Hello Keith,

There are easy ways to create phase and correlation maps with AFNI. The program 3ddelay (and its plugin version) was designed to do just that. The algorithm uses the Hilbert Transform of the cross-correlation function to determine the delay between a reference time series (modeling the stimulus) and each voxel's response. For each voxel, you'll get 4 parameters, delay, covariance, correlation coefficient and variance. The delay is the same as phase if your stimulus is made up of one (or mostly one) sinusoid, the covariance is a measure of the response's amplitude and the correlation coefficient can be used to determine the significance of activation. A lot more detail can be obtained with the -help option which also cites literature references on the algorithm and on delay estimation in FMRI.

If you prefer to use your own methods, you could import your matlab results to AFNI using AFNI's matlab library found here: [afni.nimh.nih.gov]

At this point I should mention that I wrote most of the tools I have just shamelessly promoted above.

cheers,
Ziad