 |
 |
Hello SPMers In a recent fMRI experiment each participant performed the same task twice in the scanner (to increase power). There was no experimental manipulation between the 2 runs, they just simply repeated exactly the same task (albeit using a new set of stimuli) I considered 2 ways of combining these data 1 - Model them as one huge long run 2 - Model them as two distinct sessions within one fixed effects analysis When I tried option 2, the prompt questions asked me whether sessions 2 was a repetition of session 1 Although it was the same person, I had to say no because the order of presentation of blocks was different between the 2 runs (I'd earlier noticed that if I said yes it just modelled the same block order/timings for both runs) Now mechanically it runs fine What I am concerned about is what is going on in the background. For instance, if one were analysing behavioural data in say SPSS, one knows that there are differences in the underlying algorithms So, as I have the model set up (i.e. because I had to say it was not a replication), how can spm know that it is in fact a within-subjects design rather than a between-subjects design? Thanks Laurie