I am puzzled with the ROI analysis with FSL. Note that I have already had my ROI masks. 

1) does it mean that we must use Featquery to do ROI analysis? Does any one write scripts to do this work? I mean that we extract the ROI signal changes. 

My problem is that I only have one run for each subject. Hence for the FEAT, I only do the first level analysis, then the group analysis (higher level analysis). That it to say I do not have the standarized 2nd higher level analysis to combine the different runs. Because of this, it seems that I can not input the cope. feat to the Featquery. Could I input the feat directories (e.g., Subj1/Output.feat), rather than the individual Cope1.feat?  I guess so, but hope to confirm it. My understanding is that we must do the ROI analysis with each subject. 

2) why I can not click the "Do not binariese mask (allow weighting)? My FSL version is 5.0.9.

3) How to understand the output from featquery. Someone said that we need consider the "report. txt". I do not understand what do the output mean. For instance, what does the file (with 14 columns) below mean? Could I understand the output as the plot which I made below?

1 stats/cope1 96 -199.1 -87.43 -9.282 -1.656 71.7 192.4 66.82 26 28 12 -21.8 -16.4 -17.1