No subject

of EEG.

Durka PJ.

for review articles.

and you could probably use the hilbert transform as well for the similar
reasons over fourier:

Hum Brain Mapp. 2003 Aug;19(4):248-72.

Aperiodic phase re-setting in scalp EEG of beta-gamma oscillations by
state transitions at alpha-theta rates.

Freeman WJ, Burke BC, Holmes MD.

There are other Freeman references this is just a recent one of very
interesting data.

Biomed Tech (Berl). 2002;47 Suppl 1 Pt 2:577-80.

MEG-analysis using the Hilbert transform.

Link A, Elster C, Sander T, Lueschow A, Curio G, Trahms L.

Clin Neurophysiol. 2002 May;113(5):754-63.

On the time resolution of event-related desynchronization: a simulation
study.

Knosche TR, Bastiaansen MC.

What do others make of this argument 10K acquisition to see brief power
changes?

hilbert and wavelet transforms of single-trial eeg data have been shown to
very comparable:

J Neurosci Methods. 2001 Oct 30;111(2):83-98.
Related Articles, Links

Comparison of Hilbert transform and wavelet methods for the analysis of
neuronal synchrony.

Le Van Quyen M, Foucher J, Lachaux J, Rodriguez E, Lutz A, Martinerie J,
Varela FJ.

Take care,

-Morgan

On Fri, 31 Oct 2003, Arnaud Delorme wrote:

> I agree with Vladimir. I think that computing an AR model will only
Ã> construct a parametric model for your data and will not change the way
> you apply coherence. As far as I understand AR models, the main
> advantage of using AR models concerns the speed of computation (with AR
> models, the activity at one time point is a linear combination of the
> activity at previous latencies; since FFTs are linear, you do not need
> to recompute the FFT at each point, you may simply use a linear
> combination of the FFT decompositions from previous latencies; so you
> may compute spectral decompositions on-line when the subject is
> performing which might not be possible using more time-consuming
> standard approaches).
>
> Computing coherence on a single-trial does not make sense. You may
> compute the cross time-frequency spectrum of the two processes though
> (to do so, try editing the crossf function and remove the normalization
> part line 970 "Coher  = cohercomppost(Coher, trials);"; Then use 'type',
> 'coher' when you launch crossf and that might do the trick).
>
¶ller, B> Best
>
> Arno
>
> ps: for computing power spectrum using AR models, Dennis McFarland
> recommended me chapter 13.7 of the numerical recipes in C, available
> online at http://www.library.cornell.edu/nr/bookcpdf/c13-7.pdf
> --
>
> *Arnaud Delorme, Ph.D.*
> Computational Neurobiology Lab, Salk Institute
> 10010 North Torrey Pines Road
> La Jolla, CA 92037 USA
>
> *Tel* : /(+1)-858-458-1927 ext 15/
> *Fax* : /(+1)-858-458-1847/
> *Web page *: www.sccn.ucsd.edu/~arno <http://www.sccn.ucsd.edu/%7Earno>
Ã>
>
> _______________________________________________
> Eeglablist mailing list Eeglablist at sccn.ucsd.edu
> Eeglablist page: http://www.sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsub at sccn.ucsd.edu
> To switch to digest mode, send an empty email to eeglablist-digest at sccn.ucsd.edu
>
¤rbel Schack, Matthias Arnold and Herbert Witte