[Eeglablist] Why is Phase Coherence Greater than 1?

[Ramón Martinez]

Hi,
I've been checking the function and I think there is no big issue here...
probably only the label. What we are getting here is the cross-spectrum.
This function make a check  for continuous data (single trial, no epoched)
and then, if positive, calculate the cross-spectrum.. 'not the coherence'.
In this case, when cross spectrum is calculated, probably does not have
sense to call  'Coh' to the marginal mean on the lower side...we need to
change this for that option. But.. anyway.. I'm just informing to not rise
the red flag on this.
 Regards,
 Ramon


On Thu, Jul 24, 2014 at 10:26 AM, Makoto Miyakoshi <mmiyakoshi at ucsd.edu>
wrote:

> Thanks Arnon for the deteiled report.
> Ramon, do you think this is related to the bootstrap issue again?
>
> Makoto
>
>
> On Thu, Jul 24, 2014 at 10:22 AM, Arnon Yang <abarnyeu at hotmail.com> wrote:
>
>> Hi Makoto,
>>
>> Thanks for the quick reply.
>>
>> The steps I took were this:
>> 1. Import Netstation binary file into EEGLab.  (The file has only 1 epoch
>> and filters and ICA have not been ran yet.)
>> 2. Plot -> Time-frequency -> Cross-coherence pop_newcrossf()
>> 3. I compared channels 1 and 2, in an epoch range 100 ms after the start
>> and 100 ms before the end.  Log scale for frequencies was set and phase
>> coherence was used.  Bootstrap significance level was set to default and
>> 'padratio' = 1.
>> 4. From the output diagram, phase coherence reaches values as high as 1.3
>> e+007, as shown by the color side bar on the upper panel.
>>
>> I've attached screenshots of pop_newcrossf() and the resulting figure.
>>
>> Thanks very much, Makoto.
>>
>> Arnon
>>
>>
>>
>> ------------------------------
>> From: mmiyakoshi at ucsd.edu
>> Date: Thu, 24 Jul 2014 10:01:38 -0700
>> Subject: Re: [Eeglablist] Why is Phase Coherence Greater than 1?
>> To: abarnyeu at hotmail.com; nucleuscub at gmail.com
>> CC: eeglablist at sccn.ucsd.edu
>>
>>
>> Dear Arnon,
>>
>> Thanks for reporting it. Please describe how to replicate it in our
>> environment so that we can work on it. Thank you for your cooperation.
>>
>> Makoto
>>
>>
>> On Wed, Jul 23, 2014 at 11:12 AM, Mister Arnold <abarnyeu at hotmail.com>
>> wrote:
>>
>> Hi all,
>>
>> I have been trying to compare the phase coherence (phase locking value)
>> of 2 channels of the same data set.
>>
>> From my understanding, coherence should always be a value between 0 and
>> 1.  However, when I use the pop_newcrossf() function to plot the coherence,
>> I obtain values which are as large as 1.30 e+007.
>>
>> Is this expected and how can I change it back to a value that is between
>> 0 and 1?
>>
>> Any suggestions is appreciated.  Thanks very much.
>>
>> Arnon
>>
>>
>>
>>
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
>>
>>
>>
>>
>> --
>> Makoto Miyakoshi
>> Swartz Center for Computational Neuroscience
>> Institute for Neural Computation, University of California San Diego
>>
>
>
>
> --
> Makoto Miyakoshi
> Swartz Center for Computational Neuroscience
> Institute for Neural Computation, University of California San Diego
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20140724/58fee4f2/attachment.html>