Hello Dr. McLaren,

I was trying to implement gPPI on our data. So far it has been great.

In our data, we have 3 sessions, each session with a 2x2 design plus regressors for the time derivatives, motion parameters and outliers (from ART).

As you mentioned in the previous threads, I understand for P.Task I would only need to include the 4 regressors. Since they have the same names across sessions, I wrote:

P.Tasks={'1' 'hit_high' 'hit_low' 'miss_high' 'miss_low'}

Which worked out fine. But for some participants, they could miss one regressors in one session. If that happened, the regressor was omitted in our first level design for that session and first level contrasts were modified accordingly. For PPPI, do I need to set P.Task= {'0', ...} and manually add _run1 _run2 for the regressor names to generate the right contrasts? 

Thanks for the help!
LT

Yes. You just need to change 1 to 0. 0 means that the task does not need to
occur in every run.

Hi Dr. McLaren,

Thanks for the reply.

A follow up question about setting the task when including motion and outlier regressors (from ART) in this multiple session study. Since the number of outliers would be different each session, even if the tasks are the same, I should set Task=0 correct? 

LT

Thanks

You can set it to 0 or 1 as usual. The number of regressors added for bad
volumes and motion does not effect the task settings AND the task settings
don't effect the number of regressors. The number of regressors is always
the same as the 1st level task model.

Hope that helps.