Loading...

Topology and Network Pproperties of a Resulted Network from a Model for Brain Based on STDP Plasticity

Nowrouzi Nezhad, Hossein | 2024

0 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 57018 (04)
  4. University: Sharif University of Technology
  5. Department: Physics
  6. Advisor(s): Moghimi Araghi, Saman
  7. Abstract:
  8. The proposition of the critical nature of brain activities has been discussed for approximately two decades. This notion has been empirically observed through measurements of neuronal ensembles on brain slices of mice. Subsequently, numerous models have been introduced to rationalize and interpret this brain property. Additionally, it has been demonstrated that the neural system exhibits optimal performance at the critical point, both in computational and memory aspects. Therefore, the hypothesis of the critical brain is supported both empirically and theoretically. Typically, critical behavior occurs at phase transition points that require adjustment of external parameters such as temperature. However, in the case of the brain, no external parameter is explicitly adjusted, leading to the proposition that self-organization is intrinsic to this critical brain behavior. The most recognized model for self-organized criticality is the sandpile model of Bak-Tang-Wiesenfeld. Although this model exhibits acceptable similarities to neuronal population behavior, it also presents discernible differences. Consequently, generalized models have been developed to better approximate brain behavior. Another aspect regarding neuronal populations is the network type that connects different regions of the brain. One of the well-established networks for the human brain is called the connectome, wherein the brain is divided into approximately 1000 regions, and the connectivity strength between each pair of regions is obtained through Diffusion Tensor Imaging (DTI) measurements. Interestingly, the weight distribution of these connectivity strengths displays power-law behavior. In one of the models proposed in the physics department, both dynamic critical behavior and power-law behavior in the distribution of connectivity strengths have been observed. In this model, the dynamic network nodes resemble sandpile hills, and the network connections are dynamic, akin to SDTP dynamics in the brain. However, various aspects of the obtained network have not been explored. The question arises as to whether power-law behavior exists in connectivity strength but not in other network properties, prompting an investigation into the topology of the network in this model. In this research, we attempt to extend the mentioned model and derive the topology and properties of the obtained network from the Moghimi-Mahdisoltani model
  9. Keywords:
  10. Network Topology ; Self Organized Criticality ; Plasticity ; Critical Brain ; Critical Behavior ; Spike-Time-Dependent Plasticity (STDP)

 Digital Object List