[Eeglablist] std_stat vs statcondfieldtrip syntax

[Arnaud Delorme]

Dear Taylor,

Based on what I know Fieldtrip does not allow for n x m design correction, which is why statcondfieldtrip does not support it. The only way I know around this problem is to use LIMO which also uses cluster correction and allows arbitrary complex design. Unfortunately, the learning curve for LIMO is very steep and we want to make a couple of video tutorials but we do not have them yet.

Best wishes,

Arno

> On Feb 6, 2019, at 11:03 AM, Taylor Hearn <thearn at asu.edu> wrote:
> 
> Hi Everyone,
> 
> I'm using EEGLAB to perform a cluster-based permutation test (Maris and Oostenveld) , and I was hoping to get some assistance in regards to syntax/data structure.
> 
> My study involves 5 groups (stimulation location) and 2 conditions (pre- and post-stimulation, paired) with a different number of subjects in each group. I would like to perform a multi-sensor analysis as well because I do not know at which sensors the effect will be observed. This would be for both time-series and time-frequency data. Given this information, what would be the ideal way to structure the "data" cell array?
> 
> Also, I'm a little unclear as to whether I should use the "std_stat" or the "statcondfieldtrip" function. I know I do not want the "statcond" function because I am looking to perform the cluster correction which is only in the fieldtrip function. Additionally, if someone could clarify the "fieldtripclusterparam" and "fieldtripchannelneighbor" (both in "std_stat") as well as the "neighbours" ("statcondfieldtrip") inputs.
> 
> Any assistance is greatly appreciated!
> 
> Thank you,
> Taylor
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu