Stimulus Timing Files and GLTSYM

AFNI Message Board
1

Hi AFNI,
I am running a deconvolution script via afniproc.py with a task that has many condition types (24 conditions). I realize there are possibilities in which this would result in multicollinearity in my matrix, however, another issue is that in the AFNI GUI there are so many trial types (and GLTSYMS) that I am unable to even properly view the stats file output (…they T-stats and betas go outside the viewer window).

One option is to simplify my design matrix and concatenate stimulus timing files to create a simpler task, however, is there an option in afniproc.py that reorganizes or even suppressed some of the output in the stats file that I am not interested in?

I am interested in the stats I mention in the GLTSYM lines, but AFNI also calculates betas and T-statistics for each individual trial type which doesn’t allow me to view anything else.

Any thoughts here would be greatly appreciated and if it helpful to supply my afniproc.py script out or screenshot of the stimulus timing files I am happy to!

thanks in advance,
Steve

2

Hi Steve,

There are 2 things you can do. One is to add “-regress_fout no” to afni_proc.py, so all of the F-stats are not output. They are not so necessary anyway.

But the more simple thing to do it with the afni GUI. When choosing “OLay” or “Thr” volumes, put your mouse over the text to the left of the button (e.g. put the mouse cursor directly over ‘OLay’) and right-click. That should give a scrollable menu that is much easier to deal with.

  • rick