Abstract

Engineering principles of cerebral information processing

The evolutionary invention of the cerebral cortex 205 million years ago constituted a fundamental innovation in brain architecture. Within the brain, the cerebral cortex functions as a flexible learning machine, in which information is processed by large and densely connected populations of nerve cells. In humans, this brain region is critical for cognitive functions ranging from sensory perception and working memory to language processing and executive control. Although perhaps the energetically most expensive tissue in the mammalian body, the cortex excessively expanded during the evolution of primates and humans, which likely reflects unique advantages of this type of neural circuit architecture.  

Concurrent progress in neurotechnology and in the theory of neural computation is currently opening new avenues towards understanding the operation of cortical networks. In this talk, I will give an overview of our research into cortical network function and design. In particular, I will highlight (1) studies indicating that optimization principles for neuronal population codes shape the design of living nerve cells down to the bio-molecular level, (2) theoretical work exploring the dynamical basis of flexible information routing in large-scale cortical networks, and (3) research revealing that the architecture of cortical networks can converge to a unique and potentially optimal design with almost “mathematical” precision. These findings illustrate that both groundwork in mathematical and computational theory as well as trans-disciplinary research in collaboration with experimental neuroscientists are essential for progress in circuit-level neuroscience. I will conclude by discussing the prospects of bidirectional neural interfaces for next-generation studies of cortical network function.

Short Bio

Fred Wolf studied physics and neuroscience at the Goethe University Frankfurt and at the Max Planck Institute for Brain Research. Following his PhD, he was selected as the first awardee of the Altdorfer Leibniz Prize in 1999, received an Amos de-Shalit Fellowship of the Minerva Foundation, and a Schloessmann Fellowship of the Max Planck Society and subsequently worked at two international centers of theoretical neuroscience and theoretical physics, the Interdisciplinary Center for Neural Computation and Racah Institute of Physics (Hebrew University) and the Kavli Institute for Theoretical Physics (UCSB). Returning to Germany in 2001, he established the Research Group Theoretical Neurophysics at the Max Planck Institute for Dynamics and Self-Organization in Göttingen, with support by the Volkswagen Foundation and HFSP. In 2008, he was appointed honorary professor of physics at Göttingen University and in 2011 visiting professor at the Centre International de Rencontres Mathématiques, University of Marseille. Wolf has been a regular participant in research programs of the Kavli Institute for Theoretical Physics, where he served in 2010 as Program Director of the Program on "Emerging Techniques in Neuroscience". From 2010 to 2014, he served on the board of the German Neuroscience Society as section coordinator for computational neuroscience. Since 2014 he is a member of the research grand review committee of the Human Frontier Science Program. He heads the Bernstein Center for Computational Neuroscience Göttingen and the PhD program in Theoretical and Computational Neuroscience of the University of Göttingen. In 2014, he was elected Fellow of the American Physical Society (APS) and in 2017 received The Mathematical Neuroscience Prize, the world’s most highly endowed prize for mathematical contributions to understanding the brain.