3dLMEr model creation

AFNI Message Board
1

Hello! I'm trying to specify a model in 3dLMEr and am getting the following error:
** Error:
Incorrect level coding in variable Reputation : 2

I am building this off of the example from the help page (example 2) https://afni.nimh.nih.gov/pub/dist/doc/program_help/3dLMEr.html

And have pasted my code below. I would appreciate any help! Thank you so so much!

Example 2 --- LME analysis for one group of subjects each of which has
three effects associated with three emotions (pos, neg and neu), and the
effects of interest are the comparisons among the three emotions at the
population level. In addition, reaction time (RT) is available per emotion
from each subject. An LME model can be formulated to include both random
intercept and random slope. Be careful about the centering issue about any
quantitative variable: you have to decide which makes more sense - global
centering or within-condition (or within-group) centering?

3dLMEr -prefix LME -jobs 12                   \
      -mask myMask+tlrc                       \
      -model  'emotion*RT+(RT|Subj)'          \
      -SS_type 3                              \
      -bounds -2 2                            \
      -qVars  'RT'                            \
      -qVarCenters 0                          \
      -gltCode pos      'emotion : 1*pos'                       \
      -gltCode neg      'emotion : 1*neg'                       \
      -gltCode neu      'emotion : 1*neu'                       \
      -gltCode pos-neg  'emotion : 1*pos -1*neg'                \
      -gltCode pos-neu  'emotion : 1*pos -1*neu'                \
      -gltCode neg-neu  'emotion : 1*neg -1*neu'                \
      -gltCode em-eff1  'emotion : 0.5*pos +0.5*neg -1*neu'     \
      -glfCode em-eff2  'emotion : 1*pos -1*neg & 1*pos -1*neu' \

AFNI version info (afni -ver): Precompiled binary linux_openmp_64: Jun 23 2023 (Version AFNI_23.1.09 'Publius Helvius Pertinax')

3dLMEr 																								                        \
 -prefix 3dLME.Antx3dLMErDemeanRepAgg.n44.102223										                                    \
 -jobs 30																													\
 -model 'Reputation*Aggression+(Aggression|Subj)'																			\
 -qVars 'Aggression'																										\
 -qVarCenters 0                                                                                                             \
 -num_glt 12																	                                            \
 -gltLabel 1 Ant.Mean 					        -gltCode 1 'Reputation : +1*Mean'			                                \
 -gltLabel 2 Ant.Nice 					        -gltCode 2 'Reputation : +1*Nice'			                                \
 -gltLabel 3 Ant.Unp				            -gltCode 3 'Reputation : +1*Unp'					                        \
 -gltLabel 4 Ant.Mean-Nice 					    -gltCode 4 'Reputation : +1*Mean -1*Nice'					                \
 -gltLabel 5 Ant.Mean-Unp 					    -gltCode 5 'Reputation : +1*Mean -1*Unp'							        \
 -gltLabel 6 Ant.Unp-Nice				        -gltCode 6 'Reputation : +1*Unp -1*Nice'					                \
 -gltLabel 7 Ant.MeanxAgg 					    -gltCode 7 'Reputation : +1*Mean Aggression : '			                    \
 -gltLabel 8 Ant.NicexAgg 					    -gltCode 8 'Reputation : +1*Nice Aggression : '			                    \
 -gltLabel 9 Ant.UnpxAgg				        -gltCode 9 'Reputation : +1*Unp Aggression : '					            \
 -gltLabel 10 Ant.Mean-NicexAgg 				-gltCode 10 'Reputation : +1*Mean -1*Nice Aggression : '			        \
 -gltLabel 11 Ant.Mean-UnpxAgg 					-gltCode 11 'Reputation : +1*Mean -1*Unp Aggression : '			            \
 -gltLabel 12 Ant.Unp-NicexAgg				    -gltCode 12 'Reputation : +1*Unp -1*Nice Aggression : '					    \
-dataTable																													\
Subj        Reputation 	   Aggression   InputFile				     														\
s305        Mean	       1.31   $subjdata/s305/s305.results.SEATReputation.1mm/stats.s305+tlrc'[Ant.Mean#0_Coef]'       \
s305        Nice	       0.21   $subjdata/s305/s305.results.SEATReputation.1mm/stats.s305+tlrc'[Ant.Nice#0_Coef]'       \
s305        Unp	           0.77   $subjdata/s305/s305.results.SEATReputation.1mm/stats.s305+tlrc'[Ant.Unp#0_Coef]'        \
s318        Mean	       1.54   $subjdata/s318/s318.results.SEATReputation.1mm/stats.s318+tlrc'[Ant.Mean#0_Coef]'       \
s318        Nice	      -0.05   $subjdata/s318/s318.results.SEATReputation.1mm/stats.s318+tlrc'[Ant.Nice#0_Coef]'       \
s318        Unp	           0.37   $subjdata/s318/s318.results.SEATReputation.1mm/stats.s318+tlrc'[Ant.Unp#0_Coef]'        \
2

The specification for contrasts in 3dLMEr is slightly different from 3dMVM/3dLME. Replace the following

 -num_glt 12																	                                            \
 -gltLabel 1 Ant.Mean 					        -gltCode 1 'Reputation : +1*Mean'			                                \
 -gltLabel 2 Ant.Nice 					        -gltCode 2 'Reputation : +1*Nice'			                                \
 -gltLabel 3 Ant.Unp				            -gltCode 3 'Reputation : +1*Unp'					                        \
 -gltLabel 4 Ant.Mean-Nice 					    -gltCode 4 'Reputation : +1*Mean -1*Nice'					                \
 -gltLabel 5 Ant.Mean-Unp 					    -gltCode 5 'Reputation : +1*Mean -1*Unp'							        \
 -gltLabel 6 Ant.Unp-Nice				        -gltCode 6 'Reputation : +1*Unp -1*Nice'					                \
 -gltLabel 7 Ant.MeanxAgg 					    -gltCode 7 'Reputation : +1*Mean Aggression : '			                    \
 -gltLabel 8 Ant.NicexAgg 					    -gltCode 8 'Reputation : +1*Nice Aggression : '			                    \
 -gltLabel 9 Ant.UnpxAgg				        -gltCode 9 'Reputation : +1*Unp Aggression : '					            \
 -gltLabel 10 Ant.Mean-NicexAgg 				-gltCode 10 'Reputation : +1*Mean -1*Nice Aggression : '			        \
 -gltLabel 11 Ant.Mean-UnpxAgg 					-gltCode 11 'Reputation : +1*Mean -1*Unp Aggression : '			            \
 -gltLabel 12 Ant.Unp-NicexAgg				    -gltCode 12 'Reputation : +1*Unp -1*Nice Aggression : '					    \

with

...

 -gltCode Ant.Mean 		 'Reputation : +1*Mean'                       \
 -gltCode Ant.Nice 		 'Reputation : +1*Nice'                       \ 
 -gltCode Ant.Unp		     'Reputation : +1*Unp'                        \
 -gltCode Ant.Mean-Nice     'Reputation : +1*Mean -1*Nice'               \              
 -gltCode Ant.Mean-Unp      'Reputation : +1*Mean -1*Unp'                \
 -gltCode Ant.Unp-Nice	     'Reputation : +1*Unp -1*Nice'                \
 -gltCode Ant.MeanxAgg      'Reputation : +1*Mean Aggression : '         \ 
 -gltCode Ant.NicexAgg 	 'Reputation : +1*Nice Aggression : '         \
 -gltCode Ant.UnpxAgg	     'Reputation : +1*Unp Aggression : '          \
 -gltCode Ant.Mean-NicexAgg 'Reputation : +1*Mean -1*Nice Aggression : ' \
 -gltCode Ant.Mean-UnpxAgg  'Reputation : +1*Mean -1*Unp Aggression : '  \
 -gltCode Ant.Unp-NicexAgg	 'Reputation : +1*Unp -1*Nice Aggression : '  \

Gang

3

Oh wow thank you so much for catching that! Will give this a try!