[FieldTrip] How to handle results from cluster based permutation?
martabortoletto at yahoo.it
Thu Jan 17 12:39:10 CET 2019
Dear Fieldtrip community,I have a queston related to further exploration of results after a significant effect of a cluster based permutation analysis.
I am using cluster based permutation test to look for differences in evoked potentials amplitude before and after a learning task.
I found that there is a significant difference between the two conditions, with the largest cluster between 80 and 140 ms on frontocentral electrodes.
Now, I would like to test if the change in evoked potentials between pre- and post-task is related to the behavioral improvement by running correlation analyses.
My question is how can I do that. My understanding is that taking the mean activity in the cluster to run the correlation with behavioural data is problematic.
Shall I rather run correlation analyses with cluster based permutation on the evoked potentials, perhaps restricting the analyses to the time window highlighted in the first analysis (80-140 ms)?
I would really appreciate your help with this.Marta
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the fieldtrip