Guest speaker: Diogo Santos-Pata, postdoctoral researcher at the Institute for Bioengineering of Catalonia, Spain
We continuously form memories of our experiences by binding available sensory information into compressed representations. Those memories are later recalled and used to perform problem solving, such as computing the trajectory from one location to another. However, the neural mechanisms behind associative learning and episodic problem-solving are not fully understood. In the first part of this talk, professor Diogo Santos-Pata will present evidence suggesting that, during navigation, human hippocampal theta oscillations (3-9Hz) impose time windows for associative learning. Next, Santos-Pata will present recent results from human hippocampal recordings during semantic content generation, advocating that high-frequency oscillations (sharp-wave ripples, 70-140Hz) are involved in linking dissociated memories/concepts.
Furthermore, Santos-Pata will present theoretical and computational work highlighting the role of GABAergic neurons in driving learning in the hippocampus and how it relates to current machine learning techniques.