Dear list,
The issue of within-trial orthogonlisation and FIR modeling has been mentioned previously (https://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=SPM;347bbbf.0905). However, I remain curious as to why within-trial type orthogonalisation is hardcoded into SPM.
We've commented out the offending lines (276 to 279 in spm8r4010) but I remain curious as to what type of situations one would desire within-trial orthogonalisation. Parametric designs?
If someone could shed some light, that would be brilliant.
Cheers.
|