Importing referenced ABR data into EEGlab ?

Post Reply
Alexandre
Posts: 13
Joined: Tue Jul 07, 2009 3:22 pm
Location: Montreal

Importing referenced ABR data into EEGlab ?

Post by Alexandre »

Hello,

I am a newbie and use the ABR recording software (Auditory Brainstem Recording) from BioSemi to carry out psychophysical experiments. When importing my BDF data in EEGlab I get the following message

"We highly recommend that you choose a reference channel IF these are Biosemi data (e.g., a mastoid or other channel). Otherwise the data will lose 40 dB of SNR! "

This issue had been adressed previously in the forum but only partially and remains unclear.

With BioSemi ABR I record using 3 electrodes (Cz and both Mastoids) and I end up having two channels in my recording, which I guess are two referenced channels.

So my question is : "Do I need to care at all about this warning and if yes what should I do ?"

Thanks a lot in advance

Alex

Coen
Site Admin
Posts: 1124
Joined: Fri Mar 26, 2004 7:00 pm
Location: Amsterdam, Netherlands
Contact:

Post by Coen »

The warning about referencing applies to all data recorded with the standard ActiView software (AD-box speedmodes 0-8), but it does not apply to ABR recordings (speedmode 9)

With the ABR software, the signals on file are indeed already referenced, and should be analyzed without additional referencing in the review software. The BDF file from an ABR recording always contains 3 channels: 2 bipolar leads (Left versus Ref, and Right versus Ref), and the status/trigger channel. A further difference between the ABR and standard recordings is that the ABR signals on file are high-pass filtered (-3 dB at 100 Hz), whereas the standard data is always DC (regardless of the filter settings in ActiView). So, electrode offstes cannot be retrived afrterwards from an ABR data file. Finally, the scaling is different: ABR uses a LSB value of 0.4883 nV (1/2048th uV), whereas standard data uses LSB = 31.25 nV (1/32th uV). The review software should adjust automatically to the different scaling factors (the necessary information is in the BDF file header), but it is probably good to check anyway.

Best regards, Coen (BioSemi)

Alexandre
Posts: 13
Joined: Tue Jul 07, 2009 3:22 pm
Location: Montreal

Post by Alexandre »

Thanks, I will try this out !
Best

Post Reply