Kurzfassung

The concept of neuronal assemblies proposes that behaviorally relevant representations
(percepts, motor programs, memories) take the form of transient activity patterns within
distributed neuronal networks. Recent progress has opened new possibilities to study the mechanisms underlying cognitive states at the level of small networks. For a truly mechanistic understanding of the neuronal processes underlying cognition and behaviour, however, we need to link both approaches at the level of...The concept of neuronal assemblies proposes that behaviorally relevant representations
(percepts, motor programs, memories) take the form of transient activity patterns within
distributed neuronal networks. Recent progress has opened new possibilities to study the mechanisms underlying cognitive states at the level of small networks. For a truly mechanistic understanding of the neuronal processes underlying cognition and behaviour, however, we need to link both approaches at the level of neuronal networks. This approach requires the use of different
preparations and experimental approaches, reaching from cellular physiology in vitro to
research on behaving animals in vivo.
In our consortium, we will focus on three distinct, well-defined mammalian networks. We will
share equipment and techniques (including mathematical descriptions of networks) for
similar approaches in the neocortex, hippocampus and striatum. More importantly, we share
the same understanding of the importance to study transient neuronal assemblies and we
believe that similar questions have to be solved in all three (and further) model regions.
Leading questions are:

• Which network states constitute functionally relevant representations?


• Which network architecture is crucial for information processing in a given task?


• How are the rapid switches between different states achieved without loss of relevant information?


• What are the cellular and network mechanisms of neuromodulators?


• What is the role of noise and how are transient, semi-stable states ("attractors") generated?