The central goal of our research is to understand the synaptic and neuronal dynamics originating from the interactions of diverse adaptive processes on different time scales and the resulting emergence of complex, cognitive behaviors. In more detail, based on experimental data, we analyze with mathematical tools from various scientific fields (e.g., nonlinear dynamics, graph theory, machine learning) the interactions of different, experimentally well-known plasticity mechanisms depending on environmental stimuli and their relations to cognitive processes such as learning, computation, and memory formation, which serve as the basis of complex behaviors. Thereby, one important part of our research is to link the mathematical models to experimental findings by, on the one hand, analyzing experimental data to derive the theoretical foundations of the models and, on the other hand, by deducing experimentally verifiable predictions from the mathematical models to test the overall hypotheses. In addition, the identified theoretical principles are transferred to technological applications such as robotic platforms or neuromorphic chips to verify derived hypotheses and to advance neuro-inspired technologies.

Currently we are working on the following projects:

Molecular principles of synaptic plasticity

Changes of the synaptic strength play a fundamental role in shaping the behaviorally relevant dynamics of neuronal networks. Different activity-dependent plasticity processes such as short-term plasticity or long-term synaptic plasticity govern these changes and they are implemented by various protein and ion dynamics on the molecular level. In general, we investigate the relation between different activity-dependent plasticity processes and the spatiotemporal dynamics of molecules in the synapse.

Calcium-dependent synaptic plasticity in single and network of neurons

Calcium is an important messenger in neurons, being regulated by neuronal activity. Especially in a synapse the calcium concentration determines the shape of synaptic plasticity at this synapse and its neighboring synapses. In addition, diverse dendritic mechanisms such as dendritic spikes or organelles acting as local calcium stores manipulate the calcium dynamics and, by this, synaptic plasticity. We investigate the influence of such manipulations on learning and computation of a single and a network of neurons.

The impact of synapses on neuronal network dynamics

The mammalian brain hosts a huge repertoire of different synapse types. Clearly there is a correlation between the properties of different synapses, their numbers, and their distribution and the resulting capabilities of neuronal networks to process and memorize information. We combine and use a wide variety of tools from different scientific disciplines to map and better understand the underlying complex relations.

Advancing neuro-inspired technologies

There is a long history of the impact of neuroscientific insights on technology with its current peak in the development and application of neural networks as machine learning tools for data processing. Following this example, we are continuously seeking to transfer our insights about the working of neurons to technologies like robotics and neuromorphic computer chips, with a special focus on making these systems more versatile and adaptive.