Home
Contact
Short CV
Research
Publications
Software
Links
Impressum

 

University of Plymouth | Centre for Theoret. and Comput. Neuroscience | Home   

 

Previous and Ongoing Research Summary

Thomas Wennekers received his Diploma (hons, equivalent to MSc) in physics from the Heinrich Heine University Dusseldorf, Germany, and a PhD (summa cum laude) in Computer Science from the Faculty of Computer Science at the University of Ulm, Germany, both, under the supervision of Prof Guenther Palm (Ulm). During his MSc and PhD he investigated the tentative role of gamma-oscillations and spatio-temporal spike patterns in associative cortical processes, proved rigorous theorems about dimension reduction in randomly connected neural networks, and developed a theory of how higher cortical processes are tentatively based on Hebbian cell assemblies ("operational cell assemblies"). The thesis was awarded best PhD thesis of the year in 1997 by the University Society of the University of Ulm.

For 4 years he worked as a research fellow at the Max Planck Institute for Mathematics in the Sciences in Leipzig (MPIMIS), Germany, where he had responsibilities for the Neural Networks and Cognitive Systems group. During this time he investigated dynamic receptive fields in primary visual cortex using mathematical tools and developed a general approximation method for localised solutions in neural field models. The method was successfully applied to various phenomena in cortical simple cells (e.g., receptive field sharpening or contrast invariance). Together with Dr Nihat Ay (MPIMIS) he also put forward the framework of "spatio-temporal stochastic interaction" as a conceptual approach towards the selforganisation of cortical structures based on information theory. Extending the well-kown "information-maximisation principle" (Infomax), this framework still is the only neural information-theoretic approach that truly considers recurrent systems and spatio-temporal activation patterns. It has numerous implications for cortical systems, including spike-timing dependent plasticity or globally almost deterministic activity although single neurons fire highly irregular.

In 2003 TW moved to the Ruhr-University Bochum, Germany, where he became Juniorprofessor (a new German academic post roughly equivalent to Assistant Professor) at the Institute of Neuroinformatics. He was the first JP that (after just 7 months) obtained a permanent academic post.

Since November 2003 TW is Reader in Computational Neuroscience at the Centre for Theoretical and Computational Neuroscience (CTCN) a dedicated brain research centre at the University of Plymouth, UK. In close contact with experimenters, he works on large-scale models of the cortex aiming at the selforganisation of maps and higher level functional processes in the layered mammalian cortex. He also develops further his previous theories about operational cell assemblies, which he investigates in system-level computer models of multiple cortical areas that target on perceptual, behavioural, and language processes. Beside providing a better understanding of cortical processes, these works also feed into the VLSI-design for bioinspired future computing architectures done in collaboration with other institutes in the UK and abroad. TW has published over 50 papers.



General Research Interests

  • Computational neuroscience
    • Gamma-oscillations and spike synchronization
    • Dynamics of spatio-temporal receptive fields
    • Spatio-temporal spike patterns, synfire chains

  • Mathematical neuroscience
    • Biophysical, dynamical and computational properties of spiking neuron models
    • Neural field equations
    • Dimension reduction in neural networks

  • Brain theory
    • Hebbian cell assemblies and associative memories
    • Concept and similarity structure in cortical representations
    • General principles of high level information processing


Special projects/areas



 
   

   © 2004 by -thowe-