A FPGA real-time model of single and multiple visual cortex neurons

Area of Science:

• Neuroscience
• Computational Neuroscience
• Biophysics

Background:

• Biologically realistic neuron models are valuable for electrophysiology research and education.
• Simulating multi-electrode recordings requires models that can generate distinct spike waveforms.

Purpose of the Study:

• To develop an affordable and customizable Field Programmable Gate Array (FPGA)-based spiking neuron model.
• To simulate realistic multi-electrode (tetrode) recordings of visual cortex neurons.

Main Methods:

• Implemented a Field Programmable Gate Array (FPGA) system to simulate three independent neurons.
• Generated distinct analog spike waveform signals for each neuron across four output channels.
• Designed the model using inexpensive, commercially available components for ease of construction and operation.

Main Results:

• The model successfully simulated biologically realistic simple-cell-like response properties.
• Demonstrated orientation selectivity, spatial/temporal frequency selectivity, and space-time receptive fields.
• Achieved highly modulated Poisson spike trains in response to stimuli.

Conclusions:

• The FPGA-based spiking neuron model offers a cost-effective and flexible tool for visual neurophysiology research and education.
• The model's customizable nature allows for adaptation to specific research needs and exploration of different neuronal models.
• Open-source code and documentation facilitate user modification and extension of the model's functionality.