Hi all!
I'm sure I am missing something really basic here but I have racked my brains and cannot come up with the problem.
I am trying to run 3dROIstats using a mask of some ROIs that I extracted from my data on some mean irf files. I have done this type of analysis many times with no problem but this time I keep getting the message:

"Error: cannot deal with float-valued mask dataset
(consider the -mask_f2short option)"

What I did was I used 3dTstat to get the mean of the irfs that were generated by my deconvolve script and I warped these means into talairach space. Then I ran a t-test versus 0 on all my subjects and thresholded and clust_ordered the data which resulted in a map with a number of ordered ROIs (an intensity map [fim]). I then tried to use this clustordered map as a mask to overlay on my original mean irf maps in each subject.

Is there something I am missing here? I have tried to run 3dROIstats on another similar map that I made a few months ago and the program is working fine with that. I also double-checked the clustordered maps to check that I have a whole-number digit representing each cluster, which is the case and I have also check that my mean irf original files contain one number per voxel representing the mean, which they do.

Here is my 3dROIstats command:
3dROIstats -mask clstrd.ROIs+tlrc *.mean.irf+tlrc.HEAD > ROIs

Thank you!
Catherine