help > ICA component's correlation to templates
Showing 1-3 of 3 posts
Display:
Results per page:
Feb 19, 2020  08:02 AM | syq
ICA component's correlation to templates
Dear all,


Once computing the correlation to templates via ICA tools,

how do we know whether the correlations are significant ,is there p-values provided, and where?

Thanks in advance

Best regards
Yiquan
Feb 19, 2020  08:02 PM | Alfonso Nieto-Castanon - Boston University
RE: ICA component's correlation to templates
Dear Yiquan,

Sorry, those spatial correlations are simply to help in identifying/labeling individual networks of interest in your data, but there is no significance or p-values associated with them. Often people use correlations with these or alternative network templates simply to identify the component with the highest spatial correlation with an a priori network of interest. Alternatively (e.g. if your network of interest appears to be broken down into perhaps multiple components) one common alternative is to select the component(s) with the highest loading at a series of apriori spatial coordinates

Hope this helps
Alfonso
Originally posted by syq:
Dear all,


Once computing the correlation to templates via ICA tools,

how do we know whether the correlations are significant ,is there p-values provided, and where?

Thanks in advance

Best regards
Yiquan
Mar 5, 2020  01:03 PM | syq
RE: ICA component's correlation to templates
Dear Alfonso,

I see. Thank you very much :)

Best wishes
Yiquan