yes, that should work. make sure to use DOSS.

On 11/23/2022 11:50 AM, Olivier Brown wrote:

External Email - Use Caution

Dear FreeSurfer experts,

I would like to confirm my FSGD file format and mtx contrasts prior to continuing with my analyses. Here is some background information: I have two groups (treatment, control) and would like to determine whether our quantitative scale, CDRISC, is associated with thickness within each group. I would like to control for sex, handedness, and age. Seeing as sex and handedness are categorical, I entered them as Classes instead of variables (Controlleftmale was not included, as no participant fell in that group).

GroupDescriptorFile 1                           
Title CDRISC                              
Class Controlrightfemale                           
Class Controlleftfemale   
Class Treatrightfemale
Class Treatleftfemale
Class Controlrightmale                         
Class Treatrightmale                             
Class Treatrightmale                                        
Variables   CDRISC_TREATMENT   CDRISC_CONTROL   age         
Input sub-1     Controlrightfemale  0     19    10.66
Input sub-2     Controlleftfemale   0     26    14
Input sub-3     Treatrightmale      35     0    14.02
Input sub-4     Treatleftmale       35     0    10.16
[...]

My contrasts (DOSS) are as follows:

[Nregressors = Nclasses + Nvariables Nregressors = 7 + 3 Nregressors = 10]

Treatment CDRISC slope = 0 0 0 0 0 0 0 1 0 0 Control CDRISC slope = 0 0 0 0 0 0 0 0 1 0

Are there any glaring issues with this model? The only extra step of consideration here is that I have yet to demean the data. Any guidance is appreciated.

Warm regards, Olivier Brown

---

Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer