Dear Sarah,
I don't think you need to concatenate here: why not model them as
separate sessions where for the second session, you assign trials to
different conditions whether they were repeated from the first session
or are new ones - or just use the scans corresponding to the second half
of the session (you can use negative onsets if needed). The contrasts
you will define will probably average effects across sessions.
Best regards,
Guillaume.
On 18/11/2019 13:21, Sarah Buck wrote:
> Hello,
>
> I work with memory (task-based) fMRI. For two participants, we had to stop the memory fMRI task half way through, and start the whole task again, immediately after. These participants were therefore administered the same task twice (scan 1, halfway; scan 2, full). Because of the memory nature of the task, I don't want to simply consider the second scan of the fMRI task since the items will have already been presented in the first scan.
>
> To what extent is it possible to merge the images from scans 1 and 2? That is, take the first half of images from scan 1 and the second half of images from scan 2? If this is possible, how do we then deal with the realignment process? Is there any precautionary measure to take, or do we simply consider all images as though they were acquired during the same scan?
>
> Thanks for your input!
>
> Sarah
>
--
Guillaume Flandin, PhD
Wellcome Centre for Human Neuroimaging
UCL Queen Square Institute of Neurology
London WC1N 3BG
|