I see that in the code (and in the GUI) the "coherence" and the "phase locking values" are two different things. While I see where this comes from, and while the implementation is correct, I wonder if it would not be better (conceptually, and maybe practically) to unify these two, or to move some measures from the phase to the coherence part.
This is particularly true for the measures using the imaginary part, for example the weighted phase lag index is usually computed from the cross-spectrum, etc.
1 Like
Thanks Daniele: I think the logic based on the name of the metric (i.e., "phase") should prevail wrt how it is methodologically derived.
Sorry if I am missing the point.
You aren't missing any point 
There are many perspectives, if a user goes for a measure as if they would go in the index of a book (which is definitely the case sometime), then it's ok that measures with "phase" in the name are grouped.
What's interesting/instructive (and could save the users some time) is that they are potato poteto.
See for example the figure below, from
Bruña R, Maestú F, Pereda E. Phase locking value revisited: teaching new tricks to an old dog. J Neural Eng. 2018 Oct;15(5):056011. doi: 10.1088/1741-2552/aacfe4. Epub 2018 Jun 28. PMID: 29952757.
In which the measures are more similar across column than across rows
1 Like