Dear SPM-ers
I am looking to run laterality analysis on some fMRI data that I collected, I have read through the archives and found the method which involves flipping the images and comparing these with the 'unflipped' images. Here is the method that I followed:
- LR flipping the pre-processed normalised (smoothed) images, for this I used the Imcalc 'flipud(X)' function on each image
- Reslicing these 'flipped' images using one of the original 'unflipped' images as a reference (Batch - Coregister - Reslice)
- Running first level analysis on the 'flipped' images
- Running a second level analysis (paired t test) - inputting the 'unflipped' and 'flipped' con_* files as a pair for each participant.
However, when I run this analysis the output I get doesn't make a lot of sense. The spmT* file from the second level analysis shows only random voxels are activated (noise) which are at a very low threshold i.e. there are no clear regions which are greater in one hemisphere compared to another. I am using a contrast that has clear laterality effects, so I was expecting to see more defined 'blobs'.
Is there something obvious that I am doing wrong? Or is this problem to do with the fact that I am currently not using a symmetrical template? If so, how do you create one of these?
Many thanks in advance,
Grace
|