Dear Stephen M. Smith and colleagues,
I recently confronted myself with the same problem that motivated Anda van
Stegeren and Kerstin Preuschoff to write to the list earlier this year: I
was not able to make featquery (working under Windows XP/Cygwin) to run
properly either using a mask or by trying to define standard space
Shortly, what I did was to create a mask with MRIcro using the
example_func.img, or alternatively a high resolution image in standard
space. Within both cases, I received a message stating that the mask size
did not match any of the images in the FEAT directory. Just to ensure that
I was wasn't using different dimension-sized images, I then used avwhd and
fine tuned the images (the mask and its matching image) to be exactly the
same. Yet, the same error remained.
Suspecting that this could be due to a Windows/Cygwin version bug, I then
tried to run the same analysis, i.e., to use exactly the same masks, this
time using Linux (Suse9.1). And it worked uneventfully.
Therefore it seems that this problem with featquery appears to be
restricted to the Windows version of FSL.
Div of Psychiatry and
School of Psychology
University of Nottingham, NG7 2RD
Tel: +44 (0) 115 - 95 -15308