NEURAL DYNAMICS AND INFORMATION TRANSMISSION: from cortical circuits to the hippocampal modelling

In this doctoral thesis, we dive into the field of computational neuroscience, ex- ploring various aspects of neural network modeling. Our work unfolds in three interconnected parts, each contributing to a comprehensive understanding of in- formation processing in the brain. In the first part, we investigate the complexities of information transmission within cortical circuits. The focus revolves around unraveling the essential conditions for effective communication between intercon- nected cortical populations mediated by a potential third element, the thalamus. We find two operation modes dynamically switchable by modulating the strength of the direct cortico-cortical connection.
In the second part, we explore the modeling of data-driven whole-brain networks with spiking neurons. Here, we aim to harness the power of computational models to investigate potential therapies based on transcranial current stimulation (tACS) patients facing various conditions. We emphasize the use of spiking models for the design of such networks, where traditionally simpler and more abstract models such as neural mass models have been considered. We have developed a methodology to optimize these networks, subsequently applying tACS to increase the alpha band power on personalized networks based on patient data. This experiment represents a pivotal step towards a proposed clinical treatment.
Finally, we model the hippocampal region using multicompartmental neuron models, capturing the intricate dynamics of the CA3 and CA1 subfields. In this circuit, we study the interaction between theta and gamma rhythms, crucial in facilitating cognitive processes such as memory formation and spatial navigation. With a simplified hippocampal model, we can reproduce observed phenomena, such as the long delay between CA3 and CA1 pyramidal cells and the role of inhibition in the enhacement of theta-gamma cross-frequency coupling.