Dear Ganesh
There is an example script for specifying DCMs in SPM12 at the bottom of this page: https://www.fil.ion.ucl.ac.uk/spm/data/attention/
Alternatively, there is a streamlined scripting interface for specifying fMRI DCMs in the development version of SPM. This version of SPM is available from https://github.com/spm/spm . See the help text at the top of spm_dcm_specify.m for an example.
However, I'd also push back and ask - are you sure you want to use the non-linear model? It does make an already difficult model fitting problem more challenging, so I'd recommend only using it for your hypotheses depend on it.
Best
Peter
-----Original Message-----
From: Ganesh Chand <[log in to unmask]>
Sent: 03 November 2023 20:22
To: [log in to unmask]; Zeidman, Peter <[log in to unmask]>
Subject: Spectral DCM Re: Batch script for non-linear modulations in DCM
⚠ Caution: External sender
Dear Peter,
I am wondering how to apply spectral DCM for the model structure example below. If there is any batch script suggestion for this example. Greatly appreciate your response for this and in the past below.
Best regards,
Ganesh
............................................
Dear Ganesh,
DCM.d should be of dimension [n x n x n] where n is the number of regions. The matrix DCM.d(:,:,1) represents non-linear connections from region 1, matrix DCM.d(:,:,2) represents non-linear connections from region 2, etc.
So to match the description you provided, model 2 should be:
DCM.d(1,3,2) = 1;
DCM.d(3,1,2) = 1;
Best,
Peter
From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]] On Behalf Of Ganesh Chand
Sent: 02 March 2016 13:52
To: [log in to unmask]
Subject: [SPM] Batch script for non-linear modulations in DCM
Dear SPM team,
I am having confusion how to write batch script for non-linear modulations in DCM.
DCM.options.nonlinear = 1;
Suppose I have THREE ROIs and they are intrinsically connected so I define
DCM.a = [1 1 1; 1 1 1; 1 1 1];
No bilinear modulation: DCM.b = zeros(3,3,0); No external input to a node: DCM.c = zeros(3,3,0);
%model 1:
DCM.d = zeros(3,3,1);DCM.d(3,2,1) = 1; DCM.d(2,3,1) = 1; % First ROI has nonlinear modulation in the connections between second ROI and third ROI save DCMM1 DCM
%model 2:
DCM.d = zeros(3,3,1);DCM.d(3,1,1) = 1; DCM.d(1,3,1) = 1; % Second ROI has nonlinear modulations in the connections between first ROI and third ROI save DCMM2 DCM
%model 3:
DCM.d = zeros(3,3,1);DCM.d(2,1,1) = 1; DCM.d(1,2,1) = 1; %Third ROI has nonlinear modulation in the connections between first ROI and second ROI save DCMM3 DCM
Are the models defined here the correct way for nonlinear modulations in DCM? If this is the correct way, my confusion is how the model 1 (for example) knows the first ROI is providing modulation effect?
Your advice is greatly appreciated in advance.
Best regards,
Ganesh Chand
|