statistic

This forum deals with BCI2000 configuration issues.
Locked
simona
Posts: 3
Joined: 14 Apr 2005, 12:16

statistic

Post by simona » 26 Jul 2005, 07:49

Hello,

We recently installed the latest BCI2000 upgrade. Our best subjects have very high performance with 1D task. Now, before starting 2D experimental tasks we would like to train our best subjects to perform the horizontal movement. So, we would like they perform their training as they do for vertical movement. It means that the task should be with the cursor starting on the bottom of the screen and moving towards the target on the top. I want the cursor always go across the screen by itself, so just I give it a constant bias/offset in YMean/Offset and set YTrendControl to 0, then I set Xmean and Xgain to have the desiderate speed and XTrendControl to 1. It seems to work well, but the problem is that I can’t make the BCI2000 save statistic information about the performance! The only things I changed are the target’s position and the adaptation!

I appreciate your help!

Thanks,
Simona

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

Re: statistics questions ...

Post by gschalk » 03 Aug 2005, 10:03

Simona,

Do I understand correctly that you do in fact get the vertical movement working just fine, but some information is not saved? Exactly what information are you looking for?

Gerv

simona
Posts: 3
Joined: 14 Apr 2005, 12:16

Post by simona » 31 Aug 2005, 11:14

Dear Gerwin,
I didn't answer you because I was on holiday. I come back to work just today.
The information I'm looking for is about the performances. When the subject performs the vertical movement, in the log file the system saves as result the number of hit/missed targets (performances) but when he performs the horizontal movement the log file is empty!
Do you Know what I miss?

Thanks for assistance!
Simona

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

Performance ...

Post by gschalk » 31 Aug 2005, 11:28

Simona,

You might have discovered a bug here. You can always determine the correct numbers by evaluating the state information in the data file. You can use this as a work-around. Specifically, you could convert data in Matlab, and then write yourself a little Matlab program to get the statistics that you need.

Gerv

Locked

Who is online

Users browsing this forum: No registered users and 3 guests