[FieldTrip] negative connectivity strength from mvgc

Jörn M. Horschig jorn at artinis.com
Wed Sep 9 13:58:07 CEST 2015

Hi Tyler,

as far as I know only the instantaneous causality term might become
negative, not the granger values x->y or y->x themselves. Reasons for this
are a bit dubious as far as I understood (within my limits). In [1], Ding et
al. explained that that the instantaneous causality term can become negative
for “certain situations” and that it thus might not have a “readily
interpretable physical meaning”. It could have something to do with SNR or
with real, third region influences. But this is speculative and in fact I
have no idea. I hope this helps, or at least provides a reference to read up
on this ;)


[1] Ding, M., Chen, Y., and Bressler, S. L. (2006). Granger Causality: Basic
Theory and Application to Neuroscience. In Handbook of Time Series Analysis,
B. Schelter, M. Winterhalder, and J. Timmer, eds. (Wiley-VCH Verlag GmbH &
Co. KGaA), pp. 437–460. Available at:
[Accessed April 8, 2014].


Jörn M. Horschig, PhD, Software Engineer
Artinis Medical Systems  |  +31 481 350 980 

> -----Original Message-----
> From: fieldtrip-bounces at science.ru.nl [mailto:fieldtrip-
> bounces at science.ru.nl] On Behalf Of Tyler Grummett
> Sent: Wednesday, September 9, 2015 10:52 AM
> To: FieldTrip discussion list <fieldtrip at science.ru.nl>
> Subject: [FieldTrip] negative connectivity strength from mvgc
> Hello fieldtrip experts,
> I decided to try multivariate granger causality today, using the steps in
> tutorial 'analysis of sensor- and source-level connectivity'. I noticed
that I am
> getting negative connectivity measures in the granger.grangerspctrm. Is
> normal? If so, why does this happen?
> I apologise if this has been asked before!
> Tyler
> _______________________________________________
> fieldtrip mailing list
> fieldtrip at donders.ru.nl
> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip

More information about the fieldtrip mailing list