Source separation and higher-order causal analysis of MEG and EEG

Kun Zhang, Aapo Hyvärinen
Proceedings of the 26th Conference on Uncertainty in Artificial Intelligence, PMLR R8:734-741, 2010.

Abstract

Separation of the sources and analysis of their connectivity have been an important topic in EEG/MEG analysis. To solve this problem in an automatic manner, we propose a two- layer model, in which the sources are con- ditionally uncorrelated from each other, but not independent; the dependence is caused by the causality in their time-varying vari- ances (envelopes). The model is identified in two steps. We first propose a new source separation technique which takes into ac- count the autocorrelations (which may be time-varying) and time-varying variances of the sources. The causality in the envelopes is then discovered by exploiting a special kind of multivariate GARCH (generalized au- toregressive conditional heteroscedasticity) model. The resulting causal diagram gives the effective connectivity between the sep- arated sources; in our experimental results on MEG data, sources with similar functions are grouped together, with negative influ- ences between groups, and the groups are connected via some interesting sources.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR8-zhang10e, title = {Source separation and higher-order causal analysis of {MEG} and {EEG}}, author = {Zhang, Kun and Hyv{\"a}rinen, Aapo}, booktitle = {Proceedings of the 26th Conference on Uncertainty in Artificial Intelligence}, pages = {734--741}, year = {2010}, editor = {Grünwald, Peter and Spirtes, Peter}, volume = {R8}, series = {Proceedings of Machine Learning Research}, month = {08--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r8/main/assets/zhang10e/zhang10e.pdf}, url = {https://proceedings.mlr.press/r8/zhang10e.html}, abstract = {Separation of the sources and analysis of their connectivity have been an important topic in EEG/MEG analysis. To solve this problem in an automatic manner, we propose a two- layer model, in which the sources are con- ditionally uncorrelated from each other, but not independent; the dependence is caused by the causality in their time-varying vari- ances (envelopes). The model is identified in two steps. We first propose a new source separation technique which takes into ac- count the autocorrelations (which may be time-varying) and time-varying variances of the sources. The causality in the envelopes is then discovered by exploiting a special kind of multivariate GARCH (generalized au- toregressive conditional heteroscedasticity) model. The resulting causal diagram gives the effective connectivity between the sep- arated sources; in our experimental results on MEG data, sources with similar functions are grouped together, with negative influ- ences between groups, and the groups are connected via some interesting sources.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T Source separation and higher-order causal analysis of MEG and EEG %A Kun Zhang %A Aapo Hyvärinen %B Proceedings of the 26th Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2010 %E Peter Grünwald %E Peter Spirtes %F pmlr-vR8-zhang10e %I PMLR %P 734--741 %U https://proceedings.mlr.press/r8/zhang10e.html %V R8 %X Separation of the sources and analysis of their connectivity have been an important topic in EEG/MEG analysis. To solve this problem in an automatic manner, we propose a two- layer model, in which the sources are con- ditionally uncorrelated from each other, but not independent; the dependence is caused by the causality in their time-varying vari- ances (envelopes). The model is identified in two steps. We first propose a new source separation technique which takes into ac- count the autocorrelations (which may be time-varying) and time-varying variances of the sources. The causality in the envelopes is then discovered by exploiting a special kind of multivariate GARCH (generalized au- toregressive conditional heteroscedasticity) model. The resulting causal diagram gives the effective connectivity between the sep- arated sources; in our experimental results on MEG data, sources with similar functions are grouped together, with negative influ- ences between groups, and the groups are connected via some interesting sources. %Z Reissued by PMLR on 04 October 2026.
APA
Zhang, K. & Hyvärinen, A.. (2010). Source separation and higher-order causal analysis of MEG and EEG. Proceedings of the 26th Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R8:734-741 Available from https://proceedings.mlr.press/r8/zhang10e.html. Reissued by PMLR on 04 October 2026.

Related Material