Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!


Forgot your password?
Hardware Hacking Toys Build

Brain-Controlled (Inflatable) Shark Attack 17

the_newsbeagle writes: This is a parlor trick, not neuroscience," writes this DIY brain hacker — but it sure is a nifty trick. The hacker put electrodes on his scalp, fed the resulting EEG data into a specialized processor that makes sense of brain signals, and modified the remote control for a helium-filled shark balloon. Soon, he and his buddies were steering the shark around the room. Why did it take his buddies, too? "EEG interpretation is not easy because, to be technical, EEG signals are a crazy mess. EEG recordings are a jumble of the signatures of many brain processes. Detecting conscious thoughts like “Shark, please swim forward” is way beyond even state-of-the-art equipment. The electrical signature of a single thought is lost in the furious chatter of 100 billion neurons." So builder Chip Audette settled on the simplest control system he could, and divvied up the actual controls (left, right, forward, etc.) among several users, so each one's brain signals could be interpreted separately.
This discussion has been archived. No new comments can be posted.

Brain-Controlled (Inflatable) Shark Attack

Comments Filter:
  • by Anonymous Coward

    No deal.

  • by Opportunist ( 166417 ) on Saturday September 26, 2015 @10:43AM (#50603655)

    Then I read that he split up the control over a couple people. Ok, folks. That's been done. If I had known that this was /. frontpage material, I'd probably even have violated the NDA I'm under.

    While detecting conscious thought may be tricky (though I don't know for sure, since it isn't the scope of my research in this field), detecting the conscious will to move something (or, more specific, want a change to happen) is well within the reach of hobbyists. What IS hard again is that after a while of doing this, the brain changes the way it "thinks" about a specific "want to change". My guess so far is that this has to do with learning and moving the newly learned trick from the "I'm learning" to the "I'm remembering" parts of the brain. The hard part is that there is no "hard" transition. It happens gradually. You're getting a lot of conflicting data that you can't even reproduce sensibly because you can't roll a brain back to the yesterday position.

    Generally a very rewarding field. And a damn lot of room to work in. We know so little about this so fascinating biological computation device.

    • by gringer ( 252588 )

      I've mentioned this before, but I might as well say it again. Our nervous system is designed to learn how to control random systems, so putting a "specialised processor" in between the nervous system and a control system is probably not going to improve things, and may make them worse. Just provide the most basic interface possible to the control system via the nervous system, and let the brain (or any other limb) do the training.

      • You still have to provide some kind of interface, i.e. some way to detect and interpret the nervous system's behaviour. That alone is already a feat and a half, because so far we're still pretty much pissing in the wind when it comes to finding out just WHAT to interpret.

        • by gringer ( 252588 )

          i.e. some way to detect and interpret the nervous system's behaviour.

          Detect and react to the nervous system's behaviour. Let the nerves do the interpretation.

          • OK, but react to what? You get an insane amount of noise and maybe a percent of signal. Finding that percent is the name of the game.

  • It's quite impressive considering the hardware. The demonstration uses a "trick" used by practically all current Brain-Computer Interfaces, which is to pick the more conspicuous brain wave patterns and use them, rather than abstract thoughts of movement, to steer the machine:

    Instead of looking for specific thoughts, I looked for an EEG signature that would be naturally easy to detect and that I could use to signal intent. The easiest such signal occurs whenever you close your eyes: For most people, when the

  • ... including readily available toys.

    https://en.wikipedia.org/wiki/... [wikipedia.org]

    If it is on wikipedia, it is not news, and there is absolutely nothing different from controlling a ball (1 D) and controlling 1 D of the motion of a toy shark. There are even more complex mindflex games out there that suggest that one could probably rewire one to allow a single person to control the shark. There is a mind controlled UFO (drone) on the market already that is probably much cooler, except oh wait! It isn't

panic: kernel trap (ignored)