Hi - if you know what needs to go into your designs (and resulting
design.fsf files) then you should be able to script this - taking a
template design.fsf and amending it for each new subject/analysis,
taking into account what low-level images/EVs are non-zero.
Cheers.
On 19 May 2009, at 18:23, Jessica Kirkland wrote:
> Thanks Steve,
>
> That's what I've done (thanks too Martin!), but now I run into the
> issue that Heather Urry posted about last week (the analysis fails
> at the post-stats, due to (I think) any time I've used an "empty"
> EV). Is there any way to get around this? I'm asking because I have
> 70 subjects, and I am already customizing each run-level .fsf with
> specific empty EVs for all 3 runs, and now will have to customize
> each .fsf file again for each subject at the 2nd level, reflecting
> these empty EVs. I can definitely go that route, but want to make
> sure I am not missing out on any more automated work-arounds before
> I proceed with this path.
>
> thank you,
>
> Jessica
>
> On Tue, May 19, 2009 at 5:11 AM, Steve Smith <[log in to unmask]>
> wrote:
> Hi - I would think that you should be able to setup a second-level
> FE within-subject model that takes in all first-level COPEs
> (possibly using the "inputs are copes" input type option, rather
> than the "inputs are FEAT directories" options). You then just need
> one EV per condition (eg one for face and one for object) in this
> second-level model.
>
> Cheers.
>
>
>
> On 18 May 2009, at 21:01, Jessica Kirkland wrote:
>
> Hi,
>
> I am currently analyzing data for 70 subjects, 3 runs each. I would
> like to set up a 3-level analysis like this: level 1 = run, level 2
> = subject, level 3 = group. My problem is that each subject has 2
> runs containing face stimuli, and one run containing object stimuli--
> so I do not want to average all 3 at the subject level.
>
> I am unsure of how to make my 2nd level fixed-effects analysis take
> this into account (e.g., I could do a 2nd level combining across the
> face runs, but what about the object run?). Is there a way to do
> this at the second level without averaging across all 3 runs? I
> tried to get around the issue by entering all 3 runs' feat
> directories, and then setting up 2 contrasts, one combining the 2
> face runs and 1 representing the object run, but Feat crashed
> because the runs have different #s of copes.
>
> Alternatively, could I run two separate 2nd-level fixed effects
> analyses--one for face runs and one for the object run? If I am
> understanding correctly, it would not be appropriate to input all
> runs separately into a higher-level ME Flame analysis, or to input
> some 1st (object runs) and some 2nd (face runs) level data at the
> 3rd level. In any case it seems that I may run into the different #
> of copes issue at the 3rd level as well...
>
> Any suggestions on how to proceed would be greatly appreciated.
>
> thank you,
>
> Jessica
>
>
> ---------------------------------------------------------------------------
> Stephen M. Smith, Professor of Biomedical Engineering
> Associate Director, Oxford University FMRIB Centre
>
> FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
> +44 (0) 1865 222726 (fax 222717)
> [log in to unmask] http://www.fmrib.ox.ac.uk/~steve
> ---------------------------------------------------------------------------
>
---------------------------------------------------------------------------
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director, Oxford University FMRIB Centre
FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
---------------------------------------------------------------------------
|