Hi all,
I'm having a bit of a teething problem with SPM12 batch mode. I'm trying to run a batch on a simple movement task, starting at slice timing etc. and including fmri model specification, estimation and contrast manager modules and the results are very strange i.e. glass brain activations are not as expected and the design matrix is different to what I see when reviewing the model.
Stranger still, if I re-run the analysis from fmri model specification i.e. using the same batch but removing all modules before model spec, the results are completely different and in fact, more like what we would expect. I've attached some outputs, the 1st one shows a contrast after running the batch once and the 2nd shows the same contrast after re-running from the fmri model specification stage.
I've also just separated the same batch into two sub-batches i.e. preprocessing and model/contrast, and re-analysed the raw data to double check whether the SPM.mat file is somehow updating funny when re-running for the second time...however, the results are identical to the 2nd image (expected results).
Does anyone know why this is happening?
Thanks very much.
Nick
|