Dear SPM experts,
What is the best way the analyse this kind of data with SPM?
Your help will be more than appreciated.
Here the description of the data set:
I have two kinds of stimuli (1. Mec et 2. Elec).
Each kind of stimuli is divided in three categories:
a) Nai
b) Sci
c) Con
And each of them can be divided (depending of the answer giving by the subjects):
i) Right
ii) Wrong
And we have a stimulus called "Rest".
Then, we have
1. Mec
a) Nai
i) Right
ii) Wrong
b) Sci
i) Right
ii) Wrong
c) Con
i) Right
ii) Wrong
2. Ele
a) Nai
i) Right
ii) Wrong
b) Sci
i) Right
ii) Wrong
c) Con
i) Right
ii) Wrong
3. Rest
The stimuli were presented in two runs.
After the two runs, we presented an instructional movie in the scan, and, just after that, we redo the two runs (repeated mesures). The subject is in the scan for the 4 runs. To resume, the runs were:
Run 1 (before presenting the movie)
Run 2 (before presenting the movie)
Movie (no data collect)
Run 1 (after presenting the movie)
Run 2 (after presenting the movie)
Anatomical scan
Above that, I have two groups of subjects: novices and experts.
I am particularly interesting to see if there is more activations in runs after the movie compare to before for the novices group
and to compare if there is more or less activation in experts (run 1 and 2 before movie) compare to novices (run 1 and 2 before movie)
I am not used to analyze this kind of data. I was simple t-tests comparing one conditions to another. But I don't know how to deal with repeated measures. Spontaneously, I think of a paired t-test between contrasts before and after the presentation of the movie, but I don't know to implement it in SPM. Do I need a factorial design? If yes, how to define it in SPM? Other ideas?
Really, thanks for your advice.
Steve
|