Loading...

Synchronization in Inhibitory Neural Networks

Mehrani Ardebili, Mohsen | 2022

112 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 55526 (04)
  4. University: Sharif University of Technology
  5. Department: Physics
  6. Advisor(s): Moghimi Araghi, Saman
  7. Abstract:
  8. Centuries passed and the human knew himself as the protagonist who searches around nature and discovers the phenomena. But after the birth of ``neuroscience", his wisdom and the process of reasoning were also added to the list of uncovered subjects. Since its arrival, many scientists started investigating ``reasoning", "sleep", ``memory disorders" etc. with a such framework. One of the main branches of this stream is the ``Synchronization" problem when the neurons get synced in the matter of spiking likelihood. ``Synchronization" means a lot to the community, because it is said that it is one major symptom of Epilepsy. With that said, we need to get to the root of this effect. It seems that the brain lies in some critical state. One promising option is the transition point between synchronization and desynchronization. Recognizing the in-between state demands a well understanding of extreme states. Numerous models have been suggested to describe this matter. However, we discuss one promising model with numerical and analytical details. In this model, we study the networks with neurons of heterogeneous inputs. Of course, there are many options to put for of dynamics of each neuron.However, we found that regardless of the taken equation of dynamic the phase-transition always holds, even at very simple ones. Moreover, we indicated that the synchronization happens to their velocity rather than their phases, better say ``velocity-lock" rather than ``phase-lock". This means that when they intend to spike, they head toward the threshold simultaneously. Numerical investigation confirms our hypothesis
  9. Keywords:
  10. Neuron ; Computational Neuro Science ; Epilepsy ; Synchronisation ; Numerical Simulation ; Neural Networks

 Digital Object List

 Bookmark

...see more