Hello,

I have carried out an amplitude modulation analysis where I am asking which voxels are have BOLD amplitudes modulated by variable A and/or variable B or are unmodulated by these variables. So my -stim_times_AM2 input file looks like this: 20*15,30 44*2,10, etc.

For my output I get a beta coefficient and t statistic for the effects of interest: 1) unmodulated effect,2) the effect that is modulated by Factor A, and 3) the effect that is modulated by Factor B.

I have two questions:

1) For a given region of interest, how do I obtain the correlation coefficient for each of the effects? I know how to use 3dROIstats to get the mean values for all the voxels in an ROI, but how do I get a correlation coefficient from the beta and tscores?

2) For the unmodulated effect, does that mean that those are voxels where activity was equal (i.e., not different) across all trials of interest? Or does it mean that activity in those voxels simply does not follow either of the two modulation parameters that I have included? That is, maybe activity there is not equal across all trials, but it varies according to some variable that I have not included as an amplitude modulator.

Regards,
Christine