P300 and CMS/DRL

MatS · Post by **MatS** » Thu Dec 07, 2006 4:45 pm

I am currently using a 64 channel Biosemi ActiveTwo system, I have been looking up the information about CMS/DRL since I seem to be getting some peculiar waveforms when I subtract the average of the two ear electrodes from the rest of the electrodes. The waveform looks like the P300 waveform but inverted. At first I thought possibly the data the system recorded was inverted but I read on the forum that the positive=up so that was not it. I then stumbled upon the questions by Ulrich Hoffmann (Uli) whose paper on using boosting for P300 detection I am currently reading. In his work he seems to reference to Oz, O1 and O2. I wish to reference to the ears for now. Others have used common average as reference. Currently my CMS and DRL are located to the left and right of POz, in reference to the effects of CMS/DRL on the signals present what would be a good reference point in your eyes? Furthermore would it be better to move the CMS/DRL positions from where they are to another position when doing experiments concerning the P300? If so where?

Post by **Coen** » Fri Dec 08, 2006 12:19 pm

You refer to a previous P300 thread, in which I have convinced Uli that the location of the CMS/DRL does NOT affect the signals measured between the other electrodes. Above, you suggest that you do not agree with me. Please post your arguments.

Regards, Coen (BioSemi)

MatS · Post by **MatS** » Fri Dec 08, 2006 4:20 pm

I do not disagree with you at all. I was just wondering if anyone who uses your devices has any recommendations in their experience or if you as the manufactures of the device would have any recommendations.

As far as I understand from your explanation the CMS/DRL system should not affect the resulting signal once it has been subtracted from a reference. As I understand it all electrodes including the EXG ones are referenced to the CMS/DRL. So when I subtract the reference electrodes (in my case the ears) then the resulting signal I see at Cz, Pz and Fz is what was at those locations without modification by the CMS/DRL loop (of course any signal on the ears will also be subtracted from those electrode locations as well). I hope I have understood this correctly, please correct me if I have not.

Post by **Coen** » Fri Dec 08, 2006 7:15 pm

There is a distribution of electrical potentials across the head. However, we can only measure potential DIFFERENCES. So, you can not speak about "a signal at Cz" or " a "signal on the ear". You can only speak of a signal (voltage) between Cz and the ear.

ActiveTwo measures the voltages of all electrodes (headcap electrodes as well as EX electrodes) with respect to a point between CMS and DRL, see viewtopic.php?t=33

By subtracting the average of the two ears from CZ, you get the signal between Cz and a virtual point in the center of the head (midpoint of a virtual line between the ear electrodes).

Regards, Coen (BioSemi)

MatS · Post by **MatS** » Fri Dec 08, 2006 8:01 pm

Yes of course you are absolutely correct about their is no signal at Cz but more a distribiution of signals which are measured at Cz. Thanks for the clarification on the operation of the CMS/DRL loop.

I liked your explenation of CMS/DRL here: http://sccn.ucsd.edu/pipermail/eeglabli ... 00930.html

Don't know where it originaly came from.

Regards, Mat