MindFlex Teardown

Maybe we’re just imagining things, but it seems to us like brainwave control is the latest trend in toys. Similar to Uncle Milton’s Force Trainer, Mattel has recently released the MindFlex, a game that involves moving a plastic ball up and down through an obstacle course that you control using your brainwaves. Naturally when [Alpha] saw this, he decided to take it apart and document what he found. After disassembling both the headset and the base, he found that most of the chips were covered in black resin making them unidentifiable. However, he was able to find identify one chip, the NeuroSky TGAT1-L64 D498Q-010 0924. Judging by the name alone, we would guess that this is the chip that makes the brainwave control possible. While there’s no mention as to whether you’ll be able to interface with this like you can with the Force Trainer, we’re sure that it’s only a matter of time before someone figures out how to use this to control more than just a floating plastic ball.

36 thoughts on “MindFlex Teardown

  1. On the german amazon-page you can find a description that it als works if you touch only your fingers with the electrodes and it is suspected to only use random movements… can somebody check if it does the real thing or is fake?

  2. It seems like brainwave control has come in and out of vogue since the days of Omni magazine. I’m sure the technology has improved, but it sort of seems like the technology equivalent of ska in that respect.

  3. Yeah, get some industrial strength paint stripper and have some acetone handy to wash it off the chip with. Recommend against rubber gloves, unless you know the paint stripper won’t melt them. Melted gloves + hands = bad.

  4. @jon
    if you do that the chip will be destroyed.

    I was thinking of an alarmclock that you could press the “sleep” button with your mind.

    OOH! better yet,
    an ipod that you can go to the next song with your thoughts. shouldnt be too hard, just take the motor control system and measure the pwm. if it reaches a threshold, set off the “next” button.

  5. @googfan
    The problem with using these devices in those types of applications is that so far they all seem to require you to concentrate on a single thought. So every time you wanted to change a song on your player you would have to sit still for a few seconds and concentrate on one thing.

    It would be a cool little project and a neat novelty, but there aren’t many realistic situations where that would be useful. For example, I am generally sitting at the computer working on something when I am listening to music. I wouldn’t want my song changing every time I think about why a function isn’t working the way I intended.

    Though I imagine there are some interesting applications for a device that reacts to concentration or relative stress level. In fact, a few nights ago I was thinking of a way to use one of these to build a meditation trainer. Punish the user with something annoying (loud noise, electric shock, whatever) if their concentration deviates for too long.

  6. actually if you take a look at uncle miltons pcb prints you’ll see neurosky made the board and chips for the mind controller. been thinkin about buying the chips and just building my own actually.

  7. wolfy02 makes a good point. It’s like this with most chip makers. It’s how you create incorporated vendor lock without breaking trade laws. I’ve dealt with this first hand with Zilog and TI.

    It’s done partially with availability, but it’s constant with special features especially pertaining to embedded security APIs.

  8. I’d be wary; If I were writing a game for the average Joe based on mind control, where you /CANNOT/ lose, I’d probably think about adding in failsafe, if tries>30 forceWin = 1… you know, in case Joe has trouble thinking for himself and stuff.

  9. You’ll notice that if you do make faces, the game becomes incredibly easy.

    These sort of devices are fine if all you want to do is flip a switch in an incredibly overcomplicated manner, but if you want anything more complex than on/off, they’re useless and you’re better off creating your own solution. I wonder how easy it would be to hack your own low-resolution fMRI hat.

  10. Lets face it… they try to record a very small signal which is coming from somewhere inside a splashy grayish mass of living cells and blood vessels…shielded by an 6-7mm thick skull surrounded by muscles, skins and hairs…

    Everybody who expect that this toys can really do more rather then playing a bit around with the noise-level here is an analogy…

    try if you could place some electrodes outside on your PC chassis and try to record the content of the ALU of the third core of your quad-core processor in real-time during that you should run a povray rendering on all four cores, play MP3 on the internal sound card. Copy the content of your partion 1 to partion 2 and burn a DVD….

    if you succed with that you might can try again to read thoughts….

  11. I’ve got to agree with tux-fan. To pin-point with such accuracy would be insanely hard, not to mention the other technological hoops we would have to jump through first. But this is most definitely the first step in the right direction.

  12. i tried one of these out at pax, it actually worked pretty well. it was controlled by the amount of brain activity. to make the ball go down you had to relax and clear your head, then when you want it to go up, you start doing some simple math or anything that excites you. it sure was hard with all the booth babes in the area though. you feel kinda silly with those things clipped to your ears.

    in the end, its really hard to control and i cant imagine it being to accurate. and i don’t think it would work for multiple actions, might work well for a mind controlled dimmer switch though.

    i suspect this thing is only measuring blood flow to the brain, but i really don’t know. i just don’t see any other reason to clamp it to your ear lobes.

  13. It’s a cool game but always out of stock.

    I just wanted to share a website that I found that will alert me the instant Mindflex is back in stock at Amazon. You can use it for any item at Amazon for that matter. It checks Amazon every few seconds automatically and alerts you the moment your product is back in stock.

    Here is the link (the item # is for Mindflex).


    You can also just go to http://az-grabber.net and search for items.

  14. Actually after taking a closer look it appears to use pressure sensors in the head band and ear clips. As you concentrate your face tightens and when you dont concentrate your face relaxes. try it with your fingers it works easier. Just tense and relax your hand.

  15. If anyone takes the time to do a bit of reading on Neurofeedback (the technology by which this game functions), you’ll find that the current practices lean towards SMR and Alpha protocols. For those without knowledge of what this means, there are various frequencies when it comes to your “brainwaves”, ranging from delta associated with sleep, to beta and high beta associated with normal every day concentration and anxious concentration respectively. Now, the current practice leans towards SMR or Alpha because SMR is associated with a relaxed kind of intense concentration, while alpha is generally more relaxed. It’s much more complex then this, employing databases to see variations in your brain and that of others, and if you care you can wiki it :).

  16. I have one mindflex toy and I was impressed on how it works- so I have started one short documentation on the product:
    – the microprocessor on the input electrodes PCB (which is on the head set) in one powerful ASIC with a lot of specific software inside ( NeuroSky TGAT1-L64); it can deliver lot of informations, not only the “state of signal” (poor _signal) , “attention” and “meditation level” used in this game-toy but also the alpha, beta, delta and theta waves level obtained with a FFT transform from the input forehead signal (the active electrode point is Fp1, and the ears refereces A1 and A2- http://www.bem.fi/book/13/13.htm); the output of this input electrodes PCB are 4 pins having the notations (T,R, V, and un-noted one) very probable Tx, Rx, Vcc, ad GND (i have not verified yet); the documentations on the “developer tools” (http://www.neurosky.com/mindset/devtools.html) is very useful to understand how the serial data_row is structured;
    – on the main PCB (on the head set also) there is also one interface from serial to bluetooth RF module ; in this side the serial data are prepared for wireless protocol – I believe that is a specific protocol for this toy and cannot be intercept with a standard bluetooth protocol – not very ease maybe..:)
    – the wireless communication of the mindflex is the “2.4GHz RF module” made by ELAN microelectronics http://www.emc.com.tw/chs/ap07_01_01_02_gp.asp

  17. I must admit, I’ve always thought that “mind control” would be far easier done by deliberately *targeting* scalp muscles. So it’s scalp control, rather than mind control, but you let the brain handle the mapping between scalp muscles and commands. Think of the BrainGate tongue vision thingy but in reverse.

  18. Hi I read the comment and they reminded me of a brain cap. I came up with that uses pic’s it is made up of two 48pin pic’s & one 24pin bluetooth pic each pic’s has 20 electrodes. I got the idea from two books bionics for the evil genius and 123 pic microcotroller experiments for the evil genius. ever seen lawnmower man or the matrix

  19. I actually bought this game about 3 days ago from Wal-Mart for $70, and I have to say that it works exactly how Matel says it does, which impressed me a lot. The headset starts to get uncomfortable after a while with the metal sensor that must be placed above your left eye, but its something I have gotten used to. I would have to say that I don’t believe it relies on any facial expressions, or any kind of blood pressure measurement. It took me about 5 minutes to really get the hang of it. I laid down in the floor with the headset on and the game in front of me. I turned on the system and selected the freestyle option. I relaxed while trying to clear my head and the ball dropped as the fan completely stopped. The moment I started concentrating, the fan turned on, and the harder I concentrated the higher the ball went up. Then once I started thinking about something else to take my mind off of the ball, it stopped again. I repeated this over and over. I have even tried this with my eyes open and closed, and still got the same result. With everything I tried, the only thing I could not rule out, was the fact that it “truly was” reading my brainwave activity. Some people say that it has a pre-programed routine, which isn’t true, because I tried clearing my head for almost 2 minutes, and it stayed still. It always responded as soon as I “randomly” concentrated, with almost pin point accuracy. So either I have telekinesis, (Which is very doubtful) Or the game actually does measures your brainwave activity with a true EEG headset, (Technology that has been around for a while) which is probably the most accurate statement. This game is not a fake nor a gimmick. (And no I don’t work for the company.) I just think that all of those who can’t believe something like this can truly do what it says, are the same ones who won’t accept that technology has truly advanced this far to make things like this possible. Heck, being in the military, I have seen cutting edge technology not available, nor known to the public as of today, that I didn’t even think was possible, until I saw it with my own eyes, and knew and understood the science behind it.

  20. Howdy just wanted to give you a quick heads up. The words in your post seem to be running off the screen in Opera. I’m not sure if this is a formatting issue or something to do with browser compatibility but I thought I’d post to let you know. The style and design look great though! Hope you get the problem resolved soon. Cheers

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.