help > RE: Contrast question
Oct 17, 2017  11:10 PM | Andrew Zalesky
RE: Contrast question
Hi Amna,

I'm not sure if I follow, but it seems that you are using the directed version to analyze undirected connectivity (PLV). The directed version should only be used to analyze directed connectivity.

Also, I'm not sure if I understand the meaning of up/down here.

In general, I am not certain that PLV and Granger will necessarily yield identical estimates. I can think of situations where this might not be the case.

Andrew



Originally posted by Amna Ghani:
Hi Andrew,

I see, that means two sub-networks were significant.
One more thing, when I am applying NBS on undirected connectivity (PLV) I am getting increased functional connections in 'down' vs 'up'
However, NBS showed totally opposite results for directed connectivity (Granger causality). 

Is it because I am using two different measures (directed vs undirected). But, I think, in principle, both connectivity measures should give close to similar results?

Thanks,
Amna

Threaded View

TitleAuthorDate
Amna Ghani Oct 14, 2017
Andrew Zalesky Oct 14, 2017
Amna Ghani Oct 16, 2017
Andrew Zalesky Oct 16, 2017
Amna Ghani Oct 17, 2017
RE: Contrast question
Andrew Zalesky Oct 17, 2017