Dear Freesurfer experts,<br><br>Sorry for the inconvenience but I posted  this message last week, but I did not receive any answer. I really need to keep  going on with my analyses.<br><br><font face="'PrimaSans BT,Verdana,sans-serif'"><br>I have runned the following command line:<br>mri_glmfit   --surf PLI_MCI_sample7_avg lh --C  /root/trabajo/freesurfer/subjects/PLI_MCI_sample7_avg/stats_MRI_PET_Biomarkers/glm_files/contrast_withinE4_neg.mat   --fsgd  /root/trabajo/freesurfer/subjects/PLI_MCI_sample7_avg/stats_MRI_PET_Biomarkers/glm_files/lh.cortex_AVGthickness_APOB_fsgd.txt   dods --label  /root/trabajo/freesurfer/subjects/PLI_MCI_sample7_avg/label/lh.cortex.label   --y  /root/trabajo/freesurfer/subjects/PLI_MCI_sample7_avg/stats_MRI_PET_Biomarkers/lh.thickness_s12mm.mgh   --no-prune --glmdir  /root/trabajo/freesurfer/subjects/PLI_MCI_sample7_avg/stats_MRI_PET_Biomarkers/resultsthickness_cortex_s12mm_BiomrkrsWithin/lh.cortex_AVGthickness_APOB_contrast_withinE4_neg/&nbsp;<br>&nbsp;&nbsp;<br></font>which is always prompts the following error:<br><br><font face="'PrimaSans BT,Verdana,sans-serif'">--------------------------------<br>ERROR: matrix is ill-conditioned or badly scaled, condno = 1.75146e+07<br>--------------------------------<br>Possible problem with experimental design<br><br><br>I'm sending attached the fsgd.txt and contras.mat files used in case you need them<br><br><br>Does anyone knows what's wrong with my design?<br><br><br>Many thanks in advanced,<br>Gabriel.<br></font>