 |
 |
Hi - I don't think your choices of masking make a lot of sense - maybe have another look at the masking text in the manual, and if you're still confused, let us know what question you wanted to be asking via masking? Cheers. On 19 Apr 2012, at 13:23, Andres Roman wrote: > HI FSL list, > > I am doing a two group comparison using the usual design matrix of 2 EVs and I am interested to study activation and deactivation of both EVs. > > The contrasts I am using are these: > Mean Activation EV1: c1= (1,0) > Deactivation EV1: c2= (-1,0) > Activation EV2: c3=(0,1) > Deactivation EV2: c4=(0,-1) > > My interest is to ask whether there are differences in deactivation and activation between two groups. For this I used this logic: > > -Activation EV1 > Activation EV2 > but the contrast c5= (1;-1) points out where the activation of EV1 is bigger than activation of EV2 and deactivation of EV2. > > -Deactivation EV1 > Deactivation EV2 > the c6=(-1;1) shows the opposite of c5. > > Im trying therefore to mask this using this logic: > > To model this I used the following assumption. > ( Activation EV1 > Activation EV2 ) where I masked Contrast 1 with c3 > > And then this: > ( Deactivation EV1 > Deactivation EV2 ) where I masked real Contrast 2 with c4 > > I used the "Z>0" option on the GUI because according to the Feat manual its the solution when you would like to further threshold a given Z statistic image by masking it with non-zeroed voxels from other contrasts. . > > The issue is that when I masked c1 (1,0) with c3 (0,1) the results I get are identical to c1. Same goes for c2 (-1,0) when masked with c4 (0,-1) where the result is identical to c2. I was wondering why this could be? Do you guys think its because I decided the threshold Z>0? Is my model right? I also wondered why is it that if the voxels which passed thresholding in the contrast of interest are only those which also survived thresholding, why is it that I get the same results even though I’m contrast masking? > > > Thank you very much for everything > Best wishes > > > -- > Andres --------------------------------------------------------------------------- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ---------------------------------------------------------------------------