Abstract
This paper presents an application of subspace projection techniques to the classification of some EEG data for cognitive psychology.
The data used come from an “odd ball” experiment, where the subject was shown targets to which they had to respond by pressing a button. The target was shown with frequency 1/10 of the frequency with which irrelevant targets (to which no response was necessary) were shown. The data used in this study concern only the recordings when the target of interest was actually shown. This means that we have the recordings of 20 EEG channels, from 1393 trials, concerning 11 different subjects. However, in the classification experiments presented, we use only a single channel at a time. The classes we try to disambiguate are those of fast and slow response. The signals in each class are further subdivided into two halves, one used for training and one for testing.
Our purpose is to predict as soon as possible after the visual stimulus whether the subject is likely to succeed (ie respond fast) or fail (ie respond slowly) in his task.
We show that if each subject is treated individually, and if we use parts of the signal that are identical for the two classes and parts that are maximally different, we obtain results that are significantly better than the pure chance level. The results presented are preliminary results of work in progress.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Baillet S and Garnero L, 1997. A Bayesian Approach to Introducing Anatomo-Functional Priors in the EEG/MEG inverse problem, IEEE Transactions on Biomedical Engineering, vol 44, pp 374–385.
Lange D H, Pratt H and Inbar G F, 1995. Segmented matched filtering of single event related evoked potentials. IEEE Trans Biomed Eng, vol 42 No 3, pp 317–21.
Lange D H, Pratt H and Inbar G F, 1997. Modelling and estimation of single evoked brain potential components. IEEE Trans Biomed Eng, vol 44 No 9, p 7
Mosher J and Leahy R, 1998. Recursive MUSIC: A framework for EEG and MEG source localisation, IEEE Transactions on Biomedical Engineering, vol 45, pp 1342–1354.
Jung T P, Makeig S, Stensmo M and Sejnowski T, 1997. Estimating alertness from the EEG power spectrum, IEEE Transactions on Biomedical Engineering, vol 44, pp 60–69.
Oja E, 1983. Subspace Methods of Pattern Recognition, Research Studies Press, ISBN 0 86380 010 6.
Roessgen M, Zoubir A and Boashash B, 1998. Seizure detection of newborn EEG using a model based approach, IEEE Transactions on Biomedical Engineering, vol 45, pp 673–685.
Trejo L J and Shensa M J, 1999. Feature extraction of event-related potentials using wavelets: an application to human performance monitoring. Brain Lang, vol 66 No 1, pp 89–107.
Trejo L J, Kramaer A and Arnold J, 1995. Event related potentials as indices of display monitoring performance. Biological Psychology, vol 40, pp 33–71.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2001 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Beltran, A.S., Petrou, M., Barrett, G., Dickson, B. (2001). Identification of electrical activity of the brain associated with changes in behavioural performance. In: Singh, S., Murshed, N., Kropatsch, W. (eds) Advances in Pattern Recognition — ICAPR 2001. ICAPR 2001. Lecture Notes in Computer Science, vol 2013. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-44732-6_30
Download citation
DOI: https://doi.org/10.1007/3-540-44732-6_30
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-41767-5
Online ISBN: 978-3-540-44732-0
eBook Packages: Springer Book Archive