Combining Channels via a subtraction method.

  • MNE-Python version: 0.23.0

For BioSemi EEG systems you have to re-reference the EEG signal via subtracting one mastoid channel “EXG5” from the other “EXG6”. I intend to do this via the mne.channels.combine_channels function, but I can’t quite work out how to write a lambda function to accomplish the subtraction of one channel from the other. Apologies for the seemingly easy question, but any advice would be really useful!

Hello, for adding a new channel that is the difference of two existing channels, you can use mne.set_bipolar_reference().

Hi Richard,

Thanks for the speedy reply. Can this also be used to combine two EOG channels or will it set these as the reference?


This should also work with EOG channels, if I’m not mistaken!

@julesneuro subtracting one channel from another will not have any impact on your remaining channels, because they are not connected in any way. All you’ll end up with is a new channel that is the difference of EXG5 and EXG6. For that, the method that Richard proposed will work, regardless of the channel type.

Perhaps you intended to afterwards re-reference all your remaining channels with respect to the newly created “subtraction channel” (EXG5 - EXG6)?

Note that re-referencing EEG channels according to some EOG channel (subtracted or not) does not really make sense, because the EOG is a different kind of signal (dipoles of the eyeballs, not brain activity)

1 Like