Research SummaryIntermediate Fellowship research summary
As animals navigate the world, they maintain an internal representation of their own spatial location. The ability to do so has been attributed, in part, to a structure known as the hippocampus. A group of cells in the hippocampus, termed place cells, responds with higher activity every time an animal passes a specific location in space. Distinct units respond at different locations forming a patchwork that covers the entire space. As the animal moves through space, groups of place cells are sequentially activated. Retaining this spatial and temporal organization then provides a neural instantiation of the memory of a path traversed. The goal of this project is to understand how the structure of the neuronal network facilitates the formation of these sequences of activity and their subsequent storage as memory representations.
Our approach is to construct and simulate detailed and idealized models of neuronal networks to understand the peculiarities this system while also abstracting broad principles underlying information processing in the brain. The inherent nonlinearity and high dimensionality of neuronal networks makes the problem of relating the structure of a network to the collective dynamics of its constituent neurons a particularly difficult one. Using tools from graph theory coupled with our knowledge of neuronal interactions we will develop low dimensional descriptions of the dynamics of the system that can be used to construct specific sequences of activity.
Graph coloring (a prescription that assigns different colors to nodes of the network that are reciprocally coupled) provides a useful descriptor of the structure of the inhibitory sub–network. In the network shown, local inhibitory interneurons (LN) can be colored using a minimum of four colors. Neurons that are associated with the same set of colors tend to spike synchronously. The inhibitory interneurons entrain principal neurons (PN) that spike synchronously during short epochs of time. The intrinsic properties of the neurons and the topology of the network determine the length, temporal ordering and reliability of the sequence of synchronized bursts generated by the network.