Hi - I think that what's happening with zfstat1 is that just a few of the FIR basis function EVs are *just about* reaching significance, and that the f-test across all of them is not quite.     With zfstat2 you have a lot more going on in the individual EVs, as well as the fstat, so that's straightforward.

It seems possible that the first two or three FIRs are reflecting subject motion rather than neural activity?   If you also try modelling with a more standard HRF the results may be easier to interpret….

Cheers.




On 21 Mar 2012, at 17:48, sere f wrote:



Dear Stephen,

thanks to help me!

I have uploaded the feat directory and the ID is: 233958

I can not change the design matrix because the events in custom (3 column format) files (they are in EV folder)  are the epileptic events of a patient.

I am looking forward to reading news from you.

Thank you very much.

Serena,

On Wed, Mar 21, 2012 at 2:10 PM, Stephen Smith <[log in to unmask]> wrote:
Hi - this does indeed sound surprising - please check carefully your design matrix and contrast setup, in particular the way the F-stat contrast is setup.

If you still can't see what's wrong, feel free to upload a tar file of the full FEAT output directory and we can take a look.
Please upload the files in a single compressed tarfile to
And then email us the upload ID.

Cheers.



On 20 Mar 2012, at 00:46, sere f wrote:

Dear FSL users,


I have a question about F-test applied to FIR basis functions.
I utilize 12 FIR to model 2 EVs so in total there are 24 FIR.
The zfstat map of the 12 FIR of the first EV does not show activation whereas the zstat maps of almost all 12 FIR show activation.

How can I interpreter this result?
any advice?

Thanks  

Serena 


---------------------------------------------------------------------------
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
---------------------------------------------------------------------------






---------------------------------------------------------------------------
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
---------------------------------------------------------------------------