offline analysis

[oubosarah]

Hi,

I want to do offline analysis using Matlab. After installing the Matlab v7.0 and double click the file of offline analysis.bat file, there are some errors:

Error in ==> gui_mainfcn>local_openfig at 199
gui_hFigure = openfig(name, singleton, 'auto');

Error in ==> gui_mainfcn at 94
gui_hFigure = local_openfig(gui_State.gui_Name, gui_SingletonOpt);

Error in ==> OfflineAnalysis at 38
gui_mainfcn(gui_State, varargin{:})

Could you help me figure out what is the problem? Is this related to the matlab software or something else?


Bo

[josh]

Bo,
This is an error we were getting after making an update to the OA GUI. While the update was functionally insignificant, certain versions of MATLAB produced this error. Please download the most recent copy of OA. I think that your experience will be much improved. Please let me know if that's not the case.

- Josh

[oubosarah]

Hi, Josh

I installed Matlab 7.6 and now it works OK. Thanks for your help!

I want to do Mu task and follow the instruction online. Now I have two main questions. One is the Eloc File. I modify one in the original file and make it as the cap in our own lab(40 channels), is it OK to use? Another problem is at the beginning I instruct the subject to do actions based on the arrow showing and do imagination. I wonder whether this sequence is OK?

[josh]

Bo,

Sorry for the delay. Great to hear that everything is working now. As far as the montage, you should be fine to use whatever you like. There are some changes that you'll need to make when you do the recordings with BCI2000 to account for the smaller number of channels. As far as Offline Analysis, you'll only need to specify the correct montage file. So, if you've modified the 64-channel montage accordingly, you should be okay to use that. Unfortunately I'm having trouble understanding your other question. Could you clarify?

Thanks,
Josh

[oubosarah]

Hi, Josh

Thanks for your reply. Now I have a question that in Elco File like eeg64.loc I only see the channels on the scalp, and no space for the channels around the eye like VEO, HEO. Do I need to put them into the Elco File? When I run the experiment, I just record all the 40 channels including VEO, HEO, is it correct?

[josh]

Hi Bo,
When we designed OA, the intention was to have a piece of software that made very common analyses relatively easy without having to have a lot of background knowledge. So, we haven't actually written it to deal with eye channels. Let me, however, suggest a workaround that I think might be helpful:

1. Execute the elocHelper application either as a deployed component (i.e., executable) or from the MATLAB command line.

2. Click File > Import > "ECoG Electrode Location File". Notice that you want ECoG and not EEG.

3. Open your 40-channel EEG montage.

4. Add markers for the electrodes positioned near the eyes. Ideally, you'll want these far away from the markers that correspond to the scalp electrodes.

5. Click File > Export to... > "Offline Analysis Electrode Location File".

Now, you can run an analysis in Offline Analysis. However, you'll want to choose ECoG as the acquisition type (instead of EEG). You'll probably also want to turn the spatial filtering off. The CAR filter that is typically used will probably not be good since you're including these eye-movement-related electrodes.

Eventually you will probably want to do some kind of specialized filtering based on the information you have from the eye channels. Hopefully this will be a useful starting point though. Hope this helps.

- Josh