[Mne_analysis] Phase Lag Index connectivity using single epoch

Sarang S. Dalal sarang at cfin.au.dk
Mon Feb 12 05:29:49 EST 2018
Search archives:

Hi,

The PLI requires multiple epochs/trials to give a meaningful result, since it essentially measures the variability of phase differences across trials. By definition, a single epoch has no variability so this explains your result. If you only have a single epoch, then measures like the PLI are not appropriate. (I can't look it up right now, but most likely Figure 6 did indeed show a result from several epochs.)

Cheers
Sarang


On Feb 12, 2018 11:10, dae kook do <eornrsayclub at gmail.com> wrote:
Dear everyone,

I have a problem about PLI.

I saw a paper(Phase lag index: Assessment of functional connectivity from multi channel EEG and MEG with diminished bias from common sources (Hum Brain Mapp 2007)
) about PLI in MNE example (Compute all-to-all connectivity in sensor space)

In this paper, PLI is checked by each montages (Figure 6, Influence of reference)

When I read that, I think Figure 6 is calculated by each epochs.

I tried to calculate PLI using spectral_connectivity with single epoch.

But, results just return same value "1".

How to calculate PLI using single epoch?

Or is Figure 6 calculated by several epochs?

I add my result.
[본문 이미지 1]
thanks


Do Dae Guk | Signal and System Lab.

Inje Univ. | Department of Biomedical engineering | the Degree of Bachelor

#302 Building A, Inje-Ro 197, Gimhae, Gyung-nam, South Korea, 621-749

Tel. +82-55-320-3797 | Fax. +82-55-327-3292

C.P +82-10-8353-7985

E-mail : eornrsayclub at gmail.com<mailto:eornrsayclub at naver.com> | eornrsayclub at naver.com<mailto:eornrsayclub at naver.com>
        [http://postfiles16.naver.net/20160926_31/eornrsayclub_1474860354392v6hQ2_PNG/%BC%AD%B8%ED%B1%D7%B8%B2.png?type=w3]


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20180212/0f098854/attachment-0001.html 


More information about the Mne_analysis mailing list