[FieldTrip] Conjunction analysis

[andrea brovelli]

Last comment on Conjunction analysis:
I got an email from Thomas Nichols (2005, Neuroimage). In fact, things are much easier than expected, no special code is required. The conjunction is just the minimum of the two statistics (t-valeus, for example, or, equivalently, the maximum of the two p-values).  And this is independent of the type of statistical test used  (parametric or non-parametric). So, the implementation in ft_conjunctionanalysis should be safe.
Bye
Andrea

andrea brovelli <andrea.brovelli at univ-amu.fr> wrote:Thanks a lot for your reply, Arjen. It's clearer now.
In fact, I am using a parametric tests, so I think I would need to use the Nichols 2005 method for the conjunction analysis.
Does anyone have the code to do this or does anyone know how to recover it within SPM ?
Thanks a lot
Andrea

"A.stolk" <a.stolk8 at gmail.com> wrote:Hi andrea, ft_conjunctionanalysis performs a logical AND on 2 or more contrasts, of which each contrast may be corrected for multiple comparisons. It thus shows the overlap between those nonparametric statistical maps where a voxel survives the conjunction if it survived two or more independent nonparametric tests. It differs from the spm version which is grounded in parametric testing (nichols 2005). In the latter one could find overlap between clusters that is not necessarily statistically significant at the clusterlevel. That is because the parametric method is geared toward the (fixed) intensity of the cluster (and overlaps). Those are the key differences. Hope this helps a bit. Yours; arjen
andrea brovelli <andrea.brovelli at univ-amu.fr> schreef:
Dear all,

I need to do a conjunction analysis between contrast maps (T-maps or thresholded p-maps).

I have seen that there is a function called ft_conjunctionanalysis. Does anyone know what this function does ? I cannot find the reference for the method implemented.

Or does anyone have the matlab code to perform a conjunction analysis as described in Nichols et al., 2005 Neuroimage ?

http://psych.colorado.edu/~tor/Papers/Nichols_2005_conjunctions.pdf

Thanks a lot

Andrea
_______________________________________________
fieldtrip mailing list
fieldtrip at donders.ru.nl
http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
 _______________________________________________
fieldtrip mailing list
fieldtrip at donders.ru.nl
http://mailman.science.ru.nl/mailman/listinfo/fieldtrip_______________________________________________
fieldtrip mailing list
fieldtrip at donders.ru.nl
http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20131122/230ba447/attachment-0002.html>