User Tutorial:Obtaining Mu Rhythm Parameters in an Initial Session
Although the basic properties of the Mu rhythm are identical for all humans, its spatial pattern and its exact frequency will differ. Thus, it is necessary to obtain these individual parameters prior to any feedback experiments, in an initial session.
Experimental Design
In the initial session, the subject is instructed to imagine hand and/or feet movements according to a pre-defined pattern. To identify a subject's Mu Rhythm, the amplitude of the EEG signal is then taken for different frequency bands, and correlated with the instructions. Across frequencies, this results in a line spectrum; across electrode locations, the result is a spatial distribution of activity.
Performing the Initial Session
- Start BCI2000 by double-clicking batch/StimulusPresentation_<YourAmplifier>.bat. For repeated use, placing a link to this file on the desktop might be a good idea.
- In the operator module, click "Config".
- Click "Load Parameters" to load the parameter file at parms/mu_tutorial/InitialMuSession.prm.
- Load additional parameters for your amplifier by clicking "Load Parameters" again and choosing the parameter file from parms/fragments/amplifiers that corresponds to your amplifier. If you're unable to find a parameter file that corresponds to your amplifier and your amplifier is listed in the contributions section, load the SignalGenerator.prm parameter file (also located in parms/fragments/amplifiers) and modify parameters as necessary. Please see User Reference:Parameters for a listing of currently supported parameters.
- When you have a separate monitor for experimenter and subject, load the parameter file at parms/fragments/stdlib/DualMonitor.prm.
- In your system's display properties configuration, make sure that the subject's monitor is configured to be located to the right of the main monitor.
- Make sure the WindowLeft parameter matches the main monitor's actual pixel width.
- You may need to adapt remaining window position parameters as well; click "Set Config" to try the effect of your changes.
- Under the "Storage" tab, enter the subject's initials into the "SubjectName" parameter.
- Switch to the "Source" tab; For the "ChannelNames" parameter enter electrode locations corresponding to amplifier channels as a white-space separated list (e.g., Fz CPz Cz CP3 ...).
- Click "Set Config".
- Instruct the subject as described in the next section.
- Click "Start".
- The recording will finish automatically. If you want to acquire more data, just click "Resume" as often as you like.
- Finally, click "Quit" to leave BCI2000.
Subject instruction
During the initial session, the subject's screen will either be blank, or display an arrow pointing up, down, left or right.
- When a left or right arrow is displayed, imagine movement of the respective hand. The imagined movement should be continuous opening and closing of the hand, similar to rotating a tennis ball in one's hand.
- When an up arrow is displayed, imagine simultaneous movement of both hands. This should be the same kind of movement as described for a single hand.
- When a down arrow is displayed, imagine movement of both feet. The movement should be similar to the one described for hands, i.e. imagine opening and closing your feet as if you could use them to grip an object.
- When you see a blank screen, please relax and stop any movement imagery.
Analyzing the Initial Session
In the previous section, we completed an initial BCI session using BCI2000 that culminated in an EEG data file. Now, we'll want to analyze the recorded data to determine what features will best facilitate on-line control. To begin with we'll set aside the data recorded in the previous sections and analyze some sample data that is included with BCI2000:
- will guide you through an offline analysis using some sample EEG data included with BCI2000.
- will compare the results of the previous step with other similar data.
As we know, EEG data is not the only type of data that is of interest in BCI. Following are some additional tutorials for working with different types of datasets:
- will guide you through an offline analysis using some sample ECoG data included with BCI2000
- will guide you through an offline analysis using some sample MEG data included with BCI2000
Next Step
Configuring Online Feedback shows you how to configure the BCI system using the parameters obtained in the analysis.