 |
 |
(1) No. You should never pass F-contrasts to the second-level models. (2) The F-contrast should refer to all sessions and you can use the replicate tool if all sessions have the same column order. (3) I'm not familiar with the "adjust for everything" contrast. In my gPPI toolbox, I generate an Omnibus Contrast for all tasks/PMs/derivative terms. You can take a look at the code used. The toolbox can be found at: http://www.nitrc.org/projects/gppi Best Regards, Donald McLaren, PhD On Tue, Jul 28, 2015 at 2:24 PM, Antonio Díaz Parra <[log in to unmask]> wrote: > Thanks Donald, it is more clear now. I have several more doubts regarding > F-contrasts and the extraction of time series: > > 1. Does it make sense to pass F-contrasts from first-level models to > second-level models in a similar way as t-contrasts? > > 2. When extracting time series with the VOI toolbox, why is the time > series from session 2 (or 3) corrected by setting an F-contrast only > referring to session 1? > > 3. When extracting time series with the VOI toolbox, what is the meaning > of "Adjust for everything"? I expected a noise signal but it was not the > case. > > Kindest regards, > > Antonio > >