Hi Everyone,
I have a few questions.
First, I am about to process my event related fmri data through the feat gui (1st level). Up to this point, I have already done motion correction through command line as I needed the motion corrected data as input for my distorion correction (topup/applytopup). I realize distortion correction can also be done through the gui but this was not the route I have taken for reasons I wont go into. Now that I am in the gui, I have obviously deselected motion correction as it is already done. However, I want to use my 6 motion parameters as aggressors of no interest. Unfortunately, this option is not available (greyed out) since I have turned off motion correction. I wonder if a 'workaround' would be to simply enter a text file with the the 6 motion parameters into the "add additional confound EVs". I generated this file myself when I ran motion correction via command line. Note that for some Ss I also need to do some censoring, so I assume I would need to combine the text file with the censored timepoints and the 6 regressors into one single text file. Is this a reasonable way to approach this?
My second question is about high pass filter cutoff choices. Using one run, from one subject, the "estimate high pass filter" option yields 90 s as a recommended cutoff. Can I then assume this is the value I can use for every other Ss? I know the recommendation of the high pass filter varies by design but I am wondering if I can go even lower than that. My events of interest are only a few seconds long. I have seen folks use 50 or 60 s.
Finally, my noise level percent is set at .66. When I try to estimate the noise from the data, my feat crashes. Can I leave it as .66?
Thanks,
D
|