Hello everyone,
I'm about to analyze data collected in an fMRI training study, where participants undergo two scanning sessions 10 days apart, one before and one after a 9 days-long behavioral training. I'm having few doubts on preprocessing and glm:
1. In preprocessing my data, am I right to assume that I can treat data for a single subject coming from pre- and post- learning sessions as two different datasets? I have two anatomical scans (hopefully identical!) and I'm doing all the preprocessing steps (registration etc) separately for the two ds, under the assumption that projecting everything in the same MNI space will result in an identical and common neural space. Is that right?
2. I'm having the same "conceptual" problem with the GLM. Let's say that I have two conditions, A and B, both presented during pre and post-learning. I want to investigate, for instance, where A(post) > A(pre). I think I got the idea of what I should do from a statistical point of view, but again, am I right in estimating two GLMs separately for pre and post sessions, as they were from different subjects, and only then compare the betas I extracted for my conditions? I feel like I'm missing something here...
Thank you very much and apologies for the silly questions: first fmri experience - and message - here! :D
S.
|