Computational Neuroscience

Analytical Methods for Signal Separation and Localization in Single-Trial Event Related Potential (ERP) to investigate Brain Dynamics

(T.N. Hoang, W.El-Deredy)

This is a multidisciplinary collaborative project involving the School of Computing and  Mathematical Sciences (CMS) at John Moores University and the Cognitive Neuroscience Laboratory at the University of Liverpool to develop analytical methods for signal separation of single-trial EEG recordings, with the aim of investigating brain dynamics that are lost by averaging.

New developments in blind source separation and noise reduction using wavelets will be combined with current methods for spatial localisation, to track the spatial and time course of EEG sources during cognitive tasks.  This will combine existing signal processing and software developments at CMS with expertise in neuroanatomy and neurophysiology to interrogate the mathematical models, as well as existing experimental facilities, at Liverpool University. The new tools will be initially validated using specifically designed cognitive experiments for which the signal sources in the brain are known, moving on to the study of higher-level cognitive function.