Thesis (M.D.)--Boston University

Maintaining the intrinsic excitability of neurons is crucial for stable brain activity. This can be achieved by the homeostatic regulation of membrane ion channel conductances, although it is not well understood how these processes influence broader aspects of neuron and network function. One of the many mechanisms which contribute towards this task is the modulation of potassium channel conductances by activity-dependent nitric oxide signalling. Here, we first investigate this mechanism in a conductance-based neuron model. By fitting the model to experimental data we find that nitric oxide signalling improves synaptic transmission fidelity at high firing rates, but that there is an increase in the metabolic cost of action potentials associated with this improvement. Although the improvement in function had been observed previously in experiment, the metabolic constraint was unknown. This additional constraint provides a plausible explanation for the selective activation of nitric oxide signalling only at high firing rates. In addition to mediating homeostatic control of intrinsic excitability, nitric oxide can diffuse freely across cell membranes, providing a unique mechanism for neurons to communicate within a network, independent of synaptic connectivity. We next conduct a theoretical investigation of the distinguishing roles of diffusive homeostasis mediated by nitric oxide in comparison with canonical non-diffusive homeostasis in cortical networks. We find that both forms of homeostasis robustly maintain stable activity. However, the resulting networks differ, with diffusive homeostasis maintaining substantial heterogeneity in activity levels of individual neurons, a feature disrupted in networks with non-diffusive homeostasis. This results in networks capable of representing input heterogeneity, and linearly responding over a broader range of inputs than those undergoing non-diffusive homeostasis. We further show that diffusive homeostasis interferes less than non-diffusive homeostasis in the synaptic weight dynamics of networks undergoing Hebbian plasticity. Overall, these results suggest a novel homeostatic mechanism for maintaining stable network activity while simultaneously minimising metabolic cost and conserving network functionality.

Made available in DSpace on 2017-07-10T17:11:46Z (GMT). No. of bitstreams: 1 DISSERTACAO ARTUR CARRIJO2.pdf: 2100488 bytes, checksum: 46e631e94e2bb6d4c69d8d2891f6e200 (MD5) Previous issue date: 2013-12-18 / The electricity market has many participants performing various roles. In the context of Smart Grids, it expands the number of participating agents and, consequently, the number, the competition and complexity of the interactions between them. The competition allows consumersto discover and explore energy sources of low cost but requires the consumer other interactions beyond the traditional relationship with the distribution company (distco). With the redefinition of the role of the consumer as a result of his active participation in the balance between supply and demand of energy, it becomes necessary to identify their interactions with the other participants, tasks, systems, subsystems and functions important to actively contribute to this balance. These interactions, called interfaces, are characterized by information that actors communicate itself to perform its functions of information collection and control tasks related to market equilibrium. The determination of these interfaces is not a trivial task because of the various alternatives for integration of the demand side. Different types and combinations of interactions between the supplier and its customers are possible. In this work will be used as a basis for discussion a concept of balance between generation and demand called homeostatic control, initially designed for a scenario of a vertically integrated monopoly. It will be studied the interactions of the consumer using the concept of homeostatic control expanded to consider the functions of demand control and integration of distributed generation and a market model that allows for consumer empowerment and within the paradigm of smart grids. / O mercado de energia elétrica possui diversos participantes exercendo vários papéis. No contexto das redes elétricas inteligentes, amplia-se o número de agentes participantes e, consequentemente, o número, a competição e a complexidade das interações entre eles. A competição permite aos consumidores descobrir e explorar fontes de baixo custo, mas requer do consumidor outras interações além da tradicional relação com a empresa distribuidora. Com a redefinição do papel do consumidor, resultado da participação ativa no processo de equilíbrio entre suprimento e demanda de energia, torna-se necessário identificar suas interações com os demais participantes, tarefas, sistemas, subsistemas e funções, importantes para que contribua ativamente com esse equilíbrio. Estas interações, denominadas interfaces, são caracterizadas pelas informações que os atores comunicam em si para executar suas funções de coleta de informações e tarefas de controle relacionadas ao equilíbrio do mercado. A determinação destas interfaces não é uma tarefa trivial em razão das diversas alternativas de integração do lado da demanda. São possíveis diferentes tipos e combinações de interações entre o fornecedor e seus clientes. Neste trabalho será utilizado como base de discussão um conceito de equilíbrio entre geração e demanda denominado controle homeostático, desenhado inicialmente para um cenário de monopólio verticalmente integrado. Serão estudadas as interações do consumidor utilizando o conceito de controle homeostático ampliado para considerar as funções de controle da demanda e a integração da geração distribuída em um modelo de mercado que permite o empoderamento do consumidor dentro do paradigma das redes elétricas inteligentes.