Warning Message 3dMVM

AFNI Message Board
1

Hello,

I ran 3dMVM (code below), and although I got an output I received the following warning message and I am not sure what it means and whether I need to do something. Any suggestions or comments would be greatly appreciated!

Thank you very much,
Tam

3dMVM output:

++ Smallest FDR q [0 (Intercept) F] = 6.39594e-13
*+ WARNING: Smallest FDR q [1 group F] = 0.702769 ==> few true single voxel detections
*+ WARNING: Smallest FDR q [2 instruct F] = 0.951091 ==> few true single voxel detections
*+ WARNING: Smallest FDR q [3 group:instruct F] = 0.907502 ==> few true single voxel detections
++ Smallest FDR q [4 valence F] = 5.07365e-07
*+ WARNING: Smallest FDR q [5 group:valence F] = 0.859463 ==> few true single voxel detections
*+ WARNING: Smallest FDR q [6 instruct:valence F] = 0.936071 ==> few true single voxel detections
*+ WARNING: Smallest FDR q [7 group:instruct:valence F] = 0.836563 ==> few true single voxel detections
*+ WARNING: Smallest FDR q [9 3way_inter t] = 0.888437 ==> few true single voxel detections
*+ WARNING: Smallest FDR q [11 3way_inter t] = 0.843422 ==> few true single voxel detections
*+ WARNING: Smallest FDR q [13 3way_inter t] = 0.737256 ==> few true single voxel detections
*+ WARNING: Smallest FDR q [15 group_instruct t] = 0.873187 ==> few true single voxel detections
*+ WARNING: Smallest FDR q [17 group_instruct t] = 0.841724 ==> few true single voxel detections
*+ WARNING: Smallest FDR q [19 group_instruct t] = 0.955704 ==> few true single voxel detections
*+ WARNING: Smallest FDR q [21 group_valence t] = 0.859528 ==> few true single voxel detections

Congratulations! You have got an output group+tlrc.

Warning message:
In summary.Anova.mlm(object$Anova, multivariate = FALSE) :
HF eps > 1 treated as 1

3dMVM Script:
3dMVM -prefix group -jobs 4
-bsVars group
-wsVars "instructvalence"
-SS_type 3
-num_glt 7
-gltLabel 1 3way_inter -gltCode 1 'group : 1pat -1ctrl instruct : 1feel -1Dfeel valence : 1pst -1neg’
-gltLabel 2 3way_inter -gltCode 2 'group : 1pat -1ctrl instruct : 1feel -1nat valence : 1pst -1neg’
-gltLabel 3 3way_inter -gltCode 3 'group : 1pat -1ctrl instruct : 1Dfeel -1nat valence : 1pst -1neg’
-gltLabel 4 group_instruct -gltCode 4 'group : 1pat -1ctrl instruct: 1feel -1Dfeel’
-gltLabel 5 group_instruct -gltCode 5 'group : 1pat -1ctrl instruct: 1feel -1nat’
-gltLabel 6 group_instruct -gltCode 6 'group : 1pat -1ctrl instruct: 1Dfeel -1nat’
-gltLabel 7 group_valence -gltCode 7 'group : 1pat -1ctrl valence: 1pst -1*neg’
-dataTable
Subj group instruct valence InputFile
1422 pat feel pst 1422/1422.results/Feel_Happy_Resp+tlrc
1422 pat feel neg 1422/1422.results/Feel_Bad_Resp+tlrc
…

2

Those warnings about FDR simply indicate that the associated statistical evidence is weak in terms of the false discovery rate. In other words, that is just how much information your data could provide based on the current model.

You can also ignore the following message (a numerical indicator for some statistical construct at some voxel(s)).

Warning message:
In summary.Anova.mlm(object$Anova, multivariate = FALSE) : HF eps > 1 treated as 1