Evolutionary optimization of neuronal processing
Despite spectacular progress in artificial intelligence and machine learning, the information processing capabilities of animal and human brains still far surpass man made machines. This pertains particularly with respect to the robustness and flexibility of neurobiological information processing and its energy efficiency compared to digital computing devices. Recent technological progress has opened new and stringent perspectives on the evolution of central nervous system structure, information processing and development. The development of powerful computational optimization theories for neuronal circuits has undergone a parallel revolutionary change. Together these breakthroughs are setting the stage to take a novel approach and systematically decipher the evolutionary principles of neuronal information processing. This approach will leverage the rapidly expanding experimental toolboxes of neuroscience such as optogenetics, large scale imaging, multielectrode recording and connectomics to collect data sets much more comprehensive and informative than classical comparative studies could. Genome editing and viral transduction approaches are enabling researchers to apply genetic techniques and tools even in non-standard animal models to gain unique evolutionary information. Finally, guidance from theoretical and computational neuroscience provides this new stage of evolutionary studies both with an unprecedented level of quantitative precision and with the capability to distill the engineering principles used by brain circuits into theories, algorithms and systems with technological application.
The Priority Program "Evolutionary Optimization of Neuronal Processing" concentrates on three interrelated thematic foci:
The PP provides a platform
- to foster the emerging field of neural circuit evolution
- to strengthen this research area in Germany
- to connect to scientific and strategic key partners internationally
The first 3 year-funding period of the PP 2205 started in early 2020.