background:

• Flexible silicon-neuron architectures offer novel platforms for simulating neural dynamics.
• Understanding the parameter space of neuron models is crucial for accurate neural simulations.

Purpose of the Study:

• To demonstrate the capability of a flexible silicon-neuron architecture in implementing multiple conductance-based neuron models.
• To explore the parameter space and dynamics of these implemented models.
• To investigate the versatility of the silicon-neuron platform for multi-behavioral neural simulation.

Main Methods:

• Implementation of three disparate conductance-based neuron models on a flexible silicon-neuron architecture.
• Real-time mapping of model dynamics across a wide parameter space.
• Systematic variation of model parameters to identify parameter regions and trajectories.

Main Results:

• Successful implementation of neuron models with both fast and slow dynamics.
• Identification of a contiguous bursting space spanning between two distinct model dynamics.
• Discovery of diverse parameter trajectories connecting canonical bursting points.
• Confirmation that specific parameter combinations maintain the neuron within the bursting region.

Conclusions:

• The flexible silicon-neuron architecture is a valuable tool for neural modeling.
• The architecture demonstrates versatility as a platform for multi-behavioral neurons.
• The findings highlight the potential for simulating complex neural behaviors with physical implementations.