The references to external regressors in mkanalysis-sess are a little sparse on how I would go about analyzing a continuous regressor: I have a mixed design experiment:
TaskBlock1 Rest TaskBlock2 Rest … TaskBlockN Rest x6 Runs
Within each block are jittered events (5 of condition A, 5 of condition B). A colleague of mine agreed that the A vs B contrast should be pretty robust even in a single participant (it’s an established paradigm), considering I have over 150 trials of each, but I am seeing nothing at any reasonable threshold in my single participant. My colleague suggested checking whether there were any regions with activity that increased/decreased as a function of block over the course of the experiment (essentially, a neural practice effect). Hypothetically, such a region would show a linear drift in activity across 6*N blocks, but only during the block periods (reverting to baseline during the rest periods).
From the mkanalysis-sess.new wiki page, it appears that the -extreg flag can be used to pass an additional regressor file, but it’s unclear from the rest of the description how to construct this file, nor how to view the GLM associated with this analysis.
Thanks for your time; I’ll post the solution on my lab wiki once it’s all sorted out Chris
/********************************************** * Chris McNorgan * Assistant Professor * Department of Psychology * University at Buffalo, * The State University of New York * http://ccnlab.buffalo.edu/ * Office: 716.645.0236 * Lab: 716.645.0222 **********************************************/