Hi - that should be fine.
Cheers.
On 8 Jul 2009, at 11:48, Andreas Bartsch wrote:
> Hi Steve,
>
> thanks a lot. What about if the number of contrasts and EVs are
> different? I.e., in our models the EV of interest and the related
> contrast is essentially the same across sessions but the total
> number of EVs does vary...
> Cheers:)-
> Andreas
> ________________________________________
> Von: FSL - FMRIB's Software Library [[log in to unmask]] im Auftrag
> von Steve Smith [[log in to unmask]]
> Gesendet: Mittwoch, 8. Juli 2009 10:20
> An: [log in to unmask]
> Betreff: Re: [FSL] AW: [FSL] FE across sessions with slightly
> different paradigms
>
> Hi Andreas,
>
> Probably it's simpler than that - if your original models in FEAT had
> the same initial heights, and the contrasts were equivalent, then you
> should be able to compare these across runs in general (though of
> course caveats when comparing across different stimuation periods,
> e.g. block vs brief events is problematic...)
>
> Cheers.
>
>
> On 8 Jul 2009, at 08:16, Andreas Bartsch wrote:
>
>> Thanks, Steve!
>> So you mean we should check in the design.mat?
>>
>> Cheers!
>>
>> Andreas
>> ________________________________________
>> Von: FSL - FMRIB's Software Library [[log in to unmask]] im Auftrag
>> von Steve Smith [[log in to unmask]]
>> Gesendet: Dienstag, 7. Juli 2009 21:10
>> An: [log in to unmask]
>> Betreff: Re: [FSL] FE across sessions with slightly different
>> paradigms
>>
>> Hi,
>>
>> Yes, this sounds fine to me - as long as the 'height' of the first-
>> level model*COPE can be considered comparable across sessions.
>>
>> Cheers.
>>
>>
>>
>> On 5 Jul 2009, at 12:46, Andreas Bartsch wrote:
>>
>>> Hi,
>>>
>>> I have a rather general question about second level FE analyses:
>>> Is it legitimate to run a within-subject FE analysis across sessions
>>> with slightly different paradigms and tasks?
>>> Background: We have 3 session per subject (n~30) and each session
>>> was recorded under a slightly different paradigm / task. However,
>>> the sessions share certain main EVs and we were thinking to run a FE
>>> for these to get more robust and somewhat context-independent
>>> results.
>>> I can see that this will all depend on whether the assumption of a
>>> context-independent component to each run holds true but other than
>>> that does a FE in such case sound ok? Or would people strongly favor
>>> a 'conjunction'-type of analysis here?
>>> Thanks + cheers-
>>> Andreas
>>>
>>
>>
>> ---------------------------------------------------------------------------
>> 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
> ---------------------------------------------------------------------------
>
---------------------------------------------------------------------------
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
---------------------------------------------------------------------------
|