Talk by Prof. Cees van Leeuwen
Laboratory for Perceptual Dynamics, KU Leuve
- 11.Apr.2012 14:00
- 11.Apr.2012 15:00
What is it like, for a brain to be a dynamical system?
In analyzing the temporal structure of brain activity, we readily observe the hallmark 1/f phenomenon of complexity, also known as long-term dependency, in its dynamic activity patterns. Such results are both revolutionary and yawn-inspiring at the same time. They are revolutionary in showing the insufficiency of approaching the brain as a linear system. They inspire yawns because we knew this all along. Complexity is pervasive in nature, so in order to gain ground after the initial awe has waned, we need a more explicit and detailed characterization of brain dynamics. This means we need to move from data-driven explorations of brain activity to testing of dynamical models. These models may early on be simple and qualitative in their predictions, but ultimately have to be detailed and quantitative characterizations. In developing these models, certain principles will come to the fore. Moving between experimental results and modeling, I try to sketch what these principles might be.