multi person, multi channels SSVEP on OpenBCI

SeventeenChenSeventeenChen Shenzhen
edited July 2019 in Cyton
Hi, everyone, 
We have tried to observe multi-channels SSVEP on computer with Cython, the details of our experiment are following:
1. two people , channal 1 and channel 2
2. Oz, Cz, AFz; Oz is the measurement connected to N1(N2 fot the other), Cz is the reference connected to P1(P2), AFz is GND connected to BAIS. The connection is like 'Shark Attack' experiment.
The question is:
We found the signal in one channel could influence the other, which should be calculated individually, for example, if we found a peak in channel 1, there could be the same peak in channel 2, but the peak shouldn't appear in fact. Why does it happen? How could we analysis the mutili-channal SSVEP individually with Cython?


Regards,
Chen

Comments

  • BillhBillh
    Usually the peak that should not appear in more than one place is due to a change in the surrounding environment that affects both persons, and multiple leads and channels. There is also some possibility of cross talk between channels. The crosstalk should be removable by subtracting Cz from Oz on each person.
  • wjcroftwjcroft Mount Shasta, CA
    Bill, thanks.

    Chen, hi. Have you adjusted your OpenBCI_GUI settings to disconnect SRB2 (and SRB1) from the common reference bus? And also disabled the Bias (inverse common mode signal) injection? I think in your case you only want to use Bias as a ground, and give up on the common mode injection, since you are using two subjects. I believe Chip's Shark Attack setup used such settings. If you are not receiving the data with the GUI, (for example Python), then you need to send the appropriate SDK serial port commands to affect these settings.

    Regards,

    William

  • SeventeenChenSeventeenChen Shenzhen
    edited July 2019
    Hi, William,
      We have tired to adjust my OpenBCI_GUI settings to disconnect SRB2 (and SRB1) from the common reference bus, but it seemed make no sense.
      We have read the website about cyton SDK, but we don't know how to programming the serial port. 
      Could you please help me with the situation? 
      In our experiment, we used matlab to get data and process the signals.

    Regards,
    Chen

  • SeventeenChenSeventeenChen Shenzhen
    Hi, William, 
      Could you tell how to programming the cyton SDK to control the serial port?

    Regards,
            Chen
  • wjcroftwjcroft Mount Shasta, CA
    Chen, I don't understand. Are you accessing the serial port from the GUI or from your own program? If using the GUI, adjust the Settings using the GUI menus. Just turn off both SRB1 and SRB2 and also the Bias.

    Your pasted Images (which were links to your C: hard drive), don't upload here. If you need to upload an image use a file service such as Google Drive.

    Serial port commands for adjusting the Cyton (for example from Python) are here:


    Please go through all the steps of the Cyton tutorial, it shows how to run the Settings control panel. In the tutorial for ECG, EMG, those channels are disconnected from SRB and Bias.


    Regards,

    William


  • SeventeenChenSeventeenChen Shenzhen
    Hi,William,
      Thank you~~,and I have tried to connected to serial port with openbci_matlab.py, in chip, he didn't turn BAIS off, and we have tried to turn BAIS off  or  keep BAIS on, but both made no sense.

    Regards,
    Chen
  • wjcroftwjcroft Mount Shasta, CA
    Try it from the GUI Settings panel. If signals look good there, you may need to adjust your Python code.
  • SeventeenChenSeventeenChen Shenzhen
    Thanks,William,
    We have tried to change the setting it in GUI, but it sitll can't work well

    Regards,
    Chen 
  • wjcroftwjcroft Mount Shasta, CA
    Well, then Bill made some valuable suggestions. Be sure your cables are as short as possible and the cables for person#1 do not run along side or twisted with cables for person#2.
  • SeventeenChenSeventeenChen Shenzhen
    William,thanks,
    The cables were connected to cython individually. 
    The closing eyes effects were observed in two channel individually without the interference of the other. But the SSVEPs displayed on outr computer could be affected by the other. When one student look at the flicking LED and the other was rest, we can see paeks in both #1 and #2.

    Regards,
    Chen 
  • wjcroftwjcroft Mount Shasta, CA
    Have you tried disconnecting the ground from one of the subjects? The ground function of the Bias lead, is just centering the differential amplifiers. With Bias turned off in the settings it still works as ground, but does not inject a counter common mode signal. Just grounding one person may be enough. There is really only one ground potential anyway; since both subjects are using the same amplifier. You might be seeing some ground loop effects.

    You might also try shielded cables.

    Don't forget to try what Bill said.
  • SeventeenChenSeventeenChen Shenzhen
    William,thanks,
    Good idea, and we tried your suggestion at once, it made just a little sense, but waveform synchronization still existed, compared with connecting two subjects' ground, the 

    phenomenon disappeared a little.


    Regards,
    Chen


  • BillhBillh
    Do you still see the synchronization with one person with eyes covered (no light)  and the other observing the flashing synchronizing light? Perhaps there is some leakage of light blinking.

Sign In or Register to comment.

© Vanilla Keystone Theme 2026