[FieldTrip] Extracting a timecourse from cluster of channels

[Eric Maris]

Dear Clara,

> I am using fieldtrip's cluster-based permutation test to identify a
> significant channel-time cluster showing an effect.  Now I would like
> to extract a timecourse averaged over these channels (for the purpose
> of comparing different conditions, not used to generate the cluster),
> but I'm not sure if it makes sense to do so because the channels
> belonging to the cluster change over time -- would the timecourse have
> contributions from different channels at different time points?  Do I
> need to limit my cluster to fixed channels across time? (And if so,
> how do I do that?)

Yes, you should fix channels across time. The most important point in
selecting the channels is finding evidence for the fact that the channels
reflects a single source only. This is not a statistical issue, and
therefore there is not a p-value that can guide you in this choice. If you
have collected MEG data, the best evidence is a dipolar topography of your
effect.

Best,

Eric Maris


> 
> Thanks!
> Clara
> _______________________________________________
> fieldtrip mailing list
> fieldtrip at donders.ru.nl
> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip