Dear Serena,
This makes sense to try this phase if that's what you
are interested in testing.
You can actually replace your 1 column EV file with an
equivalent 3 column version.
For the one you sent me it is:
60.0000 2.5000 1.0000
72.5000 2.5000 1.0000
75.0000 2.5000 1.0000
97.5000 2.5000 1.0000
167.5000 2.5000 1.0000
205.0000 2.5000 1.0000
207.5000 2.5000 1.0000
As this contains the times (in seconds) of the onset of all the non-zero
points in the 1 column file and replaces them with a stimulus of
duration 2.5 seconds (equal to your TR). This will do the same as your
1 column EV.
So that is one way of getting around this bug, since you are right
that you can use a phase of -4 with the 3 column format just fine.
All the best,
Mark
On 5 Jul 2011, at 10:58, sere f wrote:
> Hi Mark,
>
>
> maybe to know that FEAT with double gamma phase =-4 in event related design (3 Colum format) runs, can be useful info.
>
> I have chosen phase= -4 because I have read that FSL double gamma has Glover paper parameters, and for this subject we want to try a early response.
> Do you think that 4 s is too much?
>
> Thanks for your help.
>
> Serena.
>
>
> On Mon, Jul 4, 2011 at 6:31 PM, Mark Jenkinson <[log in to unmask]> wrote:
> Dear Serena,
>
> The problem is nothing to do with your EV file
> but due instead to the phase setting for the
> Double-Gamma HRF. It seems that Phase
> values of -1.25 (= -TR/2) or less cause this
> problem, which looks like a bug in FEAT.
>
> In the meantime you could either use a
> different convolution (e.g. Gamma) or a
> different Phase value. Is there any reason
> you are trying to use such a large value?
>
> We will hopefully be able to fix this and get
> it patched soon, but I cannot be sure when.
>
> All the best,
> Mark
>
>
> On 4 Jul 2011, at 17:10, sere f wrote:
>
> > Dear Mark,
> > the reference number for fsf file is: 964257
> >
> > thanks a lot.
> >
> > Serena.
> >
> >
> > On Mon, Jul 4, 2011 at 3:33 PM, Mark Jenkinson <[log in to unmask]> wrote:
> > Dear Serena,
> >
> > I still cannot replicate your problems.
> > If you press the "Save" button in the FEAT GUI then
> > you should be able to save a design.fsf file even
> > if there are errors. Please try and do this and send
> > us this file as I need to see if I can replicate your
> > error. Alternatively, if this won't work, please describe
> > exactly what selections you are making in the FEAT GUI.
> >
> > All the best,
> > Mark
> >
> >
> > On 4 Jul 2011, at 13:30, sere f wrote:
> >
> > >
> > > Dear Mark,
> > >
> > > The reference number for my fMRI file is 553096
> > > I have not design.fsf because FEAT does not run.
> > >
> > > Thanks,
> > >
> > > Serena.
> > >
> > > On Mon, Jul 4, 2011 at 2:20 PM, Mark Jenkinson <[log in to unmask]> wrote:
> > > Dear Serena,
> > >
> > > This looks fine to me and I cannot replicate your
> > > problem.
> > >
> > > Can you also upload your data and design.fsf file.
> > >
> > > All the best,
> > > Mark
> > >
> > >
> > > On 4 Jul 2011, at 12:21, sere f wrote:
> > >
> > > > Hi Mark,
> > > > the reference number for my txt is: 888901
> > > > Thanks for your help.
> > > >
> > > > Serena.
> > > >
> > > > On Mon, Jul 4, 2011 at 12:59 PM, Mark Jenkinson <[log in to unmask]> wrote:
> > > > Dear Serena,
> > > >
> > > > Can you send us the file you are using as your 1 column EV?
> > > >
> > > > All the best,
> > > > Mark
> > > >
> > > >
> > > > On 4 Jul 2011, at 11:46, sere f wrote:
> > > >
> > > > > Dear all,
> > > > >
> > > > > I run feat and I have an error in Stat window.
> > > > > I select "Custom (1 entry per volume)" and a Doulble-Gamma HRF.
> > > > > I choose my contrast and when I click "Done " I obtain this message:
> > > > >
> > > > > problem with processing the model: terminate after called throwing an instance of 'NEWMAT::ConvergenceException"
> > > > >
> > > > > What is the problem?
> > > > >
> > > > > thanks for the help.
> > > > >
> > > > > Serena
> > > >
> > >
> >
>
|