JISCMail - SPM Archives

I have a question about the experimental design for fast event-related fMRI experiments.
I am going to conduct a fast event-related fMRI experiment.

The trial sequence is that a phrase describing a situation (e.g., talking with other people) is presented for 3s and then a jittered ITI is presented (Mean = 3s, 2s – 7s).
The jittered ITI will be employed to avoid expected periodicity.
And, even though the stimulus presentation will be 3s, I assume that participants’ response will be pretty short even before the stimulus presentation is ended (less than 2s based on pilot tests).
Participants are asked to make a decision as soon as possible after reading the phrase during the whole trial presentation (5s – 10s).

Basically, I will make a contrast between stimulus presentation of A condition minus that of B, vice versa.
And, I will conduct a conjunction analysis of the contrast between stimulus presentation of A minus that of C (control condition) and the contrast between that of B minus that of C.

1) Whether jittered ITIs (Mean = 3s, 2s – 7s) are too short or not? (TR = 2s)?
At first, I intended to use a little longer ITIs (Mean = 7s, 2s – 12s). In many recommendations, “for the common effect, the optimal SOA is approx 16-20s; for the differential effect, the optimal SOA is the shortest SOA possible”. In my case, even though I am interested in the common effect, I will use a conjunction analysis based on two different contrasts (i.e., two differential effect contrasts) rather than using a common effect contrast (i.e., 1, 1 contrast). So, I concluded that I could use a little shorter ITIs like this. I am wondering whether my understanding is correct or do I need a little longer jittered ITI for detection power?

2) Is it possible to set the minimum ITI at shorter value (e.g., 1s)? I set 2s because I thought the minimum ITI should be at least 2s which is same to TR. I am wondering whether it’s correct understanding.
I have tried to find answers for these questions from discussions in the SPM mailing lists, webpages (e.g., http://imaging.mrc-cbu.cam.ac.uk/imaging/DesignEfficiency), and articles.
But, due to my lack of understanding, a lot of information makes me confusing more and more.
I should be very grateful if you let me know.