Dear Jodi,
I'm afraid you will have problems with temporal filtering and autocorrelation estimation.
We do not normally recommend intensity normalisation beyond the 4D grand-mean-scaling.
Therefore I would recommend not doing this and doing it via a multi-level modelling
approach, as explained in the FEAT manual.
All the best,
Mark
On 22 Feb 2012, at 19:19, Jodi Gilman wrote:
> Hi,
>
> My fMRI study has three runs that I would like to merge in time (for various reasons, I'd like to analyze it as one long run instead of three short runs). If I merge the 3 raw functional runs, will I have problems with intensity normalization?
>
> Thank you,
> Jodi
>
|