Questions about EEG frontal asymmetry

musemonitor

#1

Hi I’m a computer science student but I want to measure EEG asymmetry in people’s reaction in a VR scene. I’m using Muse 2016 (one of the only devices that fit with a VR headset), and using Muse Monitor for exporting data.

I have several questions:

  1. Which frequency band should I look at? I’ve seen people using alpha, beta, gamma, but still not sure about it.

  2. Is the change in frontal asymmetry time-related? Is it visible on the graph upon the onset of a stimulus? If not, should I compute the mean or median value of the entire duration of the experiment? (currently I visualized beta band but the graph looks very random to me.)

  3. How should I preprocess the data from Muse Monitor (compute baseline, de-noise)? Is there any method or algorithm I can implement by hand? Since the Muse band is not a scientific tool, I can’t find any toolbox (for Matlab, for example) to process the data.

Sorry I’m a Computer Science student and just started learning about EEG so any advice on how to make the most of the data from the Muse device will be great help to me!

Thanks!


#2

Hey there,

I’m a Cognitive Science student and I’ve been toying around with EEG a bit, and sounds like I’m in a similar position as you. Really cool that you’re going for EEG + VR, definitely something I want to try go for. I’ll try and answer what I can for you.

The Alpha Band is usually looked at often because it is associated with cognitive load. It’s observed a lot for learning because when cognitive load is too low or too high, there’s no learning taking place, and as such the Alpha Band is considered an important observation. Take this with a slight pinch of salt as I’ve only started looking into it this week from this paper.

As for the data, there’s a MATLAB toolbox called EEGLAB. There are some good tutorials on Youtube on how to use it. Should be able to guide people who aren’t too familiar with EEG either.


#3

Hi, thank you so much for your reply.
Do you have any idea if sudden visual stimuli (for example, showing a terrifying face to the subject) can be reflected on the Alpha band change or frontal lateralization change (Alpha_AF7 subtract Alpha_AF8, for Muse)? It’s very important to us because we are trying to measure users’ reactions to visual stimuli using Muse.
Or is there a better way you can think of to do this?
Thanks!


#4

Hey,

I’m afraid I can’t help you much there as my knowledge is stretched a bit thin on that front. Also as I said in my previous post, take my recommendation for the Alpha Band with a pinch of salt, as I believe it’s primarily concerned with cognitive load and not surprise. Also a problem you might encounter is high Delta activity due to a lot of eye movements. The problem with the AF7 and AF8 electrodes being right above the eyes means you’ll pick up a lot of unwanted electrical activity from eye movements that’s usually reflected in the Delta band, but I think this can be filtered through EEGLAB for MATLAB.

Sorry couldn’t be more helpful than that, but if something comes to mind, I’ll try input when I can!


#5

Thank you so much for your help!