Kaoru Sekiyama wrote:
> Dear SPMers,
>
> From 8 subjects, I have obtained 2 sessions of fMRI data, each session
> having 3 epoch conditions. Session 1: A, B, X; Session 2: C, D, X,
> with X as a control condition. Due to the small number of subjects,
> I did fixed effect analyses for each of Session 1 (to see activation for
> A, B)
> and Session 2 (C, D). Now I want to know across-session differences
> such as A vs C.
>
> From the archive I notice that the across session comparison is less
> powerful
> due to session effects. But if I want to see only strong activation that
> survives
> session effects, I thought this may be a possible way.
>
> Fixed effect analysis with 16 sessions (8 subjects x 2 sessions),
> with contrast (1 0 0 -1 0 0 ) x 8.
>
> Is this acceptable?
> Does anyone know better ways?
>
This looks fine.
Although (depending on exactly what your
control is) it may be better to use
1 0 -1 -1 0 1
ie. the difference between (i) A and its
control and (ii) C and its control.
If, for example, the control is a
generic baseline common to both sessions
it would soak up the session effect.
Best wishes,
Will.
> Any comments will be much appreciated.
>
> Kaoru Sekiyama
>
>
--
William D. Penny
Wellcome Department of Imaging Neuroscience
University College London
12 Queen Square
London WC1N 3BG
Tel: 020 7833 7478
FAX: 020 7813 1420
Email: [log in to unmask]
URL: http://www.fil.ion.ucl.ac.uk/~wpenny/
|