Hello everybody!
I am preparing an event related fMRI study with n = 4 different conditions
and trial length of 3.4 sec. TR=2.
I am especially interested in differential contrasts and a parametric
analysis based on reaction times in the different conditions. Because of the
(unchangable) length of each trial of 3.6 sec, I am not sure which design to
take: jitter, nullevents, both or none of both.
If I understood it right, it would be ok not to use null-events because I am
not interested in baseline effects. However, I don't feel quite well not to
use any baseline at all. The trials should at least be separated by 1 sec or
more (otherwise it would be too demanding for the subjects).
Is it maybe, therefore, reasonable to jitter between the trials (let say 1-3
secs) in order to prevent fixed SOAs? Or do I need longer ISI since trial
length is already 3.6 sec? I need an event related approach since I would
like to model single reaction times and only trials in which subjects
answered correctly.
Cheers,
Christian
-------------------------------
Dr. med. Christian Grefkes
Institute of Medicine
Research Centre Juelich, 52425 Jülich, Germany
phone: +49 2461 61 8609
fax: +49 2461 61 8297
http://www.fz-juelich.de/ime/ime_cognitive_neurology
Department of Neurology
University Hospital RWTH Aachen, Pauwelsstr. 30, 52074 Aachen, Germany
phone: +49 241 80 89613
fax: +49 241 80 82444
http://www.ukaachen.de/content/listpage/1331868
|