Some questions about triggers / markers with OpenBCI_GUI and Cyton board.
Hellow everyone,
I'm a researcher new to OpenBCI. I'm trying to use Cyton (perhaps) with Daisy to record EEG while the subjects watching visual stimulus (videos and pictures). I read the documents and former threads and still have some questions need your help.
First, as the firmware of my cyton board is 3.x.x, does it means I just need to open the digital/analog read widget and turn digital/analog read on, the GUI will send a command to the board to switch it to digital/analog read mode? So I don't need to anything else as the 'hard way'? And could I save the configuration and load it to switch the board mode more easily (as not just me will use the system in my institution) . Or I need to switch it everytime?
The second question is about the data format outputed by the OpenBCI GUI. I read the csv outputed and found that the 1st to 31st channel seemed to be used for contain the data of Bytes2 'Sample Number' to Bytes32. And there are 7 'Other' channels, the value of the 1st 'Other' channel is always '193' . Would you please tell me what it means? And what are the 2nd to 7th 'Other' channels used for? The 5th is connected to the D17 pin by default, so are the other channels connected to D11, D12, D13 and D18?
The third question is about the external trigger. How many external triggers I could use at the same time? As there are 5 pins, I think I could use 5 external triggers if I use the usb dongle. But accordiing to the csv format, it seems that one 'Other' channel is not be connected to the D1x pins. Is it possible to use this channel (I guess it maybe the 2nd 'Other' channnel)?
The last question is about how to send a 'marker' in the data stream. There might be a latency if I use a photosensor to record when the stimulus is presented on the screen. And sometimes the cables might be inconvinient. I think it will be better to send the 'marker' to the Cyton board to record the timing more precisely, or send the 'marker' to the dongle instead. And is it possible to send this kind of signal to Cyton or to the dongle? I usually use psychopy to present the stimulus.
Sorry to be a long article. Best wishes.
Comments
Tiger, hi.
Can you explain the goal of your project, what are you trying to measure / demonstrate? What are you attempting to synchronize with the external trigger(s) or 'markers'? The external trigger approach always gives more accuracy, because the trigger data, recorded in the Cyton 'Aux' channels, is sampled at the exact same time as the channel data. 250 Hz in the case of Cyton 8 channel alone.
'Markers' are inserted after the fact on your laptop, so do not have this degree of synchronization. Why do you assume you need 5 unique external triggers? This sounds like more than are typically used. This page mentions where the Aux data is located in the CSV recording file.
https://docs.openbci.com/Software/OpenBCISoftware/GUIDocs/#cyton-other-channels
Instead of using the GUI to collect your data, you could instead use Brainflow and your language of choice.
https://brainflow.readthedocs.io/en/stable/Examples.html
The Aux channels can be turned on in either 'digital' or 'analog' mode using SDK commands sent by Brainflow
config_board()method.https://openbci.com/forum/index.php?p=/discussion/2703/sending-cyton-sdk-commands-with-brainflow-resolved
Regards, William
No, the photosensor will give you the most accurate synchronization with screen stimulus. As stated before marker approach will have latencies. Which could vary based on processor loading, buffering, etc.
Hi William,
Thank you for your advances.
I have find out the connection between 'Other' columns and D1x pins.
I and my colleague usually use 1 or 2 external triggers to distinguish the stimulus. As our main purpose is to recording the EEG changes during watching some kinds of videos, the conditions would be quite complex. So it will be very helpful to use mutiple external triggers to mark when the 'event' happended in the movies.
BTW, I find whether or not to fix the FTDI buffer could make the latency quite different, from 500ms to 75ms. I suggest you can move the 'FTDI Buffer Fix' page to 'Setting up for EEG' chapter.