The colorful wolf

December 13, 2009

Mind control for the masses: NeuroSky Mindset review

Filed under: Tech — randy @ 16:46
Tags: , ,

Before I start, take a look at this video:

NeuroSky sells a device called the MindSet, which is basically a bluetooth headset with a couple of sensors added that measure your brain activity. They provide several applications, like the NeuroBoy game mentioned in the video and a program called Brainwave Visualizer, which shows your real-time brain activity in several cool charts, showing separate values for each of the important waves (alpha, beta, delta, theta). Besides that they also provide an API so you can program your own software with it. Sounds awesome, doesn’t it? I thought so too a couple of weeks ago, but was then disappointed that they didn’t ship to either Europe or to Japan. That is, until last week. I received it last Friday and have been playing with it a lot, so I thought I’d write down my impressions here.

NeuroSky did a great effort at translating everything to Japanese, much to my dismay. The first thing I did was scrounge their website to find English versions of manuals and software. Having found those I proceeded to install it. (Yes, I am one of those guys who actually reads the manual before trying something). Installation was fairly easy: NeuroSky provides a bluetooth dongle plus driver, which is easily set up. The headset then acts as a regular bluetooth headset, and provides a bluetooth virtual COM port interface for the brrainnnn data. Applications can hook in from there, or they can connect to a local service listening on a port if the provided application to do that is running.

First impressions: works straight out of the box, no trouble installing the software, and then it stops. After installing the brainwave visualizer I got some random responses on the brainwave chart, but I was unable to get any activity at all on the ‘Concentration’ and ‘Meditation’ scales, which are supposed to react by either focusing strongly on something or closing your eyes and relaxing. The manual suggested that the signal might be too weak to be used to control concentration and meditation, so I wiggled the headset around a bit, and indeed I got some response. Eventually though it took me more than 10 minutes to get the headset in a position so that the signal was at high strength (4 or 5 bars). But it worked!

Brainwave Visualizer

Indeed after practicing only a tiny bit for the past three days I am roughly able to control the concentration and meditation meters with my mind, which is an awesome feeling. Unfortunately it only worked for about 30% of the time that I used it, because the signal strength was never strong enough, no matter how I wore the headset. Even after following all the instructions on improving signal quality (clean the sensors, remove hair out of the way, adjust position) the quality of the signal seemed to depend most on luck and the position of the moon. It gets worse though. At times when the signal quality is actually good it’s still impossible to use the headset. Why? Well, the movement caused by the mere blink of an eye creates a spike about ten times greater than any brainwave. Don’t blink. Don’t move. Don’t even look left and right cause that will fuck up your reading.

Despite all this I can still see the brain interface working quite well. The thing that really kills this headset is the bluetooth interface. The Toshiba bluetooth stack provided by NeuroSky is basically shit. It works about 5% of the time on Windows 7. For 5 minutes, and then it bluescreens. I have to say that I’m quite impressed: a simple bluetooth device can crash an entire Windows 7 platform, something I haven’t experienced yet since I installed Win7 several weeks ago.

A final note on the user experience: the audio quality is just fucking terrible! I can’t believe that a bluetooth headset’s audio quality is supposed to be this poor. There is so much static and noise on the line it’s just horrible. The range also seems limited to about 1 meter from wherever you plug in the bluetooth dongle, so don’t think (…) about moving too far away. Also, I’m guessing thanks to the crappy bluetooth driver, the sound and the COM connection terminate randomly after some time. I tested this both on XP and Win7, but I was never able to use the headset longer than 20 minutes until either my PC crashed or the signal disappeared and I had to restart to get it to work again. Well done guys…

I really want to believe in this product. It’s one of the first consumer devices that can interface with the brain, which is absolutely awesome. It’s state-of-the-art technology! It’s supposed to have some quirks in the beginning, right? It’s not supposed to work perfectly right from the start. Well, the range of software available is as expected very narrow. There’s about two or three applications available right now. But perhaps the best part of NeuroSky is that they offer an API so you can create your own programs that use the brain’s information. Brilliant!

They provide you with an API in all the major languages, including Java. The API documentation is excellent, the API functions are easy to understand and the sample code is a great help. One minor negative point was that the Java class used to interface with the headset driver was located in the default package, and apparently (I found this out today) Java classes that are not located in the default package cannot access any class in the default package. In other words, my com.heidesoft.RuleTheWorldByMindControl class couldn’t access the NeuroSky class. I resorted to solving this in a nasty way with reflection, which was good enough for a test, but it would certainly be helpful if NeuroSky spent just a little bit more time testing their Java driver for their next release.

So how did my programming go? Well, while programming I was using the headset to listen to music. After 20 minutes the signal disappeared, I got pissed off and gave up.

So, to summarize, here are the conditions for being able to enjoy this device. One: you are unable to move any muscle in your head. Movement destroys the signal. Two: you are deaf. If you are not deaf then the crappy sound quality will piss you off and create suicidal tendencies. Three: you have an attention span longer than 10 minutes and shorter than 20 minutes. 10 minutes is the time it takes to position the headset properly. 20 minutes is the time you have until the software crashes your pc.

As a programmer, I really, really, believe that brain-computer interface is the only way to communicate with our PC in the future. Forget about that Wii crap or hand gestures. That’s just ineffective compared to directly controlling the PC with your brain. I wasn’t expecting the first available consumer product for brain-computer interfacing to offer something anywhere near as awesome as this. What I was hoping for was a way to take a look at the brainwaves inside my head and perhaps write a program that can analyze them. The NeuroSky is not the device I was looking for. I wouldn’t recommend this device to anyone, not even geeks like myself. I’m really sorry to say this, because NeuroSky seems like a nice company, and they seem to have the right mindset. I hope that they can improve their product in the future. Until then, there’s also Emotiv, which provides an alternative but similar approach. It’s companies like these that push technology forward, even if they don’t succeed all the time.

The headset’s lying on my desk right next to me. Whenever I see it I feel the urge to play with it. No matter how sucky, there is a strong technological attraction. Despite all the crappiness I still want to play with it! I still feel the urge to write a program for it. Maybe I’ll try again next week.

2010 update: read the continuation here.

Advertisements

11 Comments »

  1. thx, i was just about to buy this mindset for the same reasons. Now i changed my mind, I’ll rather wait for next improved edition or buy something else

    Comment by Iggy — January 13, 2010 @ 23:48 | Reply

  2. I was also considering buying one… I’d use it with a Mac, which might crash a bit less often… but also provide fewer apps, which doesn’t matter so much since I was gonna roll my own. Because of the openness (“right mindset” as you said) of the product/company I found it more appealing than Emotiv’s EPOC. I think I’ll just check out Emotiv’s lite SDK (that’s free to download but not connectable to any hardware) to see if I find their SDK appealing. If I do I might actually buy it and start booting my Mac into Windows more often. 🙂 Their SDK+developer headset is already available for worldwide delivery, while as the consumer version is only available to US destinations (in a couple of months this’ll probably change). (Another thing to play with could be OCZ’s NIA, but that one doesn’t have any official SDK yet and only reads alpha and beta waves + some facial expressions.) So Emotiv’s EPOC pretty much does all, while as NeuroSky’s MindSet and OCZ’s NIA pretty much different things. But here’s something cool that someone did with the OCZ NIA at least: http://www.ngen.us/?tag=nia (And it’s a cheap device.)

    Comment by Simon — January 18, 2010 @ 6:49 | Reply

  3. Hmmm… I have a hard time finding anyone else saying it works very poorly, but I’m not finding anyone stating that it works very well either. Do you know any other reviews, where the product was actually tried out? (Most just reference what they’ve seen or heard.) I’m just thinking that you might have some faulty hardware somewhere, or something.

    Comment by Simon — January 18, 2010 @ 8:12 | Reply

  4. It could be faulty hardware, but I am getting a response. I’d pursue it further, but I won’t have time in the near future. If you really want to get one soon I think the alternatives you mentioned are also interesting. As far as the SDK goes the NeuroSky is not bad though.

    Comment by rheide — January 18, 2010 @ 21:22 | Reply

  5. If you ever do write a program and the thing ever does work, how about doing a meditation program? I would really like to know what is going on in my brain when I try to meditate. They say deep meditation really fires up delta waves. And also that deep sleep also produces delta waves. A lot of folks don’t get enough sleep. Being able to have effective meditation with delta waves could be a way to make up for missed sleep. And a guided meditation program with this headset might make it easier to do. It the bluetooth is such an issue maybe they should make a wired headset. And it seems like they should have more than one sensor on the head to make it work.

    Comment by andy — January 23, 2010 @ 11:46 | Reply

  6. Thanks for the review…it was comical but informative….

    Comment by Me — February 5, 2010 @ 7:54 | Reply

    • Hi

      Does anybody know this mindset + their applications can improve brain power?

      Thanks

      Comment by sivgan@hotmail.com — February 14, 2010 @ 19:14 | Reply

  7. Hi

    Does anybody know this mindset + their applications can improve brain power?

    I have ordered one just now

    Thanks

    Comment by sivgan@hotmail.com — February 14, 2010 @ 19:16 | Reply

  8. bluetooth headsets are great because they are wireless and your movement is not limited by wires :-“

    Comment by Cordless Screwdriver · — November 12, 2010 @ 22:07 | Reply

  9. Hi. Based on reviews and pros/conc of all the discussed different solutions, I invested in this NeuroSky MindSet and use it with Win7. I also have had some bad experience with the connection, but it seems like it was always due to the Bluetooth setup/paring. When it is properly connected the wawes are transmitted very well after approx half a minute, and the sensors are not as depending on contact etc as I had expected with only one main forehead sensor and 3 ear sensors.

    The connection problem
    Please note that I use the standard Laptop Bluetooth (BT) connection, and not the dongle. I fear that two BT conenctions from the same PC with Win7 can only make my issues worse. What puzzled me first a few times was, that even the Visualizer informs a connection but no wave signal, it is not the sensor signals which is blocked by missing brainwave contact, but actually the bluetooth data which is not transferred. Because, when checking the bluetooth connection within the properties for the specific device in Win7, it informs the date and time for last connection, and NOT “Currently connected” (which it does when it works). In my case I expected the reason for the inproper setup was that I had also paired the MindSet with my mobile phone at some time, and maybe the pairing with the PC was damaged, but maybe this is not the full story – I don’t know. But after removal of the bluetooth MindSet device (setup) in the Win7, rebooting of the PC, and a new pairing of MindSet and laptop, it seems to connect just fine every time.

    But on top of this, when starting up data transfer it seems like it often requires more than 10 seconds to settle and show that there are data available, so if you are impatient and are moving the sensor it seems like it starts all over this settling, and the experience is that it “never connects”.

    I have subjective experience that the settling is improved if I moist the sensors a little. Don’t tell my son at 12 year, that I use spit on a finger, as he also uses the headset and enjoy playing and experimenting with it ;-). So if I moist the sensors, place the forehead sensor relatively close to my eyebrown, and be a little patient with the data settling, it seems to be very consistent and stable with Win7.

    So the first hurdle is to get the BT paired well which is possible with direct bluetooth in Win7 from a standard laptop if not trying to use the headset with other devices. The sound is so bad anyway, which is also noted previsouly. And then learn how to handle the sensors (not moving them constantly to try to get connection). Maybe moist the forehead sensor and place it just above the eyebrown.

    If this basic BT setup is in place, it seems like even a 12 year old boy can start it up and use it after a little introduction of the handling.

    Just a note rgd. wired headset. I believe that the reason why the MindSet works so relatively stable regardless of few sensors, is that it is wireless, so all the grounding noise issues are out of the way. I read that noise is the main reason for unstable behavior by other wired sensor-solutions, where you have to touch something or sit very tight. The signals from the sensors are very weak, and these solutions are inheriently very sensitive to gound/chassis noise, unless designed 100% balanced with a very high common mode rejection. Especially in PC environments with switching power supplies and clock circuitries with fast rise times, making the ground jumping up and down many places in the signal path. Also gound loops are common, and the long wires pick up noise from various electromagnetic sources around the PC. So NeuroSky, if you read this, keep the wireless solution, but make the BT pairing / connection easier to use and more stable. It should also be easy to improve the sound quality ;-).

    Actually I wish the MindSet unit did not have the headset capability, but only the sensors, so I could use it without covering both ears, as I always turn down the sound anyway (terrible sound quality and I prefer to meditate without music). Also, the earcups makes my ears warm, and limit my meditation time, because they my attention are turned towards them instead. And it would last longer on the battery as the power for sound circuitries etc are eliminated.

    Has anyone seen an application where I can log the “Meditation” level e.g. every half second, and get a graph of a meditation session over time (like one hour)? Or is this easy to develop with the SDK?

    Regards
    Sven

    Comment by Sven — December 16, 2010 @ 16:40 | Reply

  10. Thanks for the review. I’m not sure if the product has improved since you used it last, but I see it all over online these days. i think this morning I found it on twitter and gizmo.com.

    Have you heard or tried the new version?

    Comment by Ken — January 12, 2011 @ 19:50 | Reply


RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Create a free website or blog at WordPress.com.

%d bloggers like this: