Hi,
What you are doing with the registration and template is fine - although
make sure you select 6 DOF and not 12 DOF for the registration to
your single-subject "template".
As for the predictor (or regressor or EV - all equivalent names) you
can have different numbers in different sessions (or subjects ) - our
framework is fine with that. Just make sure that you have
equivalent *contrasts* in each and that the contrasts are numbered
the same way.
All the best,
Mark
On 18 Oct 2010, at 18:53, Aditya Kumar Kasinadhuni wrote:
> Hi all,
>
> I have a question regarding FSL operating in the funtional
> space. I understand all the first level analyses are carried out in
> the subject's space and that it is during the higher level analysis
> that the data gets resampled to the standard space.
> Does this mean that if I am not going to run any higher level
> analysis and am interested in the data across the same subject for
> different runs in one session, I don't have to specify as standard
> space template in the registration tab of the FEAT gui? The way I am
> currently working is to use the brain extracted structural data
> collected during the same session as a substitute for the std space
> template and am leaving the "MainStructural" tab unselected.
>
> Can anyone please address this and tell me if what I am doing is
> correct?
>
> The second question I had was that pertaining to the predictors. I
> understand that the number of predictors in a first level analysis
> participating at a second level must be of the same number across
> all the subjects? If there are less predictors in one of them what
> do I do? I am thinking of putting a "0" for all time points for the
> extra predictor. Would this work? And if the number of predictors
> vary across runs for the same session, would this approach still be
> valid? Is it also necessary that all analyses that enter the second
> level have the same names for their predictors?
>
> I would be very obliged if someone could kindly answer these
> questions.
>
> Thank you.
>
|