Muse suggestions


#1

Hi Muse Team,

I got my muse headband last week and I’m really liking it so far, so I thought I would share some feedback on my experience so far, and offer some suggestions to make this a better experience. I’m sure you guys are working on some (or all) of these things, but I thought you would like the feedback anyway:

  1. The users could really use their own forum, or sub-forum located at the front of the website so its easily accessible. We could use a place to talk about how muse has worked for us so far, the different experiences we’ve had, the dos and dont’s, etc. Even a subreddit would work.

  2. The calibration needs more clarity in its purpose and execution.

  • I believe that too much of the session data relies on the calibration, and it is not very well explained how the calibration procedure works. I had my knee resting on the table for a calibration event, then took it off for the start of the session, and received an 80% calm score with 60 bonus birds. I noticed that the data was off in the middle of the session and tried to spike it up with some random thoughts with no success. The calibration corrupted the sessions data.

  • If I do multiple sessions one after the other, I see a slight decrease in my overall calm level by 5-10%. I don’t believe the data is accurate. I feel that the only thing that is happening is that I am calmer on my calibrations then I was in the previous session, and so that sets a lower baseline for me to stay at. Its not indicative of my overall calmness level, just in relation to the calibration. Maybe that’s the way it’s supposed to work, but it seems counter-intuitive to me.

  • When we are thinking about the multiple categories, and we can’t think of anything else, should we fill it in with random thoughts? Is it ok to think of things twice? If we can’t come up with anything but we’re still thinking does that affect our calibration?

  1. How do the wind measurements affect our session? Do stronger winds option mean the headband is more sensitive? Its not exactly clear.

  2. Counting breaths feels counter-intuitive to what we’re doing because the band registers the count as brain activity. I’m not sure if it’s supposed to differentiate between the numbers we’'re counting and other random thoughts, but if it does it doesn’t seem too accurate. I’ve had multiple people try a session, and they all agreed that it was easier to remain calm and get a better score by watching the breaths without counting them. One of my co-workers started panicking in the middle of the session because she counted past 10, then she felt bad for counting past 10, then she panicked about panicking cause she was supposed to remain calm… you get the idea.

  3. We could really use more then 12 minutes per session. I’ve mentioned this before but its really crucial for those of us who meditate daily to get longer sessions in.

  4. I’m not sure if the band can register the different brain waves (alpha beta delta theta and gamma) but it would be amazing to see that information on our readings. I know this is a more commercial product and you don’t want to confuse the general public with those graphs, but an advanced data option (disabled by default) would be amazing.

  5. I transferred my headband from one phone to another and lost at least one session in the process, maybe more. No idea what happened.

  6. A way to delete sessions would be nice. The session where I scored 80% is throwing off my data/calm bird count.

Thanks for listening, and thanks for the amazing product.


#2

Hi Andre,

You’re correct that the session data relies on the calibration. It’s not sensitive to things like a knee resting on a table, but it will be sensitive to the state of your brain at the time of your session, and to the ‘quality’ of your calibration.

It’s definitely ok to think of things twice, and yes, it’s also ok if you can’t think of anything and are still thinking. The exercise we have there is designed to provoke an active mind. So as long as you are actively thinking about things it should provide good quality calibration data.

Things that can negatively affect calibration are muscle movements and external distractions. If you find yourself moving around a lot during a calibration, it may be best to end the session and try again for the most accurate data. We are also working on improving our ‘noise alerts’ which will automatically interrupt a calibration with poor signal quality. When we launched we set the threshold really high, but as we get more data in through the research program we’re able to set it more accurately.

Our R&D team has also found an improvement to the algorithm that can alleviate the dependence on the calibration. It should greatly reduce the variance between sessions. The next step is to implement it in the app, and that’s going to take a couple weeks.

As for some of the other points:

3 - The wind intensity affects the way that the sounds respond to your brainstate. It won’t affect the scoring of the session.

4 - It’s possible to use Muse without counting your breaths, and simply placing attention on your breath. We found that for novice users, having something to “do” during the session helped them feel more confident, so we chose to give the breath-counting instruction. We are looking at adding more exercises, and attention on the breath without counting is definitely on the shortlist.

5 - We are just working on building a graph that can accurately represent longer sessions (we need to add zoom-in / zoom-out functionality.) Once we’ve done that, you’ll see some longer session options!

6 - You can currently look at the “raw” brainwave data (alpha/beta etc.) with MuseLab which is included as part of the SDK. It’s not on the roadmap Calm right now, but I’m going to keep looking for a way to provide that information, either through a second app, or like you suggested in some kind of advanced data view.

7 - If you still have the old phone, open up the app while connected to the internet and leave it open for five minutes or so. It can take a while for data to sync with the cloud and the missing session might be “trapped” on the phone. We’ve speced out an indicator that will give feedback as to when data is syncing with the cloud, and it will be implemented in an update soon. This should make the syncing process a lot clearer.

8 - This is something we’ve heard before as well – deleting causes some issues for us on the back-end, but we are looking for ways to allow for it.

Sorry for the late reply – things are (as usual) pretty busy around here!


#3

@trevorcoleman
Thanks for this information. It definetly helps that you can just focus on your breath I didnt know that. I have one question, will the source code of Calm or some basic version of Calm be available? I used to develop but I cannot write a whole iOS application myself, so I was wondering if there would be a basic iOS app that developers then can further make their own and share.


#4

Hi, I thought I’d contribute my feedback to this thread too.

  1. as per 5. An ‘infinite’ setting would be a great feature, but it would need to take into account blinking and other head movement. I’m sure its possible to compensate for blinks after a bit of calibration (in muse-lab its really shows).
  2. an ios SDK with HealthKit support (I made a comment elsewhere on this) would be a desired software design (and timeframe)
  3. The sdk is based on several libraries and languages which makes it a little ugly. Having used the Leap Motion SDK,there seems to be a similarity on the interpretation of raw data…it would be great if there was a port to Node.Js for some of the functionality, since visualisations would be very flexible. (Its quite fragmented at the moment)
  4. I originally pre-ordered my Muse last year, when i was considering building a Controlled Remote Viewing app, and wanted bio-feedback on Theta patterns (Which seem to be a big factor in remote viewing and transcendental meditation);
  5. so i’m interested in being able to use Muse throughout the day (when i’m working), and measure my levels of focus, efficiency, productivity etc. hence functionality BEYOND just calming my brain is what i’m hoping additional Apps/SDK can bring
  6. i had another abstract app visualisation idea (rather than wind/birds) for visual feedback of multiple rings than float around and become aligned in calmer states (sort-of like the ios built-in sprit level app)
  7. the app could work in background mode perhaps and just collect data
  8. I am thinking how raw data could be collect and applied to machine learning systems and what could be produced from that…

as for the hardware, I wear glasses, and sometimes the headband has problems getting good contact around the ear

  1. i do like the gamification aspect of the app… i think its vitally important to start slowly, but it would be great if there was more sophistication in some of the exercise: ie you can go for longer sessions until your good with shorter ones.

  2. being able to train for gestures/thoughts… i’m not sure if this is possible… but would be awesome.

Adam


#5

Right now our platform team is fully committed supporting the application, but as that load lightens the SDK is the next thing we’ll have them working on.

  1. The sdk is based on several libraries and languages which makes it a little ugly. Having used the Leap Motion SDK,there seems to be a similarity on the interpretation of raw data…it would be great if there was a port to Node.Js for some of the functionality, since visualisations would be very flexible. (Its quite fragmented at the moment)

I’ll bring that up to the dev team.

  1. I originally pre-ordered my Muse last year, when i was considering building a Controlled Remote Viewing app, and wanted bio-feedback on Theta patterns (Which seem to be a big factor in remote viewing and transcendental meditation);
  1. so i’m interested in being able to use Muse throughout the day (when i’m working), and measure my levels of focus, efficiency, productivity etc. hence functionality BEYOND just calming my brain is what i’m hoping additional Apps/SDK can bring

An app that lets people track their brainwaves through the day is something a lot of people have asked for. We’re looking at ways that we can make that happen, although I don’t have any particular timeframe for you.

  1. i had another abstract app visualisation idea (rather than wind/birds) for visual feedback of multiple rings than float around and become aligned in calmer states (sort-of like the ios built-in sprit level app)

That sounds pretty cool! :smiley:

  1. the app could work in background mode perhaps and just collect data

Interesting – definitely a good idea for the data-recording app.

  1. I am thinking how raw data could be collect and applied to machine learning systems and what could be produced from that…

Actually, machine learning is a big part of what we do here, and it’s something we used a lot in developing the alogorithm that powers Calm.

as for the hardware, I wear glasses, and sometimes the headband has problems getting good contact around the ear

All three founders wear glasses, so it’s something we thought a lot about! It should be possible to get good signal with glasses on, although it does take a bit more fiddling.

  1. i do like the gamification aspect of the app… i think its vitally important to start slowly, but it would be great if there was more sophistication in some of the exercise: ie you can go for longer sessions until your good with shorter ones.

We’re currently re-examining the motivation and gamification elements of the app, so look for some changes coming in updates soon!

  1. being able to train for gestures/thoughts… i’m not sure if this is possible… but would be awesome.

Definitely a someday feature! There are some technologies out there that can detect left/right, but we’re still a long ways off from the type fo control applications you’re imagining. At Interaxon, we believe that there’s a lot more value in learning and discovering things about yourself using this technology, and that’s where we’re focussing our efforts over the coming years.


#6

thanks for the response!
and one other suggestion would be to move to github for the code, wiki, issues management,branches etc. — much better than the google site you’re hosting on


#7

I strongly support the need for a user forum!


#8

We’re in the middle of rebuilding our website – look for improvements all over our web presence soon!


#9

You’re posting in it!


#10

hello everyone,
Since am very new to this muse device, I would like to know what all are the API’s that I can use to get the functionalities of the device?
I would like to buy the device for the sake of my Final year Project. So am in need of the specification details of the device. please reply for the same.
Thank you.


#11

You can find information about the SDK in this post:
http://forum.choosemuse.com/forum/developer-forum/1070-setting-up-the-sdk


#12

The calibration is frustrating for me at occasionally. Sometimes it makes me re-try up to 6 times. Eaxh time I need to listen to the voice prompt telling me something that I know really well. “Think of as many things possible…”


#13

Calm vs sleepy, has the difference been measured/ compared? Question is generic, but how has the target state been defined? Machine learning is one of the key development tool, have “real” meditation masters (e.g., Tibetan monks, Japanese zen masters) been tested?
Is anything been done on identifying quality of the meditation, beyond, calm and active?