Hi all,
Sorry, this is a pretty basic question but I couldn't find a direct answer in the archives.
I would like to analyze whether activity within two ROIs correlate with behavior (RT) on a trial-by-trial basis. Each subject completed two runs. Each run contained 3 conditions. Only 2 conditions are of interest for the correlations. I extracted the mean ts for the two runs for each subject using fslmeants. So I have N*2 time series files for each ROI. I also have the RTs corresponding to each time point.
What's the best way to proceed at this point?
1. Do linear or mixed regression models within R or some other statistical package?
2. Use Feat in some way?
3. Compute basic correlations between the variables for each subject and then ANOVA?
On a related but different note, should the two time series from each subject be collapsed or treated independently?
Thanks in advance,
Malathi
|