Hi Sheena,
Your 6th contrast is not a differential contrast - it is just the sum of EV2 and EV3, which will be significant across all voxels - it looks like there is a minus sign missing maybe?
Kind Regards
Matthew
> On 24 Jun 2016, at 23:09, Sheena Sharma <[log in to unmask]> wrote:
>
> Hi all,
>
> I have just run VBM analysis and received a very significant result. When I run randomise as such, the whole brain lights up Yellow:
> fslview template_GM -b .1,1 tmp_tfce_corrp_tstat6 -l Red-Yellow -b 0.99,1
>
> While I would love to think this is a real finding, I am a bit suspicious that I've done something wrong as it seems the WHOLE BRAIN is different between groups. I followed the instructions in the FSL user guide and practicals so I'm not sure what has gone wrong.
>
> The steps I took are as follows:
> 1) Copied design files into VBM folder.
> 2) 'cd' into VBM folder and then run:
> fslvbm_1_bet –b
> fslvbm_2_template -n
> fslvbm_3_proc
>
> The design files are attached.
> <Screen Shot 2016-06-25 at 10.05.15 AM.png>
> <Screen Shot 2016-06-25 at 10.05.21 AM.png>
>
>
> Because VBM processes the data in alphabetical order, my design files aren't grouped nicely (in 3 chunks) and rather subjects from the different groups are all mixed in, but I don't think this should matter. Does anybody have any ideas here? Thanks you SO much in advance. I have been wracking my brain!
>
> Thanks
> Sheena
> <design_alpha.fsf>
|