 |
 |
Hey all, I've been trying to use ICA/DR on our data set. We have 323 inputs, where we have 19 subjects who were scanned 17 days. The middle 9 days were a treatment while the other days were a control. I wanted to see if this design matrix and contrast looked correct for this. I have 19 columns where I have 1's to indicate a given subject then 0's for everything else. The last column I have -1 everywhere there's a control scan, and 0.8888889 everywhere there's a story scan (I chose 8/9 because we have 8 control days per person but 9 story days, so 8/9 to give 0 if no effect). My contrast has 0s everywhere except the last column. Does this look right? Also, I used the DR script and randomize. I read on the randomize website that the tfce_p_tstat and tfce_corrp_tstat values are 1-p, so if we want p<0.05, we need values > 0.95. Is this true for the tfce_corrp_tstat file as well. I notice that most of my values are ~0.4-0.6. I assume these are lower because the p-value has been corrected? Thanks. Brandon /NumWaves 20 /NumPoints 323 /PPheights 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.888889e+00 /Matrix 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 -1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 -1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 -1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 -1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.8888888889 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.8888888889 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.8888888889 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.8888888889 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.8888888889 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.8888888889 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.8888888889 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.8888888889 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.8888888889 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 -1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 -1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 -1 /ContrastName1 "test effect" /NumWaves 20 /NumContrasts 1 /PPheights 1.000000e+00 /Matrix 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1.0