Hi Conny
You have me confused here:
it's true that because the conditions are separated in time you don't
have a problem with overlapping 'on' conditions - but by putting all the
'on' periods into a single EV you impose strong constraints that prevent
you from looking at what you want.
Specifically, when creating the single EV you specify the 'height' of
modulation under conditions A,B and C. You therefore fix the relative
'activation' strength -- but this is exactly what you're trying to
extract later on by looking at a contrast. If you only allow for a
single EV, the 'on' blocks either jointly modulate a voxel's time course
or don't.
If you want to compare conditions A and B (in the way that I assume you
want, that is ' where is response under condition B larger than under
condition A'), you need different EVs.
In your example it seems that you want 3 EVs, one for each condition.
Then you can use contrasts to compare these, e.g.
[1 0 0] looks at the voxels that have an increase in signal intensity
under condition A.
[1 -1 0] shows those voxels where the signal change under condition B
exceeds signal change under condition A
etc...
Viele Gruesse
Christian
Conny Schmidt wrote:
> Hi all,
>
> again I got a question about paradigm description in FEAT, it is somehow
> similar to my last one, but I didn't find a way to resolve the problem
> in this case....
>
> I got two tasks in one paradigm, which follow the same scheme of on/off
> periods, but are shifted in time. So the stimulation could be described
> with one EV. As the on periods are shorter than the off periods I
> couldn't use the trick with two contrasts [1] and [-1] again. So how
> could I compare both conditions?
>
> I just make an example to illustrate, what I mean:
>
> R rest condition
> A condition 1
> B condition 2
> C condition 3
> in brackets the duration
>
> The paradigm would look like this then:
>
> A(10) R(10) C(10) R(20) B(10) R(40) A(10) R(10)
> C(10) R(20) B(10) R(40) A(10) R(10) C(10)
> R(20) B(10) R(40) .....
>
> and I want to compare condition B with condition A.
>
>
> Thanks in advance!
> Conny
>
>
--
Christian F. Beckmann
Address: Oxford University Centre for Functional
Magnetic Resonance Imaging of the Brain,
John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK
Email: [log in to unmask] http://www.fmrib.ox.ac.uk/~beckmann/
Phone: +44(0)1865 222782 Fax: +44(0)1865 222717 Mob: +44(0)7980 691852
|