Hi Steve,
thanks for the reply. It hadn't been obvious to me, but it is now.
Thanks for your help as always!
Cheers,
Cornelius
On Mon, Feb 15, 2010 at 10:31 AM, Stephen Smith <[log in to unmask]> wrote:
> Hi - I didn't quite follow all of your logic below, but I suspect it is
> reasonable - my main comment would be that if you want to break down
> double-differential effects (e.g. one group is less negative than another
> wrt baseline) then it's always good to look at the individual
> effects/differences in order to know what the signs and magnitudes (i.e.
> within the "lower-level" differencing) were to help interpretation - hope
> that's clear and sorry if it's obvious! Another related point - e.g. where
> you say effects are "empty" - I guess you mean nothing survives thresholding
> - but it may still be informative to at least look at the sign of the
> unthresholded effects.
> Cheers, Steve.
>
> On 12 Feb 2010, at 13:12, Cornelius Werner wrote:
>
> Hi all,
>
> We've been running an experiment employing a simple block-type motor
> task with patients and controls. Now we were looking for changes in
> connectivity of motor regions. MELODIC identified a nice anatomical
> region with relevance to the condition studied ("A") as a single
> component, so we went for that. Dual_regression showed that this
> component had a stronger connectivity to area B (also meaningful) in
> patients than controls, which was nice. We confirmed this by running a
> seed-voxel connectivity analysis using an anatomical seed ROI covering
> the same anatomical structure A as the MELODIC component did, getting
> very similar results (significantly higher connectivity in patients
> between A and B).
> We now were interested in whether connectivity was modulated by the
> task. So, we ran a PPI according to Jill Kelly's recipe, convolving
> the anatomical ROI timecourse with the task. The "main PPI effects"
> for each group came up empty. But the contrast controls>patients gave
> a significant blob, highlighting basically area B! Being only
> physicians, we're somewhat dumbfounded by that :-)
>
> We came up with this interpretation, and I'd be glad to hear some
> opinions on it:
> Patients show higher connectivity between areas A and B, independently
> of task (MELODIC, seed voxel analysis). No areas correlate
> significantly with area A along the motor task in each group (PPI
> negative for each group). Nevertheless, modulation of connectivity
> between A and B is different between groups when given a motor task
> (positive contrast of the PPI data, perhaps due to a nonsignificant
> anti-correlation in patients and non-significant correlation in
> controls, the difference becoming significant). Baseline: there is
> more "static" correlation between A and B in patients, and the ability
> to modulate along a task differs, favoring controls. Is that ok so
> far? Or did we run into some logical circle somewhere?
>
> Is there any way/tool within FSL to further elucidate this issue, in
> particular shedding light on the modulation issue? What could
> differences be, particularly regarding the empty main effects of the
> PPI?
>
> Any help is greatly appreciated as always!
> Best regards,
> Cornelius
>
> --
> Dr. med. Cornelius J. Werner
>
> Department of Neurology
> RWTH Aachen University
> Pauwelsstr. 30
> 52074 Aachen
> Germany
>
> Institute of Neuroscience and Medicine
> MR Physics - INM4
> Research Centre Juelich
> 52425 Juelich
> Germany
>
> ::: Please encrypt confidential data :::
>
>
>
> ---------------------------------------------------------------------------
> 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
> ---------------------------------------------------------------------------
>
>
>
>
--
Dr. med. Cornelius J. Werner
Department of Neurology
RWTH Aachen University
Pauwelsstr. 30
52074 Aachen
Germany
Institute of Neuroscience and Medicine
MR Physics - INM4
Research Centre Juelich
52425 Juelich
Germany
::: Please encrypt confidential data :::
|