NSF Award
8610353
The purpose of this research is to model empirically the distributed processing network responsible for certain higher brain functions in humans. The models to be achieved will have medical relevance (EEG analysis) and relevance to the design of parallel pattern recognition and decision systems. In previously funded research, the group has used a large array of scalp electrodes to specify the origin of EEG signals associated with specific functions. To solve some of the problems encountered in prior work, a larger array (128 channels) will be implemented along with an appropriate finite-difference head and brain model. These will allow deconvolution of the scalp-recordings to just above the cortical surface. Magnetic resonance images will be used to arrive at the model which will be validated against independent data sets.
