If I have a simple fMRI motor (hand squeezing) task, what is the simplest tool to use to make functional connectivity estimates between, for example, the primary motor and primary somatosensory activation foci?

Wouldn't the simplest approach would be to extract mean intensity data for the two regions or peaks and compute the correlation?

Yes, operationally that is simple, but which 'tool' makes this easiest for me to do, by the time I create the ROIs, extract the signal, and perform the correlation? FSL, SPM, BrainVoyager, AFNI, other???

PLSGUI has a nice built-in intensity extraction feature for voxels / regions and so does BrainVoyager, but I think SPM requires the MarsBar add-on and requires some fiddling. I'm not sure about AFNI. I think these tools would be about equally difficult to use as the bulk of the preparation time will be in programming your design matrix. That said, if you go with PLSGUI, you have the advantage of being able to set up exploratory connectivity analyses later without much extra effort (e.g., task analysis, seed analysis).

Excellent suggestions. In the spirit of the NITRC projects, let's annotate how to get at these various tools you mention:

PLSGUI - hmmm, not in NITRC! It should be! But no worries, it's at http://www.rotman-baycrest.on.ca/index.p...

BrainVoyager - also not in NITRC, darn. I keep asking them, but to no avail as yet. If anyone else can try to suggest their adding a project, feel free. Still no worries, it's at http://www.brainvoyager.com/

SPM - ah yes, here we go: http://www.nitrc.org/projects/spm/

MarsBar - alas, not at NITRC, but I've pinged them on occasion, so if anyone has any influence there, feel free to exert it. https://sourceforge.net/projects/marsbar... Since it's SourceForge, it'll be easy to include in NITRC!

AFNI - yeah! http://www.nitrc.org/projects/afni/

I like the connectivity toolbox by Sue Whitfield-Gabrielli: http://web.mit.edu/swg/software.htm. It does preprocessing and seed based analyses.

I have just added the latest version of Sue's functional connectivity toolbox to the NITRC site: http://www.nitrc.org/projects/conn . In addition to extracting the ROI timecourses and estimating ROI-to-ROI correlation measures, it performs seed-to-voxel analyses (maps of functional connectivity values between an ROI and every voxel), includes a pre-processing step that removes confounds (regressing-out motion parameters and noise signals obtained from white matter and CSF areas), and performs a number of second-level random effect analyses. Hope you find it useful