Imaginary coherence output

[gianmarcoduma]

Dear brainstorm community, I’m using imaginary coherence NxN with EEG data, 32 electrodes. The output is fantastic, so first of all thanks for your gigantic work, but I’m finding some problems in the interpretation.
Chance are is more usful if I describe first my purpose in the research and then the problem I’m finding. I want to compare the changes in the coherence, between 2 conditions, in the frequency bands(delta, theta, alpha, beta). I excluded purposely gamma frequency.

Whe I run the function, I must put a maximun frequency resolution, so if I put 2Hz, and I want to study coherence until 30Hz, I will have 15 different output. Is there a way to group the data in the classic frequency bands( delta, theta, alpha, beta), or should I check manually basing on the maximun frequency band resolution?

When I export data, it is possible to export only the coherence matrix, but this is a matrix NxN channels, without the information about which frequency coherence is being reported. Is there a way to export the information about coherence each maximum frequency resolution step?

Moreover, it is possible to see how coherence changes on the scalps, each step of the maximun frequency band resolution, only when 1xN coherence is plotted, but there is not the possibility to do with NxN. Given that I cannot report in a paper the coherence of 1xN plot for each channel, how could I get an image of the global changes in the coherence for each frequency bands?

I know that this question are not very precise, and that in general the metrics of the connectivity data are very complex, but I hope you could give me some suggestions so that I could work on it.
Thanks
Gian Marco

[Francois]

Hi Gian Marco,

Is there a way to group the data in the classic frequency bands( delta, theta, alpha, beta), or should I check manually basing on the maximun frequency band resolution?

You can use the process "Frequency > Group by time or frequency bands", but you would need a frequency resolution much higher than 2Hz, otherwise, you wouldn't even have one value per frequency ban.

When I export data, it is possible to export only the coherence matrix, but this is a matrix NxN channels, without the information about which frequency coherence is being reported. Is there a way to export the information about coherence each maximum frequency resolution step?

If you look in the coherence .mat file, you can get the frequency vector in the field .Freqs.
To reconstruct the NxN matrix for each frequency bin, see this forum thread:

How the Corr matix is saved?

I calculate the correlation matrix by the way NxN. when I check the saved file content, I did not find a N*N matrix. instead of it, I find a variable TF , stored M * 1 matrix. My question, how you experts stored the correlation matrix? In which way? where the pvalue associated with Correlation matrix is stored? How the cohr and other connectivity metris store the results. I tried to do further analysis , so I ask such question. In fact, your software is user friendly and almost perfect. …

Moreover, it is possible to see how coherence changes on the scalps, each step of the maximun frequency band resolution, only when 1xN coherence is plotted, but there is not the possibility to do with NxN. Given that I cannot report in a paper the coherence of 1xN plot for each channel, how could I get an image of the global changes in the coherence for each frequency bands?

The popup menus "View as image" and "View as graph" represent the NxN connectivity matrices for one frequency.
The popup menu "View as spectrum" shows all the values at once, but gets really messy if you have too many pairs of signals.
There are currently no menus to represent the full connectivity graph on the anatomy of the subject.

Cheers,
Francois