**RE: CONN functional toolbox conn_withinbetweenROItest question**

*Boston University*

Hi,

Of course, to do that you may edit the conn_withinbetweenROItest.m file and change lines 69-71 to consider the absolute value of the connectivity strengths there, e.g.

Z_within1=permute(sum(sum(abs(Zrois_within1),2),3)./sum(sum(~mask_within1,2),3),[1,4,2,3]);

Z_within2=permute(sum(sum(abs(Zrois_within2),2),3)./sum(sum(~mask_within2,2),3),[1,4,2,3]);

Z_between=permute(sum(sum(abs(Zrois_between),2),3)./sum(sum(~mask_between,2),3),[1,4,2,3]);

Hope this helps

Alfonso

*Originally posted by tlohman:*

Hi everyone,

I'm new to Conn, and have been using the conn_withinbetweenROItest function to find the average correlation between ROI pairs in various networks.

As far as I understand it, the function leaves the sign (+ or -) when it averages the correlations. So in a hypothetical network with only 4 ROI pairs, if two ROI pair's correlation is -.5, and the other two ROI pair's correlation is .5, then the resulting value would be 0.

Is there a way to run the same function but have it average the absolute values of all correlations in the RRC matrix?

Any help is appreciated!

## Threaded View

Title | Author | Date |
---|---|---|

tlohman |
Apr 3, 2024 | |

Alfonso Nieto-Castanon |
Apr 11, 2024 | |