Independent Fellow, P.I., Complex Systems Group (Boehringer Ingelheim Foundation)
Lab Homepage: www. synapticwiring.com
Social stress is one of the main risk factors for depressive episodes. Individuals, however, react very differently to stressors. The underlying mechanisms that generate individual resilience to social stress are poorly understood. In these processes, the subjective valuation of rewards appears to play a critical role.
These mechanisms can only be studied in a social context and in longitudinal designs. We have therefore developed a sensor-rich non-invasive habitat that serves to monitor individual social behaviors and reinforcement learning in mouse colonies with high dimensional readouts and minimal interference of the experimenter as originally proposed in the 3R principles. Causal interactions are revealed by computational modeling and deep learning tools. We particularly aim at determining influences of genetic polymorphisms and possible behavioral and pharmacologic interventions to individual expression of behaviors within populations.
We examine the fundamental question how we remember others and attribute positive or negative memories. We could reveal the mechanism by which the neurohormone oxytocin increases signal extraction in neuronal networks and thereby enhances the formation of memories of other individuals. To answer this and related questions, we employ large scale single unit recordings in transgenic mice and a newly developed functional MRI approach to capture the dynamics of brain activity and discover new entry points to develop better therapies.
Value needs to be attributed to environmental events to make useful decisions. The formation of such predictions and their outcome-based evaluation are impaired in severe mental disorders. We examine, with high dimensional network recordings, functional MRI, computational modelling and genetics in mice, how distributed, but closely interacting brain networks how value is assigned to environmental cues through reinforcement learning. Through this approach, we recently identified a distributed reinforcing network loop that generates reward prediction and clarified the underlying mechanisms. These insights are integrated into our overall question on the shaping of behavior and stress resilience in complex social Environments.