[Mne_analysis] PSI, frequency bands, and edge artifacts

Mina Jamshidi jamshidi at cbs.mpg.de
Thu Sep 3 05:03:08 EDT 2020
Search archives:

        External Email - Use Caution        

Hi Josh,

I have not checked the function for MNE. but the original function form Guido Nolte uses FFT, and then takes the desired frequency bins. If the segments are very short (i.e. time domain multiplied by a very short rectangular window), then we know that at the frequency domain the frequency representation of the signal is convoluted by a sinc function, which makes the frequency representation not so perfect! A solution to alleviate this problem is to multiply the segments by a window shape like hanning, or others. 

I suggest you to first check inside the respective functionin MNE, check how it is done there. 

Best,
Mina

----- Original Message -----
From: "Alexandre Gramfort" <alexandre.gramfort at inria.fr>
To: "Discussion and support forum for the users of MNE Software" <mne_analysis at nmr.mgh.harvard.edu>
Sent: Wednesday, September 2, 2020 9:19:38 PM
Subject: Re: [Mne_analysis] PSI, frequency bands, and edge artifacts

External Email - Use Caution        

hi,

I don't reallly know. I would try a simulation to see how it behaves.

maybe someone else has more experience

Alex

On Wed, Sep 2, 2020 at 5:45 PM Bear, Joshua <JOSHUA.BEAR at cuanschutz.edu> wrote:
>
>         External Email - Use Caution
>
> Dear MNE Community,
>
> When you run PSI (or other connectivity measures) on very short epochs within a given frequency band (using fmin and fmax rather than filtering beforehand), how does the function limit to those frequencies? Is there still a risk of introducing edge artifacts? Is there a common approach to quantify this risk?
>
> Thank you,
> Josh
>
>
>
>
> _______________________________________________
> Mne_analysis mailing list
> Mne_analysis at nmr.mgh.harvard.edu
> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis

_______________________________________________
Mne_analysis mailing list
Mne_analysis at nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis



More information about the Mne_analysis mailing list