Problems while calibratin p300 speller

[ikara]

I am currently using BCI2000 (February 2011 Build) since there was a problem with the May 2011 version (check the corresponding post in the "Known Issues" section) Although I followed the P300Speller Tutorial you provide, I had zero accuracy in my calibration tests. I have done everything as proposed, and the only other settings I tweaked were the SourceChDevices (I'm using 8 electrodes instead of 16) and the Notch filter (reduced it from 58-62 to 48-52 since all the power lines i'm using are european.

Stimulus Duration, ISI Min/Max, Number of Sequences, EpochLength and EpochsToAverage were left as they were, after loading the appropriate calibration files (the InitialP3Session.prm and gUSBamp.prm). No matter how many tests I run, trying to spell "THE" (from the proposed "THE QUICK BROWN FOX" string) I got nothing. Also attempts in tweaking the aforementioned values had no result.

I am attaching my .prm file and also the data file acquired with this configuration, hoping that it would help you understand what's going on and why I'm not getting any accurate results.

PS: Tests with the original g.tec software provided some accuracy (copy spelling, without a trained classifier)

Thanks in advance,
Ioannis Karavasilis

[ikara]

Sorry, I was ignorant enough not to load the .prm generater by the classifier, everything is working now.