Dear Jon,
It is perfectly possible to have a design matrix for the whole 5
minutes of scanning, within which you model the 'activations' in the
first minute in one set of covariates (e.g. one covariate for
'activation' and one for 'baseline') and the 'activations' in the
rest of the scan as another set of covariates. So you don't need to
do anything 'fancy'. E.g. if your covariates were ordered:
activation (1st min), baseline (1st min), activations (mins 2-5),
baseline (mins 2-5)
then you could test for the condition by time interaction (which is
what you described) with the contrast 1 -1 -1 1.
I am not altogether sure, though, how interesting this contrast is in
terms of the physiology. One could imagine a condition by time
interaction occurring non-specifically, perhaps even because of the
physics of the scanner. An interpretation in terms of physiology
would be more compelling if one could re-set the physiology in some
way, so that the 'early' and 'late' activations are repeated several
times during the course of an experiment.
[Incidentally, you are probably aware that you need to throw away the
first few scans because of T1 saturation effects, but I thought that
I would mention it just in case!]
Best wishes,
Richard.
>Hi All,
>
>Having acquired fMRI time series data (block design)
>and performed a first pass SPM analysis, I would now
>like to examine whether there are significant differences
>between the activations observed in the first minute to
>those acquired in the remaining 4 minutes of the scan.
>
>Is it possible to break down a 5 minute scan into minute
>intervals for comparison? I have done a quick search on
>the mailbase archive and drawn a blank.....
>
>I realise that I could divide the raw data up and generate
>5 * 1 minute long scans - but wondered if I could
>do anything fancy with SPM?
>
>Hope someone can help.
>
>Yours,
>
>Jon.
>
>_____________________________________________________
>Jonathan Brooks Ph.D. (Research Fellow)
>Magnetic Resonance and Image Analysis Research Centre
>University of Liverpool, Pembroke Place, L69 3BX, UK
>tel: +44 151 794 5629 fax: +44 151 794 5635
--
from: Dr Richard Perry,
Clinical Research Fellow, Wellcome Department of Cognitive Neurology,
Darwin Building, University College London, Gower Street, London WC1E
6BT.
Tel: 0171 504 2187; e mail: [log in to unmask]
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|