EEG REF Channel question

[mcp0228]

Hi BST community,

I am currently using Biosemi ActiView system to acquire auditory brainstem responses (ABRs).
The Biosemi ActiView system saves the ABR signals in the following manner. The below is an excerpt from the Biosemi ActiView ABR user manual.

As you can see from the above, when the ActiView software saves an ABR signal it saves it in an already-referenced state (bipolar form; Left and Right channels) with respect to the REF channel.

Now, the issue becomes when you import this BDF file into Brainstorm.
The following is what you see when you first import this ABR BDF file into Brainstorm.

Brainstorm, by default, labels Lmon (monopolar) and Rmon (monopolar) as EEG with Ref as EEG REF channel.
This is incorrect as only the Left and Right channels (bipolar) should be used.
Therefore the channel types were modified to the below.

The Left and Right channels are now EEG (as it should be) and Lmon, Ref and Rmon are now labeled as MISC to exclude it from further analysis and viewing. Below is the resulting ABR.

What I find difficult to understand is that IF you change the channel type of Ref channel into EEG REF like the below (i.e. effectively referencing twice as the BDF file is ALREADY referenced to the REF channel when it gets saved), there is no difference in the ABR trace.

What are your thoughts about this? To me this seems to indicate that specifying the Ref channel type as EEG REF did not do anything and therefore I was wondering what the purpose of the EEG REF channel type would be.

[Raymundo.Cassani]

@mcp0228

Brainstorm, by default, labels Lmon (monopolar) and Rmon (monopolar) as EEG with Ref as EEG REF channel.
This is incorrect as only the Left and Right channels (bipolar) should be used.

In Brainstorm, channel types are inferred from the information that is the raw file (.BDF in this case).
As such Lmon and Rmon are labelled as EEG and Red as EEG REF. Since Left and Right are already re-referenced, their information does not match with regular EEG channel then they are labelled as MISC (other channels)

@mcp0228

Therefore the channel types were modified to the below.

This is correct

@mcp0228

What I find difficult to understand is that IF you change the channel type of Ref channel into EEG REF like the below (i.e. effectively referencing twice as the BDF file is ALREADY referenced to the REF channel when it gets saved), there is no difference in the ABR trace.

Not really, the EEG REF channel type is only a label used to keep track of the channel type in the raw file. I.e., re-referencing is not performed. You could try to explicitly re-reference Lmon and Rmon to the Ref channel, and such signals should look like Left and Right.

[mcp0228]

Interesting - thanks for that answer!

@Raymundo.Cassani

channel type is only a label used to keep track of the channel type in the raw file

So then is the purpose of having different channel types only to label each channel?

If you want to re-reference any data I am guessing you do that either through custom (re-referencing) montage or by using the process “Re-reference EEG”? Would this be correct?

[Raymundo.Cassani]

@mcp0228

So then is the purpose of having different channel types only to label each channel?

The channel types are useful as they allow to grouping channels, for visualization and for processing.

@mcp0228

If you want to re-reference any data I am guessing you do that either through custom (re-referencing) montage or by using the process “Re-reference EEG”? Would this be correct?

That's correct!