Computing noise covariance with/without EMG data?

Discussions
1

Hi Francois,

We want to compute both cortico-cortical and corticomuscular coherence in our analysis in brainstorm. Before importing our data into brainstorm, we combined EEG & EMG data in EEGLAB, created .set & .fdt files, then imported these files into brainstorm. As I was importing these files into brainstorm, I began wondering if the EMG data is included in the noise covariance computation for the EEG data. If this were the case, should the EMG data be included in the noise covariance computation?

Thank you for the help!
Adam

2

The noise and data covariance matrices are useful for the modalities that are used to estimate brain sources, which is not the case of the EMG.
Make sure in the channel file that EMG are correctly labelled as EMG. Same for the EEG.
https://neuroimage.usc.edu/brainstorm/Tutorials/Epilepsy#Prepare_the_channel_file

When computing the noise covariance from these files, only the EEG channels would be used.

Make sure you read these tutorials: