[Eeglablist] Synchronization of EEG data with auditory stimulus

[ali zahedi]

Hi Tarik,

Thank you very much.
Yes I am trying to find some documantion for Psychtoolbox, and also for
LSL. I will also share it with the list.

I want to ask another question about the location of reference and ground
electrodes. My device has 32 EEG channels and I may use 19 of them, or
perhaps all of them. Also, a separate electrode is marked as a Ground. But
there is no separate electrode as A Reference. My question is that where
should I place the ground electrode to record my EEG data?
Also, do I need to select one of the 32 EEG electrodes as a reference? (In
this case I have maximum 31 EEG electrodes).
And where should I place it?

Thank you very much in advance for your support.

Regards,
Ali



On Sat, 21 Apr 2018, 18:16 Tarik S Bel-Bahar, <tarikbelbahar at gmail.com>
wrote:

> Hi Ali,
>
> This process can be pretty tricky, and you're better off chatting with an
> engineer hardware specialist, since the millisecond precision synchrony of
> your experimental events in relation to the EEG recording are critical for
> valid results in your case.
> In situations where researchers seek extreme accuracy, they often rely on
> precision instrumentation and displays, as well as important things that
> allow accurate alignment of all signatures, ideally to to one main clock.
> Timing issues and hardware are not often discussed though, and in some
> cases sidestepped, so kudos to you.
>
> Psychtoolbox and other matlab and python based tools (see also Kothe's Lab
> Streaming Layer)
> all have extensive documentation where one can learn how to use them. One
> can also google for
> specific things like "Psychtoolbox code for synching with eeg amplifier",
> or similar searches. I think LSL might work well for you.
> Perhaps an easier alternative is to simply find a way to translate your
> "real clock time" into "EEG time" accurately. If you can just do that, load
> the translated beep times into your EEG after recording as new events.
> Here's perhaps something easier too: if your amp has the capability of
> adding a sensor into it's output, the plug in an audio sensor and record it
> with your amp. Your EEG recording software may also have an option for such
> an integration.
> Good luck with your EEG timing adventure. You may hear from some of the
> timing folks on the list soon too.
> When you have a solution, please share with the list so others can benefit
> from your adventure.
>
> On Fri, Apr 20, 2018 at 11:01 AM, ali zahedi <ali.zahedi.bham at gmail.com>
> wrote:
>
>> Dear Tarik,
>>
>> Thank you very much for the information you have provided, and sorry for
>> my late reply.
>> Can you please give me the detail on how I can use psychtoolbox and sync
>> it up with my EEG amplifier?
>> (I am really new in working with EEG hardware...).
>>
>>
>> I tried to find GUI in psychtoolbox but I only found a demo to produce an
>> audio.
>> Actually I do not have any idea to work with psychtoolbox and use it to
>> sync the beeps with my EEG data.
>> My beep generator (written in c#) works in real clock time and I get the
>> beep times in real clock time. However, my EEG data does not work with real
>> clock time and it has its own timer starting from 0 anytime I start
>> recording.
>>
>> I really appreciate your support and I would be very grateful if you
>> could guide me to use psychtoolbox with more details.
>>
>> Thank you very much.
>>
>> Regards,
>> Ali
>>
>>
>> On Tue, 10 Apr 2018, 00:14 Tarik S Bel-Bahar, <tarikbelbahar at gmail.com>
>> wrote:
>>
>>> Hello Ali, some thoughts below, best wishes.
>>>
>>>
>>> A simple method is to use an experimental software such as psychtoolbox,
>>> and have it "synch up" with your eeg amplifier, and generate events with
>>> each beep.
>>> Another more principled/accurate way to do this is to have a audio
>>> sensor that picks up the sound and records it's time. Of course the sensor
>>> also needs to be fully aligned with eeg time.
>>>
>>> If you trust your beep event times, and they are aligned with the EEG,
>>> then you might be good as you are, just import your event list into eeglab
>>> (into the eeg file).
>>>
>>> One thing to be careful of here is that your eeg and time signal are
>>> properly aligned (find a way to make sure that 0 time is 0 time in both the
>>> eeg record and the beep events record).
>>> Also watch out for jitter (e.g., inaccurate event timing info, unevenly
>>> spaced beeps), and for "uncoupling or drift" of your eeg time and beep time
>>> over the whole period of the recording.
>>>
>>>
>>>
>>>
>>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20180421/f7191352/attachment.html>