In December 2011, Dr. Matthias Kaschube became the new Professor of the Goethe University in Frankfurt for “Computational Neuroscience / Computational Vision” at the Bernstein Focus Neurotechnology (BFNT) Frankfurt and fellow at the Frankfurt Institute for Advanced Studies. On the basis of quantitative models, the physicist will explore the perception of complex stimuli in the adult brain and the development of different visual systems of mammals. He also will investigate the embryonic development and organization of organs. The research professorship was established by the Federal Ministry for Education and Research (BMBF) as part of the BFNT Frankfurt.
How do we sort and process stimuli, why does the brain develop in a certain way and how do cells organize into tissue? Kaschube wants to examine such questions over the coming years by using mathematical and quantitative approaches. He mainly deals with the principle of self-organization. This describes processes that attune themselves without any external influence. Everyday examples are flocks of birds and a whirl of water in the sink. “Self-organization could play a significant role in merging different stimuli in the brain, for example the fact that we perceive color, shape and movement of an object as a unit,” explains Kaschube. He also examines the role of self-organization in early childhood development of the visual system and the formation of a functional organization in the brain.
In this context, also the comparison between primates and rodents, whose visual systems show fundamental differences, is of particular interest: “By examining the structure and data processing in the visual system of different animals with quantitative models, we can better compare them with each other. This allows us to describe both general principles and to identify very specific adaptations,” says Kaschube. In the past, he already closely cooperated in this area with scientists from the Bernstein Network such as Siegrid Lowel, Fred Wolf and Matthias Bethge.
In addition, the Frankfurt scientist will develop a physical understanding regarding tissue and organ formation. Together with Eric Wieschaus, in the United States, he developed a method to calculate the three-dimensional shapes and shape changes of cells from two-dimensional microscopy images of the fruit fly embryo. It enables one to follow exactly the emergence of body structures through collective remodelling processes of thousands of cells. Such detailed measurements make the design of computer models possible, which in turn can be used to investigate the complex processes during early embryonic development. This might also help in understanding wound healing and tumor growth.
At the Bernstein Focus Neurotechnology (BFNT) Frankfurt, Kaschube will be involved in projects with Visvanathan Ramesh, Christoph von der Malsburg and Jochen Triesch where he will bring in the principles of self-organization in the context of autonomous learning and stimulus processing in natural and artificial visual systems.
Kaschube first studied philosophy and physics in Frankfurt, before moving to Göttingen for to get a diploma in physics. “I have been very interested in the neuroscience-oriented topics in philosophy, such as perception and consciousness. But the very concrete tools of mathematics and physics fit to me much better,” says Kaschube. In the laboratory of Professor Theo Geisel at the Max-Planck-Institute for Dynamics and Self-Organization and the Bernstein Center Göttingen, he first came into contact with the interdisciplinary field of computational neuroscience. In Geisel’s group, he wrote his diploma and PhD thesis and was subsequently Bernstein Fellow for one year before he conducted research and taught for five years at the Lewis-Sigler Institute for Integrative Genomics, Princeton, USA.
The Bernstein Focus Neurotechnology Frankfurt is part of the National Bernstein Network Computational Neuroscience (NNCN) in Germany. The NNCN was established by the German Federal Ministry of Education and Research with the aim of structurally interconnecting and developing German capacities in the new scientific discipline of computational neuroscience. The network is named after the German physiologist Julius Bernstein (1835–1917).