Construction, characterization and simulation of a cortical neural network model
| Grant number: | 17/05874-0 |
| Support type: | Scholarships in Brazil - Master |
| Effective date (Start): | July 01, 2017 |
| Effective date (End): | August 31, 2019 |
| Field of knowledge: | Engineering - Biomedical Engineering |
| Cooperation agreement: | Coordination of Improvement of Higher Education Personnel (CAPES) |
| Principal Investigator: | Antonio Carlos Roque da Silva Filho |
| Grantee: | Vinícius Lima Cordeiro |
| Home Institution: | Faculdade de Filosofia, Ciências e Letras de Ribeirão Preto (FFCLRP). Universidade de São Paulo (USP). Ribeirão Preto , SP, Brazil |
| Associated research grant: | 13/07699-0 - Research, Innovation and Dissemination Center for Neuromathematics - NeuroMat, AP.CEPID |
| Associated scholarship(s): | 18/07673-4 - Information dynamics during working memory tasks, BE.EP.MS |
Abstract The objective of this project is to construct stochastic models of cortical neural networks with three different architectures: a random network with Erdos-Rényi topology, a random network with hierarchical and modular topology and a random network with layered topology and determined data connectivity to the local microcircuit of the cortex. Networks with these three architectures using deterministic models of neurons have been extensively studied in the literature of computational neuroscience and the goal here will be to build versions of these networks with stochastic models of neurons for comparison. The stochastic model of neuron to be used is the one recently proposed by Galves and Löcherbach (J. Stat. Phys. 151: 896-921, 2013) in the simplified version considered by Brochini et al. (Sci. Rep. 6: 35831, 2016). The motivation for the use of stochastic models of neurons is the experimental observation that neurons of the brain present variability in their responses to repetitions of the same stimulus. The hypothesis adopted is that the stochastic behavior of the neurons has influence on the dynamics of the neural network, and should not be ignored in simulations. For the comparison with the stochastic models to be constructed, published versions of deterministic models with the three architectures will be considered as reference standards. The deterministic models are composed of excitatory and inhibitory neurons with parameters adjusted so that the networks are in a balanced state between excitation and inhibition presenting asynchronous and irregular spontaneous activity. The stochastic models to be constructed will also have their parameters adjusted so that they exhibit asynchronous and irregular spontaneous activity similar to the deterministic versions. The comparison between the deterministic and stochastic models will be made by statistical measures commonly used to characterize trains of neuronal firings. (AU) | |