Dear Vasudev,

The two analyses encompass different number of connections, so they have different family-wise correction criteria. In particular, the analyses in the 'preview' display are specific for each selected seed-ROI as they only include connections between the selected seed ROI and all other ROIs (e.g. N-1 connections, where N is the total number of ROIs), so these analyses are best/most-appropriate when you want to analyze the connectivity with one given (a priori) seed ROI. In contrast, the analyses in the 'results-explorer' window consider all pairwise connections between pairs of ROIs (e.g. a total of N*(N-1)/2 connections), and because of this they typically require a more conservative criterion for family-wise error correction. These analyses are best/most-appropriate when you don't want to constrain your analyses to a single seed ROI and prefer to include all connections among a set of ROIs (e.g. when you want to analyze the entire connectome; see also http://www.conn-toolbox.org/fmri-methods... to learn more about the different types of family-wise error correction methods used in CONN for this type of analyses)

Hope this helps
Alfonso


Originally posted by Vasudev Devulapally: