[Eeglablist] PLV in eeglab available yet?

[Agatha Lenartowicz]

Hi guys - Has there been any development on building a cross-channel/IC PLV routine yet? Last I see is a note by Tim (April 2011, see below) that he'll try to contribute something ...but it's not clear to me whether that has been done...

Many thanks 
Agatha


[Eeglablist] single trial phase locking

Tim Mullen mullen.tim at gmail.com 
Mon Apr 18 12:21:15 PDT 2011

Previous message: [Eeglablist] single trial phase locking
Next message: [Eeglablist] single trial phase locking
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Hi,

I have not implemented any PLV code in SIFT (yet). However, SIFT
implements methods for computing time-varying spectra, coherence,
partial coherence, or multivariate Granger-causal quantities on
multiple or single-trial data using Vector Autoregressive models. You
can obtain these quantities in the frequency-domain for single-trial
data using either overlapping sliding windows and a standard VAR model
(pop_est_fitMVAR.m) or for each time point individually using a Kalman
filter (est_fitMVARKalman.m). If you are interested in driving/driver
or *effective* (causal) connectivity relationships then the GC is
generally more appropriate and interpretable than PLV or coherence
(e.g., dDTF or PDC or Granger-Geweke causality for
bivariate/trivariate systems).

I don't see how the epoching workaround proposed by Michiel for
computing sliding-window PLV using ITC would work since ITC is not the
same as the bivariate PLV. ITC quantifies the phase-locking for a
single channel across event-locked trials, but will not tell you
whether two channels are phase-locked w.r.t. each other.

I have on my TODO list to contribute a PLV function for EEGLAB, and
when I do get to this (perhaps sooner than later), I'll make sure to
include single-trial capabilities...otherwise, if anyone wants to step
to the plate, that's cool too.

In the meantime, you might take a look at SIFT for getting those
time-varying single-trial coherence or GC estimates...

Tim

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20110713/0506cd08/attachment.html>