EEG data power value rescaling

Discussions
1

Hi everyone!

First I want to thank Brainstorm for its amazing functions! Here’s a problem I have and hoping to get some help.

I run Frequency>Power spectrum density(Welch) on some EEG data. Then I exported the psd file to Matlab for power values(.TF file). However, the data showed in (.TF file) is very small.

In this case, should I rescale the power value? (I rescaled (.TF file) by simply using " *10^12 " but it’s still very small). I also tried Run>Pre-process>Absolute values and the result seems to be the same.

Thank you in advance for any assistance!
Amy

2

All the values saved in the Brainstorm .mat files are in international units, for EEG your files it would be Volts.
If you display the figure with the option “power” selected in the Display, you will get to see the actual units of the power values: signal_units^2/Hz, so V^2/Hz for EEG.
Your EEG values are in the range of 10^-6 V, so the power values are going to be in the range of 10^-12.
What you observe is expected.