EEG 4 channel reading in .csv file



Sure, I can add an “OSC Prefix” setting.

I think rather than have the MuseID hard coded in there, it’d be better to just have it as a text field where you can type what you like, that way existing users won’t have to modify their MuseLab config files.

So, it would just be blank by default then you could enter user1 and then you’d get OSC paths with user1/muse/eeg showing up in MuseLab. How does that sound?


That sound perfect! Thank you very much. Maybe you can make the existing “Filename prefix” be also the “OSC Prefix”. If it is possible… I do not see a situation in which we will need to discriminate both.



Right now the default file name prefix is museMonitor, so that would mean everyones OSC paths would change. I’d prefer to add a separate setting anyway to accommodate all workflows.


Great! Thanks!


Muse Monitor version 2.0.16 has now been submitted to the Apple app store and is pending their approval. It will likely be live in 24-48 hours :slight_smile: The Android version 2.0.9 is also done, but I need to wait for my wife to get home, so I can test it on her Google Pixel, then that’ll go live in ~ 4 hours.


Excellent, @Enigma644 Thank you so much.

Have you thought about creating a mechanism in your app to connect multiple devices only one iphone? Or developing a PC version of your app to allow it… It would be very useful. In my research, for example, I have to connect 5-6 muse simultaneously, and by using the app I need 5-6 iphones.
I know Muse Direct intent to do that. But there are various bugs in the beta.



I could let it connect to multiple devices, but the phone wouldn’t have the processing power to display the data, only keep a black screen and stream. Also the amount of devices it’s possible to sustain a connection to would entirely depend on your phone processing power. If you’re near your limit (which you would have no way of telling), some background task from another app could result in data loss. I want to make sure the experience of using Muse Monitor is flawless and that’s the kind of unknown situation I’d rather avoid!

Rather than expensive iPhones, I would get a bunch of cheap android devices. If you go to Chinese sites like and, it’ll take a while to ship, but you can get a decent Android phone for as low as 50USD, or possibly cheaper if you go on eBay and shop around.


Great idea. Is there any minimum requirement that a Android device should have to run Muse Monitor App?


You need Android version 4.0.0 (API 15), or higher and bluetooth. Bluetooth LE if you want to use the 2016 headband.

The lowest end phone I’ve personally tested on is a Samsung Ace II X, from 2012.

BTW, I just got an email from Apple saying that Muse Monitor 2.0.16 is now approved. So if you go to the app store and hit update, you’ll be able to prefix your OSC streams now :slight_smile:


Perfect! Thanks!


I tried the option muse/elements/blink and /muse/elements/jaw_clench in muse lab and it still doesn’t work. I also tried to look at the txt file but it’s not there. However, if I use the .csv from MuseDirect ios the blink and jaw clench are in the file, if I try it in Muse lab still not working. Any idea?


I don’t get it. What exactly is the auxillary input?? how is USB related to this?


You can use pins on the USB connector as an extra plug in EEG sensor.
See this post for details: Step-by-step Tutorial for Making Muse Auxilliary Channel Electrode


Is anyone selling these? Apparently Myndlift makes them, but only sells to people who sign up for their full package.


Ok, so does that means that out of the five sensors in front portion of the band, the middle one is not active? Because when I was using muse 2, I was getting 5 readings in the muselab software via OSC stream. but in the muse monitor app I was gettting on 4 signals? Can you tell why was there a difference?


No, from the 5 gold pads, only two are giving you sensor readings. Think of the others as the other end of the electrical connection. Like with a battery connected to a light bulb you have a positive and a negative wire but only one light.

The sensors are: Left Ear, Front Left Forehead, Front Right Forehead, Right Ear, Right AUX.

You can turn on AUX in Muse Monitor advanced settings (right at the bottom), but only turn it on if you actually have an Aux cable attached, otherwise you’re just looking at static.


oh! So basically the middle three are reference electrodes. The left most and the right most are the ones which are giving me reading. Have I got it right?

So out of eeg0,eeg1,eeg2,eeg3,eeg4 I have to only consider eeg0,eeg1,eeg2,eeg3 right?
These will be the reading for Left Ear, Front Left Forehead, Front Right Forehead, Right Ear.

I am asking this because in muselab software, via OSC stream I was getting 5 sets of stream (eeg0,eeg1,eeg2,eeg3,eeg4) which out of these should I take to feed in my machine learning model?

Please help me out as I still not clear which of the signals are valid and whether there are 4 or 5 signals.

In CSV.jpg as you can see in the raw signals, I am getting 5 reading including AUX, but I haven’t connected any external input. Which of these should I consider then for feeding into my model?



EEG0 - TP9
EEG1 - AF7
EEG2 - AF8
EEG3 - TP10
EEG4 - Right Aux (even if not connected)




Do you know if you can download the data in any other format except from an excel file?
I am trying to find a way to insert the EEG in an another software in order to visualize them in a creative way.

Thank you!


In Muse Monitor’s settings you can toggle between .CSV and .Muse which is a Google Protocol Buffer file formatted to Interaxon’s spec.