Hi there
Thanks for reporting this. We generally test DCM for fMRI using a dataset that has a non-integer TR, so I think the problem might lay elsewhere. Would you be able to send me an example subject's DCM to look at? (Zipped, as I can't received .mat files).
Many thanks
Peter
-----Original Message-----
From: SPM (Statistical Parametric Mapping) <[log in to unmask]> On Behalf Of Anonymous
Sent: 02 December 2022 16:54
To: [log in to unmask]
Subject: [SPM] Interscan Interval in DCM Analysis
⚠ Caution: External sender
Hello, I am trying to follow the manual for DCM analysis. I am using resting state data with 0.8 TR. However, it gives an expecting an integer error at line 67 in spm_csd2ccf.m when I run DCM/estimate (cross-spectra). Here is what I think is the problem: in the specification step VOI timings were taken as 0.4 (half of the TR) and I can't change it (I tried giving 1 1 1 1 instead) and during the estimation since 0.4 is not an integer it gives an error. All examples were given on TR of 2. How can I use my correct TR value?
|