An Efficient Simulation Environment for Modeling Large-Scale Cortical Processing

Micah Richert,Nikil Dutt,Jeffrey L Krichmar,Jayram Moorkanikara Nageswaran

doi:10.3389/fninf.2011.00019

Micah Richert, Nikil Dutt + Show 2 more

Open Access

https://doi.org/10.3389/fninf.2011.00019

Copy DOI

Abstract

We have developed a spiking neural network simulator, which is both easy to use and computationally efficient, for the generation of large-scale computational neuroscience models. The simulator implements current or conductance based Izhikevich neuron networks, having spike-timing dependent plasticity and short-term plasticity. It uses a standard network construction interface. The simulator allows for execution on either GPUs or CPUs. The simulator, which is written in C/C++, allows for both fine grain and coarse grain specificity of a host of parameters. We demonstrate the ease of use and computational efficiency of this model by implementing a large-scale model of cortical areas V1, V4, and area MT. The complete model, which has 138,240 neurons and approximately 30 million synapses, runs in real-time on an off-the-shelf GPU. The simulator source code, as well as the source code for the cortical model examples is publicly available.

Highlights

The mammalian nervous system is a network of extreme size and complexity (Sporns, 2011), and understanding the principles of brain processing by reverse engineering neural circuits and computational modeling is one of the biggest challenges of the Twenty-first century (Nageswaran et al, 2010), see (National Academy of Engineering-Grand Challenges for Engineering1)
RESULTS we provide several complete examples of the spiking neural networks (SNN) developed with our simulation environment
We demonstrate a complete Spiking Neural Network that demonstrates typical spike dynamics found in random networks having the appropriate balance of excitatory and inhibitory neurons, and spike-timing dependent plasticity (STDP)

Summary

Introduction

The mammalian nervous system is a network of extreme size and complexity (Sporns, 2011), and understanding the principles of brain processing by reverse engineering neural circuits and computational modeling is one of the biggest challenges of the Twenty-first century (Nageswaran et al, 2010), see (National Academy of Engineering-Grand Challenges for Engineering). There are several spiking simulators, which are currently available, that fall into different categories based on their level of abstraction and on the computer hardware in which they reside (for a recent review see Brette et al, 2007) Simulators, such as GENESIS and NEURON, incorporate molecular, detailed compartmental models of axons and dendrites from anatomical observations, and various ion channels to biophysical details (Hines and Carnevale, 1997, 2001; Bower and Beeman, 2007). A major goal of these models is to study detailed ionic channels and their influence on neuronal firing behavior While these models are biologically accurate, they incur tremendous computational costs for simulation.

Methods

Results

Conclusion