Hacking The MindFlex, More!

[vimeo http://vimeo.com/10184668%5D

Reader [Eric] sent us a powerfully informative, yet super simple hack for the MindFlex toy. Don’t worry, it’s not another worthless shock ‘game’, And it’s using an actual interface instead of the built-in LEDs.

With two wires for the serial protocol, and an Arduino, you’ll be able to view “signal strength, attention, meditation, delta, theta, low alpha, high alpha, low beta, high beta, low gamma, high gamma” brainwaves. While it’s not medical grade, it’s a lot more intuitive than previous interfaces.

The original intent was for a system called MentalBlock, but we’re wondering what would you do with brainwave data?

59 thoughts on “Hacking The MindFlex, More!

  1. @CornBrown

    I am really glad I am not the only person who has wondered what that would be like. Depending on the effectiveness of TMS and the ability to isolate the sensor from the effector . . . well, things could get interesting.

    More immediate, with the ability to sense various waves and muscle movements, this could make an interesting addition to a sleep log. Biofeedback and the like for any purpose, medicinal or meditative.

  2. For only a little more money you can get a NeuroSky MindSet, which combines the exact same single-channel EKG with a pair of headphones, all connected to the PC via bluetooth and with a published API for collecting the data.

  3. @ErebusBat

    I might not go so far as to call it empirical data of effectiveness. It will tell you if you hit one of the widely known waves, but not how you specifically react to it. Still, I may get one just to see the effects of binaural, dream machines, music, etc.

    Putting it next to a pair of headphones just seems . . . bothersome. Sure it is only looking at 50Hz and below, and while the music it self will be on a higher frequency, I wonder how it would deal with, say, a 2hz drum beat. Wouldn’t look much like a sine wave but a very low duty cycle square wave, and I have no clue how the built in chips would filter that.

  4. @Eric: I tend to include in my estimation of such things the cost and risks of doing the hack itself. Granted the risk of destroying your mindflex is much less now that someone’s figured out how to do it, but it still takes time, labor, and that board everyone loves that starts with an “A” to get the data where you want it; the MindFlex comes out of the box ready to use.

    Now the MindFlex also has limited battery life, and can be a nuisance if you *don’t* want the headphone function. But if you want to see your brainwaves, with the MindFlex you can be doing that about 15 minutes after signing for the package.

  5. : If you want to pipe audio to your ears, while scanning your .wavs (heh), might I suggest a stethoscope? Simple mechanical audio transmission, put the speaker on the pickup. (maybe even a crappy little piezo?)

  6. Use to to measure your sleep cycles to see if your CPAP is functioning properly and ensure your alarm doesn’t go off so you are the least groggy.

    Cheaper than a Zeo and more functional.

  7. Things like this make me wonder if “The Wire” from the movie Strange Days is just a few years away…

    How far can this stuff go now? I mean, are we an OpenCL application from being able to read/record the output from the eyes/ears to the brain, and then being able to playback on demand?

    A friend has just turned me onto this site, and I can see a few hours of archive crawling in my near future!

  8. Well with it attached to an audrino I can make an LED flash!

    Jk Jk.

    You dont have to use the audrino if you can understand the programming. Use a Basic Stamp or a Propeller or pic 18 if you want. The code is open source so hack it!

    I could see this being used to control a robot or to suspend resume a computer. Hell turn some lights on and off. Have your kitchen make you a snack when your stuck on a problem. If you can think it you can do it.

    Pun intended

  9. could wire it up some rgb leds attached to jordy laforge glasses to the arduino as well and test effects of color therapy. maybe make a game that when calm, colors cycle gently and eveny and get chaotic when aggitated.

  10. @lowlysoundtech: Love the idea, but wouldn’t chaotic-like LED’s almost…make the problem worse?

    Building on your idea…how about doing cool little patterns when somebody’s calm, but upon agitation, running a calm-like color cycle.

    Look up chromatherapy, that’ll help.

  11. for what Xeracy said, i was going to buy the Neural Impulse Actuator from Newegg like a year ago but then some reviews or something said it was just some coils of wire for $100, with bad grounding, so i decided to wait for a later revision. plus, all my interesting brainwaves have been terminated. if i was still hearing voices and seeing things, i would have bought the damn thing and did some X, etc.

  12. There is no real Brainwave meter to buy. This is all a big waste of time and money. They put this Mindflex on a plastic barbershop head, and the Waves looked quite the same as if we put it on a real persons head … well either a smart plastic head or just dumb humans? No. Plastic Toy waste – made by Mattel ;)

  13. @hackbert130, that is completely untrue. The Mindset outputs the raw data stream which is fed to the FFT layer. At the suggestion of a friend who is an actual medical doctor I tried putting the headset on my knee. Very different data. Now, it does generate the two derived outputs, “attention” and “meditation” from your knee, which is kind of funny, but you can easily tell it’s not your head from the other frequency band outputs. And as for a plastic head, you will get no data at all.

  14. seems like you could just fit a teensy in the case there insted of the arduino

    then maybe have a multicolored LED on the outside, and people could see your brain state
    like insted of a moodring, a brainlight
    or a lightbulb over your head for when you get an ‘idea’lol

  15. I’ve watched far too many painfully poor hacking videos. it’s quite refreshing to find one done right. Eric did an excellent job. very clear providing all of the needed information at a decent pace. and, no stupid intro. he got right to work the moment the video started. everyone should pay close attention to how this was done, especially if you ever plan to make a hacking video of your own. let this be the new standard.

  16. @Dox: +1

    Mood indicator…figure out how to sense the brainwaves from a small distance, transmit signal to an RGB place discretely ouside the boss’s office.

  17. @AgentRitzel He shorted the contacts with a wet towel, it varied between high and low without other input, then declared it a hoax.

    Considering he does transcranial magnetic stimulation research, you’d think he’d be aware that EEGs will report false activity when not attached to anything. Heck, fancy fMRIs will report brain activity in dead salmon, even when correctly calibrated.

    I rather wish he taken a little effort and compared the device’s behavior to a real EEG and shown it to be statistically uncorrelated instead of showing that a circuit with no input responds to the noise in it.

  18. I think one of the differences between the MindFlex and the MindSet is that the MindSet outputs the raw data as well as the FFT frequency bands and the derived “attention” and “meditation” values. It is quite obvious from the raw data that there are brainwaves in there. There’s also a lot of noise, but NeuroSky’s big claim is that their fancy software can filter that out.

    One can debate whether one noisy EEG channel is worth anything, but it seems very obvious that they really are amplifying the low-level signals that contain EEG data, they really are doing FFT on it, and they really are using that as the basis for their other data products. They are not doing GSR and by any reasonable definition what they have produced really does produce output based on brainwaves.

    So stop lying about this. The device may not be as useful as neurosky thinks, but it definitely is doing what they say it does.

  19. I haven’t seen this done anywhere yet… but how hard would it be to get signals from the brain that you could contol X and Y with. Think simple mouse movement. (without right click or left click)

    There’s lots of Lou Gehrigs and Muscular Dystrophe victims that would love to see anything down this line developed.

  20. A few electrodes on your heads are like a 8-pixel camera looking through a centimeter thick opaque piece of nylon, and trying to describe a scene from the results of that, and yeah you might conclude there’s movement or large shadows possibly humans in that analogy, but don’t think that it can ever be tweaked to be much more than that with any kind of processing, although obviously in such a case you could combine readings and filter them and improve things a bit with such tricks it’s limited how much you can do nonetheless.
    So getting back to electrodes: don’t worry, this won’t ‘read your mind’ or end up ‘recording your dreams’, don’t confuse TV shows with reality when science or technology are in play.

  21. @carzRfun

    Getting two channels and mapping them to X/Y would not be hard. Teaching someone to use those two patterns as X/Y is the hard part.

    As for ‘hoax’ . . . if you short out sensors by connecting them together, or to a wet towel, they go a bit random and act more like loops of wire. And loops of wire, with some capacitance from the towel, will pick up external signals they are not designed for, like mains hum, radio, etc. Cheap EEGs have been tested, they have known weaknesses like muscle movement in the patient’s face, someone moving near the sensors. Trans-dermal EEG works better, but a non-medical device will probably not require putting needles into the scalp.

    The toy itself, not the MindFlex headband but the floating ball part, is meant to be simplistic and not really controlled by individual bands of brainwaves. If the headband has that extra data available, why not use it.

  22. @Whatnot

    If you took my jests that it would be a good match for home-brew TMS for sharing thoughts, you over read it. While the idea is cool, this isn’t it. Yet.

    For reading minds or recording dreams, it will not be something crystal clear, it won’t even tell you if you dreamt of a person. But it might tell you what mind set you were in during a dream. REM should show up if you had the raw data, different bands will show up and let you see certain responses of panic, happiness, etc. It’s a small bit of vague information, but more information than is available without similar toys and tools.

  23. I am so there! As one who’s made binaural beats available online for some time now [www.JetCityOrange.com/meditation/] I can see the advantage of a setup like this. Much more fun that Journey to Wild Divine, which honestly never caught my attention.

    My dream hack? Using something like this to keep track of where you are *AND* setting up your goal so you can see how close you are by visual feedback. Next step, adjust the binaural beats audio you’re listening to in real time response to what your brain is doing. Can’t get out of beta brainwave range? Lower the binaural beat audio file being generated going to your headphones.

    Heck, I’d never come up for air! YMMV


  25. This hack is awesome. Ok, so the data you get out isn’t reliable enough to analyse for scientific purposes. But that is completely besides the point of this project. Al they needed was an estimated measure for Alpha and Beta brainwaves.

    @Allan Cottrill
    Try using biofeedback to teach yourself not to use all caps. As for the muscles you didn’t know you could use, next session, practice on your brain.

  26. @Quin
    It was more a general statement, I also see in TV shows and such rather overstated possibilities of a few head electrodes, and it’s OK for fun but from various posts on various places on the internet a bit too many people seem to buy into TV-show exaggeration.
    Of course on hackaday you hopefully see people with some more tech sense so you won’t see many here I expect that buy that much into the overdrawn stuff I guess.
    But let my post then be there in case a senator stumbles upon this article ;)

  27. @Whatnot
    senators . . .
    on hackaday . . .
    now that is a truly frightening thought.

    although, mind-reading with a kit like this is a lot more possible than enhancing two pixels to find out that the killer has their back to the camera, then turning the image around to see their face.

  28. Maybe one of you can make something this faster than I can: http://www.emc.com.tw/twn/database/Data_Sheet/COM/EM198810.pdf

    It’s the transceiver chip used in the wireless module used to connect the headest with the console. A half hour search turned up that document, after trying to figure out the website. The actual module’s datasheet is a bit more elusive, but it shouldn’t be hard to reverse engineer one if you have a Mindflex laying around.

  29. To correct my previous post, please note page 16 of the datasheet, where it gives a PCB layout for the schematic in page 15. The board layout appears to be different from that of the Mindflex, but at least the pins connecting to the leads should be the same, if not in the same place.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.