Hello,
We recently ran a parametric modulation analysis. We used two EVs: one EV with a parametric modulator (i.e., " Subjective Ratings") and another EV using the same onset and duration data yet with a constant intensity (i.e., "Main Effect"). We utilized orthogonalization in an attempt to visualize the brain activity that uniquely correlate with the trial-by-trial changes in Subjective Ratings. The first-level analysis was run with the Main Effect EV as the first EV, and the Subjective Ratings EV as the second EV. In this second EV tab, orthagonalization was selected and the Subjective Ratings EV was orthagonalized with respect to the Main Effect EV.
The results show robust activity in the Main Effect contrast to which the subjective rating EV was orthagonalized. However, the orthagonalized contrast against baseline shows no activation, even at an uncorrected threshold. All z stats in the brain are listed as 0 in fsleyes. This may simply mean that there is no significant parametric modulation. However, I find it odd that all z stats are 0, and no activation appears even in the uncorrected image. As I have never utilized orthogonalization before, I was hoping to get some feedback from others. Did something go wrong during the analysis, or is there simply no surviving activation? Should you agree that this result is suspect, do you have any advice regarding what may have gone wrong?
Another proposed method for viewing the activity of by trial frustration ratings independently is to demean a parametric modulator and run the same analysis described above without orthagonalization. Does this sound reasonable to you?
Thank you,
Lynn
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
|