[Eeglablist] Differences between pre and post downsampled

[Steven Pillen]

Hello, EEGLABlist

When downsampling the data in a study, we noticed a strange phenomenon.

Originally, the chart for a specific participant with a specific electrode
in a specific condition looked like this with 2048 Hz:

http://imgur.com/2ZrgfjB


Then, after we downsampled all the data to 512 Hz, it was consistently the
case that the magnitude of the graphs changed, yet the shape remained the
same, as seen below:

http://imgur.com/1y2Hi9S

Notice the Y axis in each chart.  It shifts about 12 decibels downwards.

Checking the .datspec files, we noticed there was a difference of about 12
in the values for those specific indices, as well.

Is this difference in magnitude something we should expect?  What does the
downsampling function cause this kind of change?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20140207/4efe7393/attachment.html>