Mathematics and Neuroscience Seminar

We analyse the question of whether and, if so, how neuroscience allows for mathematical descriptions.

Speaker: Professor J. Leo van Hemmen, Chair of Theoretical Biophysics, Technical University of Munich, Germany

Date: 11am-12pm, 2 August 2018

Venue: Ian Potter Auditorium, Kenneth Myer Building (Melbourne Brain Centre), Building 144

Abstract: It is argued that a mathematization of natural phenomena never happens by itself. First, appropriate key concepts must be found that are intimately connected with the phenomena one wishes to describe and explain mathematically. Second, the scale on, and not beyond, which a specific description can hold must be specified. Different scales allow for different conceptual and mathematical descriptions. This is the scaling hypothesis. Third, can a mathematical description be universally valid and, if so, how? Here we put forth the argument that universals also exist in theoretical neuroscience, that evolution proves the rule, and that there is still lots of space for new developments initiated by an intensive interaction between experimental and theoretical neuroscience. Finally, the thesis is posed that major insight is provided by a careful analysis of the way in which particular brain structures respond to perceptual input and in so doing induce action in an animal's surroundings.

Bio: Professor J. Leo van Hemmen has done some of the foundational work on analysing the functional implications of activity-dependent neural plasticity and in understanding the principles underlying information processing in neural systems, particularly the sensory systems of animals. His research explores cellular theoretical biophysics, with a particular interest in both theoretical membrane physics and neural information processing of various sensory organs. He investigates mechanosensing – the sonic ranging of the barn owl, the location of aquatic objects through the lateral-line organs of the clawed frog and fish, infrared vision in certain snakes and their multimodal integration. Prof. van Hemmen has been the Editor in Chief for many years of the journal Biological Cybernetics (Springer), one of the leading journals in the field of computational neuroscience. He studied physics and mathematics at the University of Groningen (NL), and completed his doctorate there in 1976. Following a year of postdoctoral research at the Institut des Hautes Etudes Scientifiques (IHES) in Bures-sur-Yvette near Paris, he spent a further year as assistant professor at Duke University (Durham, NC). After that, he joined Collaborative Research Center 123 (Stochastic Mathematical Models) in Heidelberg, where he qualified as a physics lecturer in 1983. Prof. van Hemmen has been lecturing in cellular theoretical biophysics, in particular neuronal information processing, at TUM since 1990. He is a fellow of the American Physical Society.