[Eeglablist] different ERSP values of the same dataset used in two studies

[M Abul Hassan]

Hi everybody,
I have a problem with different readings of the ERSP of the same data in two studies.
The first study has three groups (G1, G2 and G3) and the second study has only two groups G1 and G2, which are identical for both studies. I created the STUDY with three groups by adding G3 to STUDY with G1 and G2. 

When I plot ERSP/PSD in the STUDY of ALL 3 groups, using EEGLab GUI, I get on several electrode locations very hight values for ERD over all frequencies (like noise). When I look at the same data in a s STUDY of 2 group, plots look fine. Numerical values also look different for the same dataset in two studies. 
The numerical values are obtained using command:

[STUDY erspdata ersptime erspfreqs] = std_erspplot(STUDY, ALLEEG, 'channels', {'Pz'})
[STUDY pecdata specfreqs] = std_specplot(STUDY, ALLEEG, 'channels', {'Pz'})

The scalp map is also looks different for several electrode in both studies. The numerical values are shown below:
PSD:
For G1:
Ch1: 49.1689 (study1), 51.1606(study2)
Ch2: 29.1974 (study1), 51.2525 (study2)
Ch3: 35.4119 (study1), 50.7535 (study2)
ERSP

For G1:
Ch1: -0.1233   (study1), 0.4490 (study2)
Ch2: -10.7306 (study1), 0.4834 (study2)

Have anybody experienced similar and what it the solution?
Many thanks
Muhammad Abul Hasan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20120519/22205385/attachment.html>