#muse headset

Birds (in your head and on your screen) | a prompt

If you’re in the realm of disposable income that allows you to spend in excess of three hundred dollars on a little piece of technology to assist in your meditation practice, you’d want to hear the birds that sing when you hit the bliss brain wave. The ‘Muse‘ headset application plays the lovely sound of gentle, chirping birds, when your brain remains in a state of calm for long periods of time.…

Getting started with meditation can be tough. You really need a strong motivation, as at the beginning it is quite hard to see what the benefits may be. Sure, you hear and see a lot about how meditation will improve all sorts of aspects of physical and mental health but so does a diet of sugar pills and the placebo effect. For me, it was my curiosity aroused by books such as Robert Wright’s “Buddhism is True”. I had also gone through a Coursera course given by Wright when he was teaching at Princeton. So I was motivated enough to keep at it for long enough to feel it was something worth doing. After quite a long time I began to see that there was indeed something to all the claims. Now, after almost two years of working at it, I came across the Muse, a headset that supposedly could detect the state of your mind from sensors placed on your forehead. As you can see from the enclosed picture it was stated that this would make meditation easy.

 Meditation made easy after I had put in two years of effort, very hard to take seriously. So I searched the web to find reviews and comments on the device, which it turns out had been around for several years. Among other things I found favorable comments in an essay from Consumer’s Union magazine, I source which I had a lot of trust for. Given all the effort I had put into establishing a meditation practice it seemed reasonable to give the device a try. So I put down the money and within several days I received the headset. 

To use the device you need to download an app, which in my case was found on the ios app store. The app contained detailed instructions on how to set up the headset as well as a 10 lesson beginners course on getting started with meditation. Even though I did not feel any need for the course I decided to go through it if for no other reason but to get used to the device. So I put the headset on as instructed after pushing the start button.

Several of the reviews said it was tricky to get the blue tooth connection with the device working. They also complained about getting all the probe connections on the headset working properly. So I was a bit surprised to have no problem with this whatever. So I was connected and a pleasant and very calm voice instructed me to take a deep breath and close my eyes so the session could be calibrated. This took about 30 seconds. I was then given some elementary instructions about following my breath as a meditation object. The meditation that followed lasted about 3 minutes. During this, I could hear sounds of sea surf that varied from very peaceful to quite violent. These sounds were supposed to give me feedback about when my mind was calm and focused and when it was overly active with inattention. In addition, when I was sufficiently calm I was treated to the sounds of singing birds. All this was a little too much for my taste, being used to relatively silent meditation. However, I can see how this could be helpful to a newcomer, though I am somewhat doubtful if it would build up good meditation habits. In any case, I went through all 10 lessons. All in all, they were quite standard but far from what a new meditator would need for a reasonable practice. What was most interesting is that the evaluations correlated very well with my own sense of when my attention was focused and I experienced a sense of calmness. Normal EEG apparatus uses many more sensors and a skull cap that covers the entire head. It seems that the Muse designers had indeed produced a much sparser and of course cheaper device that at least gave seemingly accurate results for meditation. According to their publicity, the device was being used by universities engaged in psychological research. I did not check on this but did find the claim believable. The app provided did not give raw data, ie voltages taken by the sensors, though they did label the graphs of mental activity as raw. Older reviews were not happy about this, though the company did provide an SDK for developers. In searching the app store I did find that they have now provided a new app that gives raw data and does offer various analysis to produce readouts of typical brain waves. In addition, at least one developer produced a rather pricey app, which I suspect is now superseded by the new Muse app, which is free.

 Thus I was convinced that it indeed does something useful. After playing around I found it was quite easy to set the time for a session and to control sound levels. Thus I eliminated the bird sounds and set the feedback at a low level. With these settings, I have found the Muse to be a useful addition to my practice. It is encouraging to get a verification of how well one’s attention is focused and to be able to see the results of a particular session. While my motivation was quite strong,  I find that it provides an interesting challenge to improve. Still, I am glad that I did not begin with it as I am not convinced that dependence on auditory feedback is useful. This, of course, depends on the individual and for someone who finds it difficult to get started the Muse might be a great solution. 

To sum up I include the results of a recent session with the headset. 

I have now been practicing meditation for almost two years. During that time I have gone through two online courses given by The Teaching Company (now using the name The Great Courses) and read five or six books. At times I have practiced for up to five hours in a day, though my daily practice averages out to about an hour. I have not attended any retreats or live courses. Thus anything I have to say should be taken with the view that I’m both new at the practice and have not followed the formal approach taken by serious meditators. I consider this both an advantage and a disadvantage. The disadvantage is that I have not learned in detail the techniques taught by the professionals or have been able to discuss my experiences on a one to one basis with a teacher. The advantages are that I have not been inculcated with the expectations and rules that have been part of the subject for thousands of years. If we look at these we see that this includes numerous religious philosophies and ideas that arose in ages devoid of the scientific outlook that we now profit by. Also, as my profession has been as a researcher and teacher I have spent a large part of my life learning new things on my own. Many of these have been far from my basic training, eg knowledge of biology used for work in the area of biomechanics. 

The classic meditation practice is usually carried out on a cushion ( zafu), or a kneeling bench. The practitioner sits in an erect posture with a straight spine and ideally in a crossed leg lotus posture. In ancient times and in the places where meditation started this made sense in this is the way most people seated themselves. The advantages of this are definitely great. It allows good breathing habits, stable seating and a tendency not to fall asleep. Also assuming a standard position it also habituates the mind and enhances ones tendency to return to the proper mental state. Not to be scorned, it also associates one to the condition of other clearly successful meditators. For these reasons in my first attempts at meditation, I set out to emulate this classic approach. Unfortunately, possibly due to age and physical condition I rapidly found myself having considerable back pain. The meditation instructions I gleaned from courses and books did cover this situation and suggested the use of a straight back chair in which one sat forward so as not to use the backrest. Adapting to this stance was little help as I still experienced back pain. To do something about this I started using the McKinsey exercise program for back pain. This had helped me in the past and was popular with a number of people I know. Part of the program involves the use of a cylindrically shaped cushion which one places in the lumbar region of the back. Following this, I rapidly recovered and from that time on I practice not only in a straight back chair but on a couch or other convenient place to sit. I have to admit that occasionally I have fallen asleep, but in general, I have not found it to interfere with my meditation. All in all, I think it wise to adopt a position that is comfortable or to not lay victim to the need to follow any particular position just because it has been practiced for thousands of years. Above all what you want is stability, attentiveness and reasonable comfort that will not divert you from the task of meditation. Here is one of the reasons that I feel it has been better for me not to attend a formal live course in which I would be exposed to most of my classmates seated in a manner would not be possible for me. This would impart a psychological feeling that I would not be able to properly practice meditation and be at a disadvantage in learning a true practice. After two years I can now follow my breath with very little diversion in just about any posture, including laying down in bed. Robert Wright, the author of “Why Buddhism is True” claims that unless he attends a retreat he is unable to accomplish this.

Speaking of Wright’s book, his claims about Buddhism come with the caveat that he does not mean that he believes in many of the creed’s stipulations. Thus he rejects reincarnation and the concepts associated with this, such as that awakening will remove one from the wheel of continual and perhaps very unpleasant existence. Paraphrasing Woody Allen who said that by immortality he does not wish that his work lives on, but rather that he himself lives on. The goal of Buddhism appears to be to end ones continual reincarnation, while I prefer the probably unreachable goal of continuing to exist. Not being a Buddhist scholar I can not speak with the surety of its doctrines. However, I do know that it has been reported that only a small fraction of Buddhists actually meditate and most worship the Buddha as a deity. At the moment I consider Wright’s book to be one of the best background guides for those interested in meditation.

An excellent guide to the actual practice is “The Mind Illuminated” by Culadasa John Yates, Matthew Immergut, and Jeremy Graves. This provides a ten step program to the meditator. The books do adhere to much of the classical practice of Buddhism and follows one of the renowned teachings of the Buddha. However, in doing this it transfers these teachings into a modern context and uses everyday language. It does contain an appendix with translations of classical Buddhist texts, which as of now I have not read. The ten step program contains many practical details on such things as how to follow the breath and how to achieve mindfulness. I believe the novice meditator can greatly profit by reading both of these books. 

Muse headsets all lined up, ready for action - photo by William Wnekowicz

In February there was a big party/exhibition at the Exploratorium in California with people walking around controlling, or trying to control various objects like robotic arms, as well as their own mental states. The team who put together the most visible mental state exhibit was Marion Le Borgne, James Bobowski, David Silva, and William Wnekowicz

Muse, a dry EEG headset (as opposed to gel or saline) allows for live time streaming data of brainwaves, and was the headset of choice for the exhibition. It also used just a few processing bands of waves like alpha, beta, theta and gamma.  The way the brainwave data was stored, analysed and visualised was through an open-source project the team named CloudBrain which is generating a lot of buzz in the BCI (Brain Computer Interface) world. For visitors who allowed it, their anonymous brain data was stored in a cloud database using the Cassandra system. 

People’s live-time brain feeds projected on a viewer. Are you relaxed, calm, tense, thinking, or are your eyes closed? You can look up on a screen and tell right away.

Cloudbrain used an analysing module powered by a custom algorithm. The visualization was set up to look like a “radar” chart, and it projected real-time EEG on a monitor. The code that powered it was written in AngularJS or JavaScript.

The tricky part of this exhibit revolved around how to stream everyone’s headset data while they were walking around. Usually BCI consumer grade headsets work with one headset, one person, and one display. The team  chose a publisher/subscriber architecture so all EEG streams (publishers), would route any one visitor’s stream to a specific booth or (subscriber). They did this through a Spark Core and a RFID tagger. They also used an app called Spacebrew. 

Spark Core device on the left and RFID taggers on the right

The Spacebrew app was not meant to be used with so much traffic, meaning so many people. That is because Muse sends out a few thousand messages per second, and the exhibition used 20 Muse headsets, which is thousands of messages per second times twenty.  The team had to rewrite the code to make it work in rabbitMQ, which scales in size much better on the backend. 

They  needed huge server (BrainServer) capacity and a lot of dedicated internet bandwidth. The team also deployed five machines to capture Bluetooth data from the Muse headsets, and forward it to BrainServer. 

Lots of Bluetooth enabled headsets at this opening party

CloudBrain has a demo of, as of this writing, just two models of BCIs - Muse and OpenBCI, and how they parse data live time. 

A screen shot of how the MUSE parses livetime data in the ‘radar’ chart.

The first screen shot shows 25 users, and the second shows “all”, though I am not sure how many “all” is.

Muse “all data, more squishy lines at top which are different channels

The CloudBrain data connects back to a repository on Github of raw code. On the repository they posted the averages of the aggregated data. 

Most people went into Alpha states of relaxation, or closed eyes, hard to say which state won out. That was followed by Beta, or alertness and/or agitated. Coming in third was Theta, or calmness. The MUSE functioned as a biofeedback device. 

This is the best representation of the visitors and the ‘radar’ screens live time. 

The developers think it is very interesting for participants to see live time comparisons of how users brains stack up against others. 

This is the first time that data banks of users brainwaves are being developed for use in the Cloud and proves how the field of neurogaming is right on the cusp of Cloud storage technology. The implications of this are far reaching. Cognitive Technology understands the profound implication this will have on issues of bioprivacy, and recently launched the first workshop for the Center for Responsible Brainwave Technologies.