Dear Helmut, just getting back to this thread, as I have a similar issue.
I have seen many indications about using ImCalc and something as the expression you suggested (i1+i2.*2+i3.*4), but I am not sure on how should I report it in the methods. Should I just say something like "we ran a conjunction analysis through ImCalc with the expression i1+i2.*2+i3.*4 using all contrasts p>0.05 FWE thresholded T maps"? Is there any reference for this method? How about a reference for the impossibility to of inference on between-subjects effects (to justify the ImCalc approach)?
One more thing: a conjunction analysis seems to me pretty conservative, so would it be acceptable to use uncorrected thresholds or not? Considering cluster thresholds, each contrast have different threshold (estimated through 3dClustSim), so is it OK to perform the analysis with these T maps, or should I save the T with the same cluster threshold?
Sorry for the many questions, but I appreciate any comments you and the others might have!