[FieldTrip] Why the general use of parametric test as observed test statistic for permutation cluster tests?

[Stephen Politzer-Ahles]

Hi Peter,

As far as I know this is just a default coded in fieldtrip, but can
easily be replaced (by using your own statfun rather than e.g.
depsamplesF, depsamplesT, etc.). According to Maris & Oostenveld
(2007), the observed statistic used to calculate clusters can be
anything, if I recall correctly.

Best,
Steve

> Message: 2
> Date: Wed, 15 Apr 2015 00:06:33 +0000
> From: Peter Goodin <pgoodin at swin.edu.au>
> To: "fieldtrip at science.ru.nl" <fieldtrip at science.ru.nl>
> Subject: [FieldTrip] Why the general use of parametric test as
> 	observed test statistic for permutation cluster tests?
> Message-ID:
> 	<A55BBE5A90EFC441A6FE147DF0DFE26E890F9252 at gsp-ex03.ds.swin.edu.au>
> Content-Type: text/plain; charset="us-ascii"
>
> Hi Fieldtrip list,
>
> I'm just wondering if someone can explain if there is a reasoning behind the
> general use of a parametric test to calculate the initial step in the
> permutation cluster test and not a non-parametric test (e.g t-test over Mann
> Whitey U)? Is it just due to the assumption of a Gaussian distribution or is
> there something deeper going on?
>
> Thanks,
>
> Peter
>
> _______________________
> Peter Goodin,
> BSc (Hons), Ph.D Candidate (submitted).
>
> Brain and Psychological Sciences Research Centre (BPsych)
> Swinburne University,
> Hawthorn, Vic, 3122
> http://www.swinburne.edu.au/swinburneresearchers/index.php?fuseaction=profile&pid=4149
>
> Monash Alfred Psychiatry Research Centre (MAPrc)
> Level 4, 607 St Kilda Road,
> Melbourne 3004
>


Stephen Politzer-Ahles
New York University, Abu Dhabi
Neuroscience of Language Lab
http://www.nyu.edu/projects/politzer-ahles/