Effects of the local dynamics in the synchronization of neural models
Visualizar/ Abrir
Data
2022Autor
Boaretto, Bruno Rafael Reichert, 1995-
Metadata
Mostrar registro completoResumo
Resumo: O comportamento cooperativo de neurônios e áreas neuronais associadas ao comportamento de sincronização se apresenta como mecanismo fundamental para o funcionamento cerebral. Além disso, níveis anormais de sincronização têm sido relacionados a estados patológicos. Ao longo desta tese, abordam-se diferentes fenômenos de sincronização que surgem por meio da dinâmica coletiva de modelos de neurônios acoplados em um a rede. Primeiramente, mostra-se um a forte correlação entre a dinâmica individual do neurônio com o comportamento global da sincronização da rede, em que a periodicidade observada no neurônio isolado é refletida em um a sincronização de fase ao considerar um acoplamento fraco. Em segundo lugar, estuda-se o papel da biestabilidade na sincronização de um a rede de neurônios idênticos, acoplados através de um esquema de campo médio. Mostra-se que a simples existência de dois estados estáveis distintos pode levar a rede a diferentes estados de sincronização, dependendo da inicialização do sistema. Por fim, é investigado o mecanismo de sincronização explosiva de um a rede neural complexa composta por neurônios não-idênticos. A presença deste regime é acompanhada por um loop de histerese na dinâmica da rede, à medida que o parâmetro de acoplamento é adiabaticamente aumentado e reduzido. Demonstra-se que as transições de sincronização abruptas estão associadas a rotas para o caos e que os mecanismos dinâmicos para a região de biestabilidade são dados em termos de um a bifurcação de sela-nó e um a crise de fronteira. Portanto, os resultados desta tese mostram uma riqueza de comportamentos de sincronização associados a pequenas mudanças na dinâmica neuronal, trazendo novos insights para o estudo teórico das redes neurais. Abstract: The cooperative behavior of neurons and neuronal areas associated with the synchronization behavior proves to be a fundamental neural mechanism. In addition, abnormal levels of synchronization have been related to unhealthy neural behaviors. Throughout this thesis, it is explored different synchronization phenomena which emerge through the collective dynamics of models of neurons coupled in a network. Firstly, it is shown a strong correlation between the individual dynamics of the neuron with the global behavior of the synchronization, in which the periodicity seen in the isolated neuron is reflected in a phase synchronization in the weak coupling region. Secondly, it is studied the role of bistability in the synchronization of a network of identical neurons coupled through a mean-field scheme. It is shown that the simple existence of two distinct stable states can lead the network to different states of synchronization, depending on the initialization of the system. Lastly, it is investigated the mechanism for explosive synchronization of a complex neural network composed of non-identical neurons. The presence of this regime is accompanied by a hysteresis loop on the network dynamics as the coupling parameter is adiabatically increased and decreased. It is shown that the abrupt synchronization transitions are associated with routes to chaos. The dynamical mechanisms for the bistability region, are given in terms of a saddle-node bifurcation and a boundary crisis. Therefore, the results of this thesis show a richness of synchronization behaviors associated with small changes of the neuronal dynamics bringing new insights to the theoretical study of neural networks.
Collections
- Teses [122]