I think I have a fix for this if you want to edit your version of
fast_selxavg3.m
Change this line:
if(~isempty(analysis) & DoGLMFit & strcmp(flac.designtype,'event-related'))
To:
% Check to make sure that each task ev as the same number
% of regressors before saving the h.dat file
evtaskind =
On 05/04/2016 04:08 PM, S.V.Shepherd [work] wrote:
> I can /also /confirm that it ran fine when the big matrix was changed
> from taskreg to nuisreg, so we can probably consider the issue
> resolved. I suspect the problem is with multi-column taskregs. If I
> want to have the regressors track
I can *also *confirm that it ran fine when the big matrix was changed from
taskreg to nuisreg, so we can probably consider the issue resolved. I
suspect the problem is with multi-column taskregs. If I want to have the
regressors tracked equivalently across runs, I'd need to add them as
taskregs, o
I can confirm that having taskregs with a paradigm works. For example, I've
run successfully analyzed the following:
mkanalysis-sess -native -event-related -refeventdur 10 -paradigm
aboutFace.para -nconditions 13 -TR 2. -rlf rlf_aboutFace_subjvid
-funcstem fm2tas1 -mask brain -analysi
I don't think it is going to work to have both a -taskreg and a
paradigm. You should be able to model it as a nuisance regressor. The
fact that it may be the same across runs won't make much of a difference.
doug
On 05/04/2016 03:33 PM, S.V.Shepherd [work] wrote:
> mkanalysis-sess -native -eve
mkanalysis-sess -native -event-related -refeventdur 10 -paradigm
aboutFace.para -nconditions 13 -TR 2. -rlf rlf_aboutFace_subjvid
-funcstem fm2tas1 -mask brain -analysis
a_401_Av_s_task_fmov_bgrnd_icadenoise -taskreg taskreg_face__zmove.dat 1
-taskreg taskreg_face__bgrnd.dat 1 -taskr
Can you send your mkanalysis-sess command? Can you also send one of the
regressor files? And can you send the full selxavg3 terminal output?
On 05/04/2016 02:46 PM, S.V.Shepherd [work] wrote:
> I've modified an analysis to include a vector of noise time courses
> estimated using ICA. Since IC ef
