Procedure for performing Paired T-Test?

Submitted by Chris Stevens on
Hello, I'm attempting to perform a Paired T-Test on the functional connectivity of two resting state conditions within-subject. I have a Placebo run and a Control run for the subject. First I converted my .nii files to nifti pairs using REST's conversion utility. Then I used the Functional Connectivity tool to create a FC and zFC Map for each condition individually using the same ROI location. Then I placed the zFC maps into their own folders (I have 1 folder for the zFC map from the Placebo run and another folder for the zFC Map from the Control run). Next I opened REST's Paired T-Test tool. I added both folders containing the run data as separate conditions in the Paired T-Test, Placebo as Condition 1 and Control as Condition 2. I set the mask to one of the zFCMaps then I ran the analysis, and another nifti pair was produced which I viewed using REST's slice viewer overlaying the image over the subject's anatomical image. I was able to see some results but I'm unsure that I've performed these steps properly. Does anyone have experience performing Paired T-Test's using REST and would be willing to share their general procedure? I admit I am new to REST and performing paired T-Tests so it is likely that I am doing something wrong. Any help from someone familiar with performing paired T-Tests would be greatly appreciated! Thank you!

YAN Chao-Gan

Tue, 11/29/2011 - 04:14

Yeah, you are doing in the correct way. The only thing you need to keep in mind is the correspondence of the file order in Placebo folder and Control folder. Paired T-Test in REST is very simple compared with SPM. Best, Chao-Gan
Forums