J. Peters (Janek)

In order to maximize efficiency and speed, the brain uses information gathered in the past to make a prediction on what will happen next. A consequence of this predictive information encoding is that some sensory areas will respond especially strongly when an unexpected signal is detected. In the auditory cortex, there seems to be a division of labour between neuronal populations when it comes to encoding raw stimulus information and matching this information to an internal model. Furthermore, there are neurons that don't seem to care much about what their neighbors are doing, while other neurons fire in close collaboration with their neighbors.
While these characteristics have been gradually uncovered over time, the spatial location of these different populations as well as their local and dynamic stimulus encoding have not been observed on a wide scale.
In our current project, we aim to find out more about these functionally separate populations and how they contribute to functional and accurate hearing of pure tones as well as natural, complex sounds. To this end, we are currenty setting up a state of the art two-photon microscope and a novel experimental procedure in order to perform optical calcium imaging of the entire auditory cortex in mice while they listen to predictable and unpredictable sounds.
We hope that insights generated by our research may help to understand sensory processing in all vertebrate brains and may lead us closer to one day uncovering the computing mechanisms that underlie our perception of reality.

• Natural Computing & Neurotechnology

• Computational Neuroscience Lab