ETD

• astrocyte homeostasis
• computational neuroscience
• hybrid modelling
• local field potential
• tripartite synapse

This thesis investigates the role of astrocytes within the tripartite synapse, describing neuron, extracellular space, and astrocyte as a coupled non-linear dynamical system. This biophysical modelling captures how the astrocyte indirectly modulates neuronal excitability by regulating extracellular ionic and neurotransmitter concentrations. The primary aim of this study is to validate the Neuron-Extracellular Space-Astrocyte (NEA) model by accurately reproducing neuronal resting state and different firing patterns (single spike, burst regime, and depolarization block). Furthermore, this work evaluates the contribution of astrocytic homeostasis - specifically potassium buffering - to Local Field Potentials (LFPs). Using a hybrid modelling framework, synaptic currents extracted from single-compartment NEA model are mapped onto a multicompartmental neuronal morphology. Simulations demonstrate the NEA model’s robustness in replicating physiological dynamics following a rigorous initialization routine. Comparative analysis reveals that astrocytic potassium buffering affects LFP dynamics primarily over long temporal scales (40 s) and under sustained high-frequency multisynaptic (100) stimulation. These findings support the physiological plausibility of the proposed NEA framework and its flexibility in capturing different dynamical regimes. Scaling the framework to neuronal populations could support the study of pathological conditions characterized by impaired astrocytic homeostasis, such as Alzheimer’s disease.