 |
 |
Hi Dirk: SPM uses a basis function approach, so you do not need to line up the trial onsets and the TR's. One special case of the basis function approach is selective averaging. In that case the TR's and trial onsets are lined up to make defining the trials and performing the averaging easier. There is no mathematical difference between the basis set approach using an FIR model and selective averaging. However, there are theoretical disadvantages to lining up the trial onsets and the TR's including restricted sampling of the HRF. Thus I would recommend appropriately jittering your trial onsets so that the TR is fully sampled. Then let SPM do the work of specifying the FIR design. I recommend choosing a bin size = TR. Darren ==============Original message text=============== On Thu, 28 Feb 2008 9:25:03 am CST Dirk den Ouden wrote: thanks Pete, yes, we are using optseq as well, to jitter the onset times of our TR wrt our event onsets. It's just that the suggestion was raised here that this was not a good idea when using a FIR model, because the time bins would consist of intenstity information from different volume acquisitions. best, Dirk %%-----Original Message----- %%From: SPM (Statistical Parametric Mapping) %%[mailto:[log in to unmask]] On Behalf Of Peter LaViolette %%Sent: Thursday, February 28, 2008 9:12 AM %%To: [log in to unmask] %%Subject: Re: [SPM] FIR and TRs %% %%Hi Dirk, %% I have no further input other than to point you to a %%program that helps in the optimization of FIR models called optseq: %%https://surfer.nmr.mgh.harvard.edu/fswiki/optseq2 %%https://surfer.nmr.mgh.harvard.edu/optseq/ %% Hopefully there is some material in there that will answer %%your questions. We've been using it for determining the %%onset times in our experiments, and ours end up jittered. %%Hope this helps, %%Pete %% %% %%On Thu, 28 Feb 2008, Dirk den Ouden wrote: %% %%> Dear all, %%> %%> We are having a discussion down here over whether the time %%bins of an %%> FIR function should be aligned with the onset of TRs in SPM models. %%> Would there be a problem with having (pseudo)randomly %%jittered onsets %%> of 2 second TRs that are not temporally aligned with 2 %%second FIR time bins? %%> %%> thanks for any input! %%> best, Dirk %%> %%> ************************************ %%> %%> Dirk-Bart den Ouden, Ph.D. %%> %%> %%> %%> Aphasia & Neurolinguistics Research Laboratory %%> %%> Dept. of Communication Sciences & Disorders %%> %%> Northwestern University %%> %%> 2240 Campus Drive, Evanston IL 60208-3066 %%> %%> Phone: (office) 1-847-467-2515; (lab) 1-847-467-7591 %%> %%> ************************************ %%> %%> %%> %% ===========End of original message text=========== -------------------------------- Darren R. Gitelman, M.D. Department of Neurology 710 N. Lake Shore Dr. #1122 Chicago, IL 60611 Voice: (312) 908-8614 Fax: (312) 908-5073 --------------------------------