Michael Herrmann
Thu 06 Oct 2016, 12:45 - 13:45
4.31/33, IF

If you have a question about this talk, please contact: Steph Smith (ssmith32)

Neural systems have been shown to exhibit critical behaviour which implies the presence of power-law distributions of neural avalanches. Criticality in neural systems is known to imply functional benefits such as maximal dynamic range of responses to inputs and optimal information transmission within the network. In order to show that this translates also into behavioural benefits, we consider an agent that interacts with its environment while being controlled by a critical network. Our approach is based on the natural assumption that internal connectivity should represent statistical properties of the environment and follows thus the principle of predictive coding. In addition, we will consider the role of correlations of the neural activity in the system which provides a basis for the discussion of the relation to different concepts of criticality.