Has anyone run into issues with SPM8 sometimes truncating data at
2nd-level analysis (Results>Select SPM.mat>select Contrast, etc.) when
looking at different contrasts from an fMRI experiment? I use exactly
the same criteria each time, and this has happened multiples times
over multiple days on both Mac and Linux computers.
Typically, the clusters of both the truncated and full data sets are
exactly the same, at least halfway or 2/3 through the data. However,
the last few clusters are omitted in the truncated set.
It seems that either the truncated set of data or the full set of data
can randomly appear. Sometimes, I have found that if I switch
contrasts several times through the Contrasts menu within the Results
window, then the full data set will appear. However, this is not
always the case. I have thus far been unable to figure out a way to
manipulate this error consistently nor does it seem to be directly
dependent on Z score or number of clusters.
Does anybody have any insight?
Thanks for your help!