Hi Alfonso,

Thankyou for this wonderful script (conn_withinbeteenROItest). I have also used this script for exploring within and between network connectivity for three resting state networks; two networks at a time. I had four groups in my study and as per suggestions, using CONN's gui second-level results tab, ANOVA was implemented for between subjects contrast. After specifying the ANOVA model [1 -1 0 0; 0 1 -1 0; 0 0 1 -1] and selecting three networks in second level results tab, I ran the script "conn_withinbetweeROItest". This promted me to select, two networks which further resulted in

Within matrix Set-1 (4 x 4 ROIs; 12 valid ROI pairs; 99 subjects);
Within matrix Set-2 (7 x 7 ROIs; 42 valid ROI pairs; 99 subjects);
Between matrix (4 x 7 ROIs; 28 valid ROI pairs; 99 subjects;
Set-1 Within-network test (average effect between Set-1 ROIs):F(3 90 )=0.79 (p=0.505040)
Set-2 Within-network test (average effect between Set-2 ROIs): F(3 90 )=1.77 (p=0.159140)
Between-networks test (average effect between Set-1 & Set-2 ROIs):F(3 90 )=4.33; (p=0.006720).

The p-value for between network test was then converted into two-tailed p value: 2*min(p,1-p)= 0.0134.

Since the model specified here is ANOVA model, I was wondering, if we need to perform post-hoc tests. Or two tailed p value of 0.0134, which is significant, can be reported.

Thanks and Regards
Himanshu Joshi