Dear SPMers
We use a sparse sampling with TR=5.6 and TA=1.4s in a
block design, with sounds of categories A,B,C,D,E plus
silense (_). The design goes like this:
xAxAxAx_xBxBxBx_xCxCxCx_xAxAxAx_ etc.
where x stands for the 1.4s of image aquisition, and A etc
for the 3s of sound presentation in the absence of a sound
Now to analyse this data set, we used no slice timing, as
it wouldn't make much sense.
Now, in building up the model, we have to tell SPM when to
consider that the data was aquired relative to onsets of
the conditions defined in the batch.
We thought of doing it as follows, but wanted your input:
SPM.xBF.T = 4; % number of time bins per scan
SPM.xBF.T0 = 1; % first time bin (see slice timing)
as a result, each TR (5.6s) is divided into 4 bins of 1.4s
each, and acquisition is considered to occur in the first
of these bins, i.e. in the first 1.4.
Does that make sense?
-----------------------------------------
Christian Keysers, PhD
Associate Professor
Neuro-Imaging Center
University Medical Center Groningen
Tel.: +31-50-3638794
|