Psychophysics and EEG Experimentation

This forum deals with BCI2000 configuration issues.
Locked
Sincrono
Posts: 5
Joined: 13 Jan 2012, 08:38

Psychophysics and EEG Experimentation

Post by Sincrono » 14 Mar 2012, 23:03

I'd like to know if its possible to ask the subject for a yes/no answer after each trial, as in typical psychophysical experiments. I want to analize EEG data offline depending on the actual perception of subjects. I understand that your software is able to process EEG signals online and make detections about those signals. Is there a simpler way for the subject to give an explicit answer instead? (for example pressing one of two keys in the keyboard).

Thanks!

mellinger
Posts: 1210
Joined: 12 Feb 2003, 11:06

Re: Psychophysics and EEG Experimentation

Post by mellinger » 15 Mar 2012, 13:09

Hi,

you can record keyboard key presses, joystick position, and mouse position and clicks using BCI2000. See http://www.bci2000.org/wiki/index.php/U ... ging_Input for more information.

Best regards,
Juergen

Sincrono
Posts: 5
Joined: 13 Jan 2012, 08:38

Re: Psychophysics and EEG Experimentation

Post by Sincrono » 15 Mar 2012, 17:03

Thanks!
Now my problem regards .dat importing to eeglab. When I do the .dat import to eeglab, several "warnings" appear, such as "Warning: 'StimulusCodeRes' is not a valid event for EEGLAB. This event is being
discarded.
> In pop_loadBCI2000>compressData at 222
In pop_loadBCI2000 at 156 "

All warnings are of the same type.
Actually I'm learning to use bci2000 and Matlab, and I just wanted to see the states corresponding to keyup and key down (I saw them on the ascii export, but in matlab I can't)

Thanks!!

gschalk
Posts: 615
Joined: 28 Jan 2003, 12:37

Re: Psychophysics and EEG Experimentation

Post by gschalk » 15 Mar 2012, 18:17

While EEGlab can be very valuable to analyze brain signals recorded using BCI2000, you can also load BCI2000 into Matlab directly using the load_bcidat mex function that comes with it.

This function loads all BCI2000 events/states, and does not filter out ones it does not know about.

Gerv

Sincrono
Posts: 5
Joined: 13 Jan 2012, 08:38

Re: Psychophysics and EEG Experimentation

Post by Sincrono » 15 Mar 2012, 18:51

Great... my last question is how can I customize presentations.
I'd like to presentate audio files and give some time for an answer. Then, an image would indicate the next trial. Each audio file would be asociated with correct response. Will I have to write some code, or just adjusting the matrix elements in the aplications configuration will do???

Thanks for your quick responses!!

boulay
Posts: 382
Joined: 25 Dec 2011, 21:14

Re: Psychophysics and EEG Experimentation

Post by boulay » 15 Mar 2012, 22:20

I think this is easy if you want to make every trial have the same duration, i.e. the trial will end after x seconds whether the user has answered 1 time, 0 times, or 20 times in that period. You could then analyze the responses (see Juergen's post above) offline.

If you want to make the trial end immediately after a button press, or not end until a button has been pressed, then you might have to code that yourself. I also have trials with indeterminate lengths and I wrote my application using the PythonApplication module.
http://www.bci2000.org/wiki/index.php/C ... s:BCPy2000

Loading up Python and all the dependencies is a lot of overhead for what you require so you are probably better off writing it in C++. My application is very different thus I am better off using Python, at least at this stage.

Sincrono
Posts: 5
Joined: 13 Jan 2012, 08:38

Re: Psychophysics and EEG Experimentation

Post by Sincrono » 16 Mar 2012, 10:36

Seems it was not the last question.

Imagine I have 10 audio files played at random. Is it possible to see the order of the files played offline? I'd need that in order to do the analysis.

gschalk
Posts: 615
Joined: 28 Jan 2003, 12:37

Re: Psychophysics and EEG Experimentation

Post by gschalk » 16 Mar 2012, 11:49

Hi,

I am not sure that I understand the question.

If you provide a deterministic presentation sequence, you specify that sequence in a parameter and thus know it.

In the case the presentation sequence is random, you do not know this sequence immediately other than looking at the screen. At the same time, all data and important system states (which include which stimulus was presented and when, button presses, etc.), are stored in the data file and available to you offline.

Gerv

Sincrono
Posts: 5
Joined: 13 Jan 2012, 08:38

Re: Psychophysics and EEG Experimentation

Post by Sincrono » 16 Mar 2012, 12:17

Which state lets me know the identity and order of the randomly played files???
Also, What is the ARSignalProcessing.exe that stars in the StimulusPresentation batch file? The only filter I need is a high pass at 0.1 kHz (for the Epoc Emotiv), selected in the filter parameters. Is ARSignalProcessing.exe doing that filter (the high pass) or is doing additional filterings???

gschalk
Posts: 615
Joined: 28 Jan 2003, 12:37

Re: Psychophysics and EEG Experimentation

Post by gschalk » 16 Mar 2012, 12:36

Here is a list of all states. Depending on system configuration, not all of them will be in a particular data file.

http://www.bci2000.org/wiki/index.php/U ... nce:States

The state in question is StimulusCode

Unless you do brain signal processing in real time, which signal processing module (e.g., ARSignalProcessing) you use does not matter.

Gerv

Locked

Who is online

Users browsing this forum: No registered users and 16 guests