Question about using sound to influence brain state


What I want to do is to use the common power bands like (Alpha/Beta/Theta/Delta) to control through OSC the selection of my pre-composed soundscapes and frequencies that in turn influence the brain state. For this I have a few questions: - Can the libmuse interface also target specific frequencies, next to the common power bands? - When will libmuse be available? - Will libmuse have a visual interface? or is it just code? Or do I have to link it to muse-lab? - Do you have any literature available (or an in house expert) on awareness states like enhanced focus, better sleep, anxiety relief, relaxation and their respective EM brain frequency bands? Or maybe you can point me in the right direction? Thanks so much in advance and keep up the great work!


MuseSDKv2 includes the powerbands as part of Muse-IO (the desktop driver), it outputs OSC. You can see all the data you get here:


Thanks for pointing me in the right direction!

(In 3.4 they are now called elements). A question though: I’m wondering how I must interpret the numbers I’m seeing. they seem to stay in between the 0.10 - 0.70 range in MuseLab visualiser.

And so for example in the Alpha range, do I need to interpret the numbers as 0.01 weakest alpha activity and 1.0 strongest Alpha activity?

Thanks in advance!


I was wondering these exact questions (post #3) in another thread. So far i only have my assumptions and they’re the same as yours.


Today I showed the Muse to a neuroscientist who has used EEG and fMRI a lot professionally and he was not very sure if it were real brainwaves that this device is picking up… So maybe we can have some more clarification what exactly this device is picking up and how that is translated into these numbers, so that we know how we can interpret them?

For example, when you close your eyes, usually you see an Alpha wave spike on EEG. So if I would want to test that in MuseLab, I would set up the four channels for the Alpha band. Which would give me some numbers like for instance all the mean numbers are in the 0.20 range. Then when I close my eyes I should see a direct increase in those numbers right?


push! i think the best alpha-blocking effect you will get at the back of the head, but it should also work with the muse.


Yes it works, when I close my eyes I get an increase in Alpha waves!

But also a lot of Delta waves, when I’m fully alert. Does anybody have some good literature on the explanations of the different bands? (Alpha Beta, Delta, Theta, Gamma?) Explanation that goes beyond the standard stuff that you see on Wikipedia.



i experienced a short increase of alpha waves when i close my eyes, but about 2-3 seconds later the alpha band drops again and fuses with the general “noise”. i think this peak of alpha is just an artifact?



forgive my poor english…

Hello,I need a month to receive my muse,[SIZE=14px]I am a[/SIZE][SIZE=14px] meditation teacher,[/SIZE][SIZE=14px] I hope I can[/SIZE][SIZE=14px] discuss with you[/SIZE][SIZE=14px],[/SIZE][SIZE=14px] would be very interesting[/SIZE]


I am software developer who worked also with EEG research. I would like to give some explanation. What MUSE delivers is most probably the calculated power in alpha, beta range etc. So this signal is not really EEG but the coefficients related to EEG spectrum in time.


So my Delta bandpower is so high because low frequency waves have more electrical power?

So I can conclude that when I compare the 4 different bands, that if there is a higher number in Delta, it doesn’t necceserily mean that there are more Delta brainwaves?

If so how would I configure for example Muse lab to visualize it so that I can see the dominant brainwave band?


thedas I don’t think MuseLab is sophisticated enough to do that.


@thedas: To see the dominant brainwave band means to try to take the values of each of the bands and see how they change during different states of mind relative to each other. One way you may be able to accomplish this is by normalizing the values you’re receiving for each band to see what is dominant. For example, if you see that alpha values are typically from 30 to 80 (I can’t remember what kind of numbers muse-io outputs so these numbers are totally made-up), then you can map the 30 to 80 values to values of 0.0 to 1.0. Similarly, if you see that beta values are typically seen from 20 to 50, then you can map those values to values from 0.0 to 1.0. Once both alpha and beta are mapped to values from 0 to 1 (normalized) then you can compare them.

You can do this with muselab though it is currently a bit convoluted. If you have alpha and beta values streaming into muselab you can create a normalizer to do what I described in the first paragraph. The normalizer tool is found in the DSP Settings. You’ll have to create DSP signal copies of the alpha and beta signals obtained from muse-io. Then you can create a “Normalizer” DSP function that can map values to a specified range using the last number of specified data points. Once you’ve completed creating the normalizer, you can then add it to the DSP signals that you created earlier and then visualize those signals.


Hi RickyD, thanks so much for the explanation!

I understand it until the point of DSP settings.

The muse-io frequency bandpower (elements) output are numbers between 0,00 - 1,00.

When I add the normalizer on the 4 electrodes per frequency band, I don’t exactly know what to input here:

Say my Alpha waves are in the range of 0,10 - 0,50, what do I type in the normalizer menu:

Number of values in nor…:
Norm max:
Norm min:


I would suggest you try using 100 for the number of values and then put the norm min as 0 and norm max as 1