[FieldTrip] Cluster-based permutation analysis statfun ANCOVA

Stephen Politzer-Ahles stephen.politzer-ahles at ling-phil.ox.ac.uk
Fri May 27 09:31:45 CEST 2016


Hi list,

Eventually we managed to figure this out off-list, but I just wanted to
post a response here to show that it's been resolved (and in case someone
in the future has a similar issue and finds this thread).
We ended up handling this by making a difference-of-differences wave to
quantify the 2x2 within-subjects interaction (per
http://www.fieldtriptoolbox.org/faq/how_can_i_test_an_interaction_effect_using_cluster-based_permutation_tests)
and then using ft_statfun_indepsamplesregrT to correlate that interaction
effect with the subject-wise covariate (as opposed to comparing it to 0
with a regular t-test). For the cfg.design matrix we just used the list of
the subjects' values on the covariate. The results we got seemed to make
sense (i.e., they were consistent with what had been observed in the same
dataset with a time-window analysis before this), but of course if we've
done anything wrong or if anyone has suggestions for a better way to do
this then we'd be happy to hear.

Best,
Steve



---
Stephen Politzer-Ahles
University of Oxford
Language and Brain Lab
Faculty of Linguistics, Phonetics & Philology
http://users.ox.ac.uk/~cpgl0080/


Message: 1
> Date: Tue, 24 May 2016 08:57:46 +0000
> From: Alex Sel <alex.sel at psy.ox.ac.uk>
> To: "fieldtrip at science.ru.nl" <fieldtrip at science.ru.nl>
> Subject: [FieldTrip] Cluster-based permutation analysis statfun ANCOVA
> Message-ID:
>         <3123912FD054FE44A023073D2EFD213C8CA53F at MBX02.ad.oak.ox.ac.uk>
> Content-Type: text/plain; charset="us-ascii"
>
> Dear List,
>
>
>
> Apologies for the reposting...
>
>
>
> I am trying to perform a cluster-based permutation test. I have an
> experimental design with 2 within variables (2x2 design) and a covariate. I
> wonder how I can define the statistics in the script to include both the
> within-subject variables and the covariate. I am using the cfg.statistics
> = 'depsamplesFmultivariate' but I am unsure how to define the covariate in
> the model.
>
>
>
> Any insight on this would be much appreciated.
>
>
>
> Best wishes
>
>
>
> Alex Sel, PhD
> Postdoctoral Researcher
> Department of Experimental Psychology,
> University of Oxford,
> 9 South Parks Road,
> OX1 3UD
> Tel: 01865 271 340
> Email: Alex.sel at psy.ox.ac.uk
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20160527/90866f7c/attachment-0002.html>


More information about the fieldtrip mailing list