Talk 14.05.2018

Speaker: Viola Priesemann
Max Planck Institute for Dynamics and Self-Organization, Göttingen

Title: Network dynamics and function

Abstract:
In most spatially extended systems, only a tiny fraction of all units can be assessed at a time. In the brain, for example, only a tiny fraction of the millions or billions of neurons can be sampled with millisecond precision. Such spatial subsampling in general can incur a strong, systematic bias when inferring collective properties of the full system. Subsampling can even bias the estimate of a measure as basic as the correlation strength between a pair of neurons. I present approaches that enable us for the first time to overcome the subsampling problem, and then put it to use to infer the dynamical state of cortical networks. The novel results provide first evidence that collective cortical dynamics switches adaptively between different regimes depending on functional needs.