[Eeglablist] Calculating ERSP from Dattimef data

[Pia Ludwig]

Dear all!

I am sorry to bother you with this, but I tried to understand the data
presented in the .dattimef files created through a study.
When looking at the channel data, I was wondering how these relate to the
the ERSP figures plotted by the study function?

I naively thought it would be just an image of
10*log10(mean(abs(ChanData).^2,3)) , but if I do this, I do not get
anything close to the plot presented in EEGlab.

I suspect two reasons for this:
(1) Do I have to apply a baseline correction on the data from the dattimef,
or is it already done? (I applied when precomputing the measures). If I
still need to do this, is this done for each trial and as a substraction or
division?

(2) Do I apply the 10*log10 for every trial, and average across them them,
or do I do I average across trials and thentake the log on it?

Best
Pia