Hi CONN users,

I wanted to display a seed-to-voxel results on a surface map. The beta is below zero. however, when I displayed it on surface, the color bar shows positive value (see attached thalamus_CL.fig). I assume that the activation on surface map should be in the same direction as the beta, or did I set up anything wrong?

The result cluster locates at Thalamus, so I only displayed the sub-cortical volume. I wonder if this could be the reason why the figure looks weird.

Thank you!
Meishan

Hi Meishan,

That's strange, how are you creating that display, is it from the second-level results explorer window using the 'Volume Display' button, or is it manually from the command-line?

Best
Alfonso
Originally posted by Meishan Ai:

Dear Alfonso,

Yes I created it from the result explorer -> Volume display --> set show subcortical ON, show cortical OFF. I just realized that I was using the wrong file - I used the Cluster_ROI.nii rather than Cluster.nii, so that's why my color bar is from 0 to 1. However, I selected the .nii file generated from conn and display it in the Volume display instead, but there's nothing showed up.

I also looked at the cluster.nii file in FSL eyes, and it seems that the values within that cluster were positive (see attached figure). But this still does not explain why I can't see anything in CONN display.

Just to clarify more, the seed-to-voxel I performed is to see the effect of VAR on the seeded functional connectivity of my ROI. The negative beta should indicate the negative relationship between my VAR and the functional connectivity. So I think maybe the positive plot I got is actually the positive connectivity between ROI and the result cluster, rather than the relationship between seeded connectivity and VAR?

Thanks so much!
Meishan


Originally posted by Alfonso Nieto-Castanon:

Dear Alfoso,

I'd like to follow up with this previous question. I tried to display the cluster on the surface. I selected the cluster I wanted to display in the result explorer, however, when I open the surface display, and selected "show color bar ON". The color bar appeared with the correct value, but the cluster did not show up on the surface. (see attached screenshot)

The cluster is on thalamus - would this be the reason why I can't display it on the surface? Thanks! It is the same cluster in the original post.

Thanks a lot!
Meishan

Originally posted by Alfonso Nieto-Castanon:

Dear Meishan,

Right, the reason why the cluster does not show up when using the 'surface display' option is because this option will only show activation near the cortex (and it will display this activation projected to the cortical surface). Similarly, the reason why the 'volume display' option does not show an activation colormap is because this option will threshold the 3D field of statistical values using the threshold specified in the results explorer window and it will then display the resulting 3D cluster. This cluster is shown as an opaque object, and the surface of this object is -by definition- an isosurface of the statistical map so all points on that surface have exactly the same statistics (up to voxel-precision rounding). You are right that perhaps it might be a better idea to display those supra-threshold clusters as pseudo-transparent objects, instead of opaque objects, so that we can see the different "strength" of statistical values inside each different cluster. I imagine that could be done, for example, by manually specifying different threshold values, then generating for each threshold value a separate display/figure, and last by simply averaging/merging the resulting .jpg figures to create that pseudo-transparency effect where higher statistical values in the interior of the clusters will be shown in darker red/blue and lower statistical values in the surface of the clusters will be shown in lighter red/blue values? 

Hope this helps
Alfonso

Originally posted by Meishan AI: