Page 1 of 1

How to calculate frontal asymmetry

Posted: Wed Apr 17, 2019 10:03 pm
by ty22
Hi I want to calculate frontal asymmetry from the AF7 and AF8 electrodes.
I see the in the output csv files there are Alpha_AF7 and Alpha_AF8. I need to compute log(AF8 value/AF7 value) but I see lots of values in the csv file are negative (usually between -1 and 1), which doesn't allow me to compute the log. So is there any conversion I need to do to the values before bringing to the log to make sure they are all positive values? Also, when I import the file to the online graph tool on musemonitor.com, the numbers turn to around 100 (dB?). How is it calculated from the csv data?

Thanks!

Re: How to calculate frontal asymmetry

Posted: Thu Apr 18, 2019 8:31 am
by James
The display values have a linear conversion from ~[-1:+1] to ~[0:100] by doing (x+1)x50.
I'm not sure why you want to log the values; To compare left vs right power, you can just use subtraction and see if the result is positive or negative.

Re: How to calculate frontal asymmetry

Posted: Thu Apr 18, 2019 12:32 pm
by ty22
Thank you for your reply. If I do (x+1)x50, does it make it in db unit? If so, then there is no need to do log, subtraction will do the work.

Re: How to calculate frontal asymmetry

Posted: Thu Apr 18, 2019 12:35 pm
by James
That formula just shifts the range.
There are no official units for what we're measuring here.
Bels are used to label any unit that has been derived from a log function and since the values are a log of Power Spectral Density this applies.

Re: How to calculate frontal asymmetry

Posted: Thu Apr 18, 2019 3:09 pm
by ty22
Thank you very much!