The error message !invalid contrast in SPM typically indicates that the contrast vector (convec) does not match the design matrix in some way.
Given the information you provided, here are a few steps to debug and correct the issue:
Step 1: Verify the Design Matrix
Ensure that the design matrix (SPM.xX.X) is correctly specified and contains 26 regressors. You have already done this, but double-check that the regressors are in the expected order.
load(fullfile(subdir, 'SPM.mat'));
numRegressors = size(SPM.xX.X, 2);
disp(['Number of regressors: ', num2str(numRegressors)]);
Step 2: Check the Contrast Vector
Ensure that the contrast vector is correctly specified and matches the design matrix. Here is an example of how to create the contrast vector:
% Initial contrast vector for the conditions of interest
initial_convec = [0 0 -1 0 0 1 0 0 0 0 0 0 0];
% Number of confounds
num_confounds = size(confounds_matrix, 2);
% Additional zeros for the confounds
additional_convec = zeros(1, num_confounds + 1);
% Complete contrast vector
convec = [initial_convec, additional_convec];
% Ensure the length of convec matches the number of regressors
assert(length(convec) == numRegressors, 'Contrast vector length does not match number of regressors');
Step 3: Specify the Contrast in the Batch
Ensure that the contrast is correctly specified in the batch. Here is an example:
matlabbatch{3}.spm.stats.con.consess{1}.tcon.name = 'nav-follow';
matlabbatch{3}.spm.stats.con.consess{1}.tcon.convec = convec;
matlabbatch{3}.spm.stats.con.consess{1}.tcon.sessrep = 'none'; % or 'replsc' if appropriate
matlabbatch{3}.spm.stats.con.delete = 0;
Step 4: Run the Batch
Run the batch and check for errors:
spm_jobman('run', matlabbatch);
