Detecting replay in multi-unit spiking data: Bayesian networks
When
Where
Hippocampal replay refers to the re-occurrence of population-wide sequences of neural spikes during sleep, similar to sequences observed in a pre-sleep task. The generation of replay during sleep is crucial for memory consolidation. We investigate the generation of replay episodes by leveraging the excitability of neurons and model the firing of place cells during sleep and pre-sleep tasks. In this presentation, I will discuss the data generated from the NEURON simulation environment using well-structured connectivity matrices as input to an algorithm that uses random spike intervals to extract subgroups of neurons with repetitive firing patterns. I will also build upon how Bayesian Networks capture the causal effect between neurons, defining neurons as nodes and the causal relationship between excitatory and inhibitory neurons as directed edges. Finally, I will motivate the role of place cells and the connectivity of neurons in the hippocampal replay that provides insights into cognitive functions such as learning and decision-making.