Dear Andre,
You can compare PLV between conditions but it might be a good idea to
transform it first to [-Inf Inf] range with something like:
arctanh(2*PLV - 1)
You can do that with imcalc. You cannot compare PLV to zero since it's
non-negative. The transformation won't help with that since then the
test will be whether the PLV is greater than 0.5 which is not very
meaningful either. If you can somehow generate surrogate data
conforming to your null hypothesis then you can compute the PLV for it
and compare to real data.
Best,
Vladimir
On Tue, Apr 6, 2010 at 11:33 AM, Andre C. <[log in to unmask]> wrote:
> Hello,
>
> In my EEG experiment I have calculated the PLV values for the phase
> estimates. I would like to know if I can use regular statistics on
> these values. For example, can I use a paired t-test to compare the
> PLV in different conditions? Moreover, can I use simple t-tests to
> know when the PLV is significantly different than zero?
>
> Thanks in advance.
>
>
> --
> Andre M. Cravo
>
> Ph.D. Student
> Roberto Vieira Laboratory of Sensory Physiology
> University of São Paulo
>
> Visiting Student
> Brain & Cognition Laboratory
> Department of Experimental Psychology
> Oxford University
>
|