Iwo,
Thank you for your response. The two images I included were the same analyses performed on two different subjects. As you suggested, the conf columns are the motion parameters (included automatically through FEAT). I'm not sure why the motion was so regular for person two. Perhaps his/her head moved slightly during breathing. For subjects in which it looks like the artifact may be related to motion I could probably keep the data then as the motion parameters should account for the "bump" and I already checked absolute and relative motion for all of my scans and they were at acceptable levels. I imagine for the others (like person two) I should assume it is an unexplained artifact and exclude the subject. Thanks again.
Best,
Meredith
Hi Meredith,
I am not sure but it looks like an motion-related artifact.
I assume that regressors labelled as 'conf' are 6 rigid-body motion parameters?
You can see some correlation especially between 4th confound and "the bump" an fthe first image.The situation illuastrated in the second image seems to be less clear and the confounds don't look like typical motion parameters (too regular patterns)... what are they? What's the difference between the two images?
Whatever it is the fact hat you get similar bump-like patter in the CSF suggests that it is a kind of an artifact.
Also it looks like you've included time drivatives in your model. I don't think it's necessary (or even advisable? maybe somebody else can comment on this) in resting state.
Best wishes,
Iwo
|