 |
 |
Dear Inken, Good question. It's just a rule of thumb. The prior on each of these connections is a normal distribution with mean 0 and variance 1/64. So you hope that at least one of your connections moves away from this prior. An estimated posterior value of 1/8Hz is unlikely given this prior (you can check how unlikely using Matlab's normcdf function). Best, Peter -----Original Message----- From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]] On Behalf Of Inken Tödt Sent: 20 August 2015 13:02 To: [log in to unmask] Subject: [SPM] DCM connection strength Dear All. actually I am working on a DCM study an used amongst others the function spm_dcm_fmri_check to check my results. In the description of this function it says, "one would normally expect to see one or more extrinsic connections with the strength of 1/8 Hz or greater. My question is, if there is any literature, so I can understand, why the connection strength is fixed on this specific value? Best regards, Inken