Ok, first of all you didn't mention what your slice acquisition is (e.g. sequential top --> bottom). Second, how does your ISI vary? Is the ISI only 5, 6, or 7 seconds? Or can it be 5, 5.5, 6, 6.5, 7? etc.
Let's assume it's only 5-6-7, and you have a sequential acquisition (doesn't matter in this example). If this is the case, then at stimulus presentation you will always be acquiring slice 1 or slice (n/2). So there isn't much distribution around stim onset. If your ISI varied [5:0.5:7], then you maybe wouldn't need to slice-time because at stimulus presentation you would be acquiring slice 1, (n/4), (n/2), or (3n/4).
See e.g. this page for more info on slice-timing: http://imaging.mrc-cbu.cam.ac.uk/imaging/SliceTiming
________________________________________
From: SPM (Statistical Parametric Mapping) [[log in to unmask]] On Behalf Of Eric Vidoni [[log in to unmask]]
Sent: Tuesday, November 23, 2010 9:44 AM
To: [log in to unmask]
Subject: [SPM] Second Posting - Slice Timing and Event-related Design
Hello,
I apologize for the second post but I have a question regarding the use of slice timing correction. My current protocol uses a event-related design with a variable ISI (random between 5000-7000ms, TR=2000ms). I'm wondering if slice timing is unnecessary when using a variable ISI. My understanding of slice timing correction is to adjust each slice as if it were acquired at the same time relative to the stimulus, rather than some slices consistently lagging. However, in a variable ISI the timing of acquisition for any given slice is likely to be distributed rather evenly around stimulus onset, correct? Therefore, would you still need to adjust for the timing of slice acquisition as it relates to the stimulus onset?
Any clarificiation or correction would be greatly appreciated.
Thanks much,
Eric Vidoni
University of Kansas
Alzheimer and Memory Program
|