• Tuesday, February 28th 2017 at 16:00 - 17:00 UTC (Other timezones)
  • General participation info   |   Participate online   |   + Phone in United States: +1 (571) 317-3116 Australia (toll-free): 1 800 191 358 Austria (toll-free): 0 800 080061 Belgium (toll-free): 0 800 78881 Canada (toll-free): 1 877 777 3281 Denmark (toll-free): 8025 0919 France (toll-free): 0 805 541 052 Germany (toll-free): 0 800 723 5274 Ireland (toll-free): 1 800 818 263 Netherlands (toll-free): 0 800 023 1954 Sweden (toll-free): 0 200 330 924 Switzerland (toll-free): 0 800 000 452 United Kingdom (toll-free): 0 800 031 4727 United States (toll-free): 1 877 309 2070 Access Code: 481-670-493 Audio PIN: Shown after joining the meeting Meeting ID: 560-403-933

Physical systems that evolve in time, including the human brain, can be mathematically described
and analyzed through tools provided by dynamical systems theory (DST). DST conceives the state
of a dynamical system as a point in a (potentially very high-dimensional) variable space (like the
spiking rates of neurons), with the temporal evolution of the system given by a specific trajectory
within this space. DST describes and predicts the short- and long-term behavior of these
trajectories. In theoretical neuroscience, DST is often seen as a language or layer of description that
mediates between biophysical, neurochemical, and structural properties of the nervous system on
the one hand side, and its computational and cognitive properties on the other. In my talk I will
argue that DST provides a central point of convergence for understanding psychiatric disorders,
with potentially profound implications for the (pharmacological) treatment of these illnesses.

An overview of formal DST concepts will first be provided using examples from neuroscience. I
will then introduce some of the highly data-driven neural system models we have developed over
the past few years for studying the relationship between the system’s physiological-anatomical and
its neuro-dynamical properties, and illustrate their application to the study of pharmacological
conditions and risk genes. The second half of my talk will focus on how to identify properties of the
system dynamics from experimental, multiple single-unit or neuroimaging, time series
measurements. Both non-parametric (based on delay embedding theorems) and parametric (based
on state space models) mathematical-statistical tools for this purpose will be discussed. I will
conclude with some thoughts on how these tools and methods could be harvested in (computational)

Daniel Durstewitz
Professor in Theoretical Neuroscience and Department Head
Dept. Theoretical Neuroscience,
Central Institute of Mental Health Mannheim

Daniel Durstewitz – Psychiatric illnesses as disorders of network dynamics

Leave a Reply