A training algorithm for optimal margin classifiers
COLT '92 Proceedings of the fifth annual workshop on Computational learning theory
EEG responds to conceptual stimuli and corpus semantics
EMNLP '09 Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing: Volume 2 - Volume 2
Classifying event-related desynchronization in EEG, ECoG and MEG signals
DAGM'06 Proceedings of the 28th conference on Pattern Recognition
Hi-index | 0.00 |
Electroencephalography (EEG) and magnetoencephalography (MEG) are closely related neuroimaging technologies that both measure summed electrical activity of synchronous sources of neural activity. However they differ in the portions of the brain to which they are more sensitive, in the frequency bands they can detect, and to the amount of noise to which they are subject. Since semantic representations are thought to be widely distributed in the brain, this preliminary study considered if the broader coverage offered by simultaneous EEG/MEG recordings would increase sensitivity to these cognitive states. The results showed that MEG data allowed stimuli in two semantic categories (mammals and tools) to be distinguished more accurately, despite some experimental settings that were optimised for EEG. The addition of EEG data did not prove informative, indicating that it may be redundant relative to MEG, even when using dimensionality reduction techniques to combat overfitting.