Dear Lucs,
> I like to conduct a 3 (positive / negative / neutral) * 2 (congruent / incongruent) event-related fMRI-design.
>
> Is it possible to have different numbers of trials for different conditions?
>
> As I do not have much scanner-time, and because I think that the positive-conditon, relative to negative and neutral, might have the smallest effect size,
> I like to increase the number of trials for this condition, while decreasing the number of trials in the other conditions.
>
> Is it possible in general and if so can the resulting data be analyzed with SPM?
Well, one could it answer in two ways:
You ask whether it is possible to have different numbers of trials per condition. The answer is 'yes'
My personal advice is, that you still should try to balance it and to have as much trials as possible. However, the recommended number of trials per condition varies in the literature, some use at least 20 trials per condition others 40.
But in principle it is possible to have different number of trials. But, in case of a null result, you might have problems in finding the correct interpretation. It could has happened that it does not became significant because you are comparing a good signal to a bad signal, i.e. low to larger variance.
Good luck,
Karsten
--
-------------------------------------------------------------
Karsten Specht, PhD
Department of Biological and Medical Psychology
& National Competence Centre for functional MRI
University of Bergen
Jonas Lies vei 91
5009 Bergen
Norway
Tel.: +47-555-86279
Fax: +47-555-89872
[log in to unmask]
http://fmri.uib.no/
|