New research project at the FU Berlin: Focus on brain structures!

New research project at the FU Berlin: Focus on brain structures!

With a grant of over 1.12 million euros from the Volkswagen Foundation, a new research project will begin at the Free University of Berlin on July 16, 2025. Under the direction of Prof. Dr. Peter Robin Hiesinger, the project is entitled “The Information Content of Brain Wiring” and will be funded for five years from January 2026. This project aims to quantitatively record the information content of biological brain structures, particularly the fruit fly brain (Drosophila). Hiesinger highlights that so far only artificial neural networks can be quantified in bits, which reveals a knowledge gap in the understanding of biological networks.

The research team plans to pursue experimental approaches to determine lower information limits at different levels of description. An important step will be the use of compression algorithms and model reconstruction based on live image data to analyze brain structure. An interdisciplinary approach creates a bridge between biological and artificial intelligence. The aim is to gain new insights into neurobiology and enrich the discussion about natural and artificial thought processes.

Interdisciplinary collaboration and financing

The project is not only financed by the Volkswagen Foundation, but also receives around 400,000 euros from the German Research Foundation (DFG) for a complementary project. The DFG project entitled “A bio-inspired electrical engineering approach to circuit development” cooperates with partners from electrical engineering. These two projects strengthen the research strength of the Free University of Berlin in the areas of biology, neuroscience, computer science and electrical engineering and are part of a broader research concept that deals with the elucidation of neuronal structures.

One of the foundations for the research project is the connectome, which was reconstructed by the FlyWire Consortium. The connectome belongs to a seven-day-old adult female Drosophila melanogaster and contains a total of 127,978 neurons and over 2.6 million threshold connections. The latest versions of the dataset show that the synapses were detected algorithmically, with a confidence score offering signs of measurement accuracy. Neurons have been classified into different categories based on their connections, which provide important insights into their functionality in the brain.

Insights into the neurobiology of artificial intelligence

As part of the Dahlem Science Discussions, Prof. Dr. Robin Hiesinger will give a lecture entitled “Neurobiology of Artificial Intelligence” on January 15, 2025. Artificially intelligent systems such as ChatGPT are based on artificial neural networks, and the lecture will discuss the comparison between biological and artificial neural networks. Questions about genetically coded development and its role in intelligence will also be discussed.

The Dahlem Science Talks offer anyone interested the opportunity to find out more about current research topics. The events take place every two months on the first Wednesday at 6 p.m. in the SupraFAB research building and are free of charge. Hiesinger, spokesman for the DFG research consortium FOR5289 (RobustCircuit.org) and author of the book “The Self-Assembling Brain” (Princeton University Press, 2021), aims with his projects not only to gain new scientific knowledge, but also to provide sustainable data and publications to facilitate access to research.

In summary, the one-year research project at the Free University of Berlin promises not only deeper insights into the fruit fly brain, but also significant insights into the interaction between biological and artificial systems in neurobiology.