Hi - thanks for sending the data.
Yes - this is all correct. You are applying a [-1 0 1 0 ] contrast
i.e. EV3 - EV1. The peak voxel for this contrast does indeed show, in
the data and the full model fit, that the "on" periods for EV1 are
significantly lower than the "on" periods for EV3, as the contrast has
requested - hopefully that is straightforward to understand.
The reason that the -partial- model fit looks confusing (i.e., is
showing EV1 "on" periods as high and the rest as flat) is that the
partial model fits are created by multiplying the individual EV fits
by the contrast. Because in this voxel's timeseries the data is low
for the EV1 on periods, once that is multiplied by the contrast, it
goes high. In general, partial model plots for differential contrasts
are complicated to interpret, as in this case.
Cheers, Steve.
On 20 May 2008, at 22:44, Steve Smith wrote:
> Hi - I'm afraid all your uploads were empty apart from the html
> files - you need to upload the complete output FEAT directory from a
> single subject.
> Cheers, Steve.
>
> On 19 May 2008, at 19:20, Carlos Faraco wrote:
>
>> Steve,
>>
>> I have noticed that sometimes this situation occurs when there is no
>> activation above threshold for that comparison.
>>
>> Hoever, I have uploaded three report files for three different
>> subjects in
>> which the situation occurs when there IS activation above
>> threshold. This
>> usually occurs on the 4th comparison which is an arithmetic
>> condition -
>> OPSAN task. Details of the task are on post 015578 (the CV's are
>> longer
>> setup up as in that post, as you will see).
>>
>> The upload # is 960782.
>>
>> Thanks you,
>>
>> Carlos Faraco
>>
>
>
> ---------------------------------------------------------------------------
> 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
---------------------------------------------------------------------------
|