Alexander Lerchner

Postdoctoral Research Fellow
UCL, Gatsby Computational Neuroscience Unit
Alexander Lerchner

[ Research ] [ Selected Publications ] [ Short Bio ] [ Contact ]

Research

I am working on a systems-level theory of brain function that proposes how the interaction between frontal cortex, posterior cortex, the basal ganglia, and the hippocampal formation gives rise to our cognitive abilities.

My current research within this endeavour is focused on two aspects at different levels of description: systems neuroscience and dynamics of large neural networks.

Systems Neuroscience
I am using experimental methods to address theoretical concepts in systems neuroscience, in particular the role of prefrontal cortex in categorization and rule switching. I am pointing out that many behavioural tasks that probe higher-level perception (such as perceptual categorization or rule switching) are confounded by a requirement for  action selection, and I am suggesting how this problem can be overcome in principle. To illustrate my point and to re-assess prefrontal cortex function, I have designed an experimental paradigm for testing rule-based categorization and switching of category rules without such a confounding element. In collaboration with Takafumi Minamimoto and Barry Richmond in the Laboratory for Neuropsychology at the NIH, we developed a specific paradigm and tested it in a study on prefrontal cortex. We have obtained exciting results that would have been difficult to explain in light of previous findings, yet are in agreement with my predictions at the outset of the project. We are currently preparing a manuscript on the study.

Dynamics of large neural networks
What can we learn from data obtained by electrophysiology in living animals? Every neuron in the neocortex is connected with tens of thousands of other neurons, making even the smallest conceivable sub-circuits much larger than the number of neurons that we can observe simultaneously. To get beyond crude behavioural correlates of neural activity towards an understanding of how networks of neurons give rise to function, we need good theories that predict the dynamics of large neural networks. I am using mathematical methods from physics and statistics, combined with numerical algorithms, to develop and investigate theories of large neural networks.


[ Top ]

Selected Publications


[ Top ]

Short Bio

Aug. 2007 - present Postdoctoral Research Fellow
University College London (UCL)
Gatsby Computational Neuroscience Unit
London, UK

Aug. 2004 – July 2007 Postdoctoral Research Fellow
National Institutes of Health (NIH)
Laboratory of Neuropsychology
Section on Neural Coding and Computation
Bethesda, USA

June 2004 Ph.D.
Computational Neuroscience
DTU and Nordic Institute for Theoretical Physics (NORDITA), Advisor: John Hertz
"Dynamics of Large Spiking Neural Networks - Modeling Circuits in the Central Nervous System"
Copenhagen, Denmark

June 1999 M.Sc.
Applied Mathematics - Mathematical Computer Science
Vienna University of Technology, Advisor: Kurt Hornik
"Analysis of Incomplete Multivariate Data with Computational Neural Networks"
Vienna, Austria

[ Top ]

Contact 

Alexander Lerchner
Gatsby Computational Neuroscience Unit
17 Queen Square
London, WC1N 3AR
United Kingdom

lerchner [at] gatsby.ucl.ac.uk
http://www.gatsby.ucl.ac.uk/~lerchner

[ Top ] [ Research ] [ Publications ] [ Short Bio ] [ Contact ]

Last updated: 13 September 2008