Hi Gang,

Thank you for the suggestion to use 3dMVM - can I ask a follow-up question? I'm having trouble conceptualizing the labels and codes for this analysis, and there aren't a lot of examples to follow for fully within-subjects designs. The design again is a within-subjects 2(Picture: pos, neg) x 2(Text: good, bad), using their emotional scores (Ambiv) as a covariate.

This is what I have so far, any guidance would be greatly appreciated on how to proceed. Thank you!

3dMVM \
-prefix 2X2_Ambiv \
-jobs 4 \
-model 'Pic*Text+Ambiv' \
-wsVars "Pic*Text" \
-qVars "Ambiv" \
-qVarsCenters '5.34' \
-num_glt 6 \
-gltLabel 1 Pos_vs_Neg -gltCode 1 'Pic : 1*pos -1*neg Ambiv:' \
-gltLabel 2 Good_vs_Bad -gltCode 2 'Text: 1*good -1*bad Ambiv:' \
-gltLabel 3 Pos_vs_Neg
-gltLabel 4
-gltLabel 5
-gltLabel 6
-dataTable \
Subj Pic Text Ambiv InputFile \
sub1 pos good 4.51 /flash/norris/tsd1/09mar12cv/FUNCTIONAL/09mar12cv.results/PG_auc_mask+tlrc \
sub1 pos bad 4.51 /flash/norris/tsd1/09mar12cv/FUNCTIONAL/09mar12cv.results/PB_auc_mask+tlrc \
sub1 neg good 4.51 /flash/norris/tsd1/09mar12cv/FUNCTIONAL/09mar12cv.results/NG_auc_mask+tlrc \
sub1 neg bad 4.51 /flash/norris/tsd1/09mar12cv/FUNCTIONAL/09mar12cv.results/NB_auc_mask+tlrc \
.....