Slashdot Mirror


Device Reads Messages From Surface of the Brain

Al writes "Technology Review has a story about a start-up company that has developed a more-accurate and less-invasive way to read a patient's thoughts. Neurolutions, based in St Louis, has developed a small implanted device that translates signals recorded from the surface of the brain into computer commands. The device, which is less invasive than implants and more accurate than scalp electrodes, uses a grid of electrodes placed directly on the surface of the brain to monitor electrical activity. This technology is currently used to find the origin of seizures in patients with uncontrolled epilepsy before surgery. But the company says it could also help paralyzed patients control a computer and perhaps prosthetic limbs using their thoughts. Tests involving more than 20 patients have shown that people can quickly learn to move a cursor on a computer screen using their brain activity."

13 of 156 comments (clear)

  1. Input device by sakdoctor · · Score: 4, Funny

    When this becomes a standard human input device...I don't want it. How will you explain when your browser suddenly navigates to your favourite porn site.

  2. Nothing new, but is it efficient? by Co0Ps · · Score: 3, Interesting

    Non invasive brain interfaces is nothing new. Here's a video of a HL2 mod where you're using your mind to pick up objects and throw them at other players. The question is if the mind reading is accurate enough to actually control a mouse pointer efficiently or reliably start macros (voice recognition style).

    1. Re:Nothing new, but is it efficient? by QuantumG · · Score: 4, Interesting

      It's hardly "non-invasive".. they have to open your skull to implant it.

      --
      How we know is more important than what we know.
    2. Re:Nothing new, but is it efficient? by venicebeach · · Score: 4, Informative

      The signals in that video are recorded from the scalp. Basically when you filter the electrical signals from the brain through the skull you lose a lot of spatial resolution. Given that spatial maps are one important way the brain encodes information having the electrodes actually on the surface of the brain makes a huge difference in the amount of information you have access to.

      That said, this is not really a new technology, merely a new application of electrocorticography. Non-invasive it is not, since it involves opening up the skull. It's only "less invasive" compared with poking an electrode deep into the brain.

  3. How much... by Darkness404 · · Score: 4, Interesting

    How much longer till we can figure out how our brain "codes" things then exploit it for our own benefit? Just think about it, custom-made drugs to make it seem like you are flying, fighting a dragon, more epic than any video game imaginable, all while being perfectly controlled with little to no side effects. Or take a pill and have the entire library of congress memorized. I wonder how much longer this will take.

    --
    Taxation is legalized theft, no more, no less.
    1. Re:How much... by FiloEleven · · Score: 4, Interesting

      The way the brain works makes such experiences nearly impossible to encode and certainly impossible with drugs, as another poster pointed out. If we take "drugs" to include "nanomachines targeting the brain and disguised as a pill" then we enter the realm of the merely highly improbable. The machines would have to collectively be smart enough to override signals from and to the body while simultaneously generating the desired experience.

      Another possibility would be a single device at the top of the spinal column and networked with lots of processing power, like a wireless Matrix or the Vertebrane system from Manna. This too requires advanced nanotech to implant, as it must splice every nerve in the spinal column plus the optic and aural nerves, and so is also highly improbable to occur at all, and certainly not for nearly a hundred years unless the Singularity frea--er, folks are right.

      Given all that, the experiences you speak of (flying, fighting a dragon) could happen, but doing the "I know kung-fu!" thing is impossible due to the nature of consciousness. If you want to learn something, you're going to have to spend the time to learn it. Reshaping synapse connections and brainwave patterns to implant memories requires godlike knowledge of the individual's brain state and history. Let's not forget that we are messy meat machines (if machines we are) whose sense of self and memory is only infinitesimally less mysterious now as it has always been. Faking an external world and letting the brain experience it, hard as it is, is orders of magnitude simpler than fabricating a past experience, especially an intellectual one such as memorizing the LOC, out of whole cloth.

      Sorry I'm such a party pooper =(

  4. "More-accurate and less-invasive"? Not so much... by DragonWriter · · Score: 5, Informative

    Technology Review has a story about a start-up company that has developed a more-accurate and less-invasive way to read a patients thoughts.

    "More-accurate and less-invasive" is misleading, since the thing that it is "more accurate" than is not the same thing it is "less invasive" than. It is more accurate than the minimally-invasive electrodes-on-the-scalp method, and less-invasive than the more accurate electrodes-implanted-into-the-brain method.

    It is, likewise, less accurate than the electrodes-in-the-brain method, and more invasive than the electrodes-on-the-scalp method, so it would be as accurate (and as hyperbolic, in the opposite direction) as TFS to call it a "less-accurate and more-invasive" method as it was to call it a "more-accurate and less-invasive" method (simply switching which existing method it was compared to for accuracy and which it was compared to for invasiveness.)

    It would be most accurate (and not at all hyperbolic) to call it a method which is intermediate between two existing methods in terms of both accuracy and invasiveness.

  5. Re:Killer App by CarpetShark · · Score: 4, Funny

    If they get a cursor to follow my eyes i'll wield the scalpel myself!

    I hope you don't twitch under extreme pain. Could end up in some kind of endlessly recursive feedback loop. Which would hurt. Muchly.

  6. Re:ohh by CarpetShark · · Score: 4, Funny

    I WOULD HAVE to say that ALL METHODS OF thought READING are INVASIVE.

    I liked the other thing you were going to say much better.

  7. Groundbreaking news! by feepness · · Score: 5, Funny

    Neurolutions, based in St Louis has developed a small implanted device that translates signals recorded from the surface of the brain into computer commands. The device, which is less invasive than implants and more accurate that scalp electrodes, uses a grid of electrodes placed directly on the surface of the brain to monitor electrical activity

    Awesome! They developed an implant which is less invasive than implants!

    Next up, a duck that quacks louder than a duck!

  8. "Less Invasive"? by DynaSoar · · Score: 4, Interesting

    Laying an electrocorticogram array (that's what they're using -- it's not new) on the surface of the brain requires removing a section of the scalp, skull and dura mater. There's nothing about it that's not invasive as well as dangerous. Single cortical or deep electrodes can be put in through very small drilled holes. The former requires a full neurosurgical suite/team. The latter can be done in a clinic visit if localization isn't critical, or else in a CT or MR scanner with no more invasive electrode technology than the clinic version. The draw back to implanted electrodes is that inserting them into proximity of the neurons of interest can cause them to die off immediately, and will cause them to die off eventually.

    Both are unnecessary for the application. In 1994 a researcher working at Radford University with Karl Pribram developed an EEG analysis program that could recognize various shapes, sizes and colors (various combinations thereof) of objects both seen and only internally visualized, with a 95% accuracy. Such accuracy among the many permutations of possible signals could very easily translate into control signals sent to another device. Fully designed but not built around this technology was such a control device intended to run an 8 stepper motor robotic arm using a standard parallel printer port. Since it rests on the scalp, an EEG electrode such as we used here is not invasive in the least. Well, the sticky glue electrode paste necessary to keep the electrode on and conducting for several hours tend to pull out hair, but that's annoying and slightly painful, but not invasive.

    --
    "I may be synthetic, but I'm not stupid." -- Bishop 341-B
  9. Re:Cool by Idiomatick · · Score: 4, Informative

    Offtopic: You know ~ is already in use to elongate words. Or make them sound more musical.
    ex. Hi~
    Hm~~~~~

    The origin is Japanese where a double vowel word like konpyuutaa is written as konpyu~ta~ (written in japanese character of course). Written to drag it out you would write one really long tilde but since the advent of computers generally people use a chain of them together. Also of note that you might be interested in from japanese culture. Japanese people often end sentences with a ;; or even shorter ; to represent a type of sadness or confusion( ;_; is a sad emoticon in japan and ^^;; is confusion (sweat drops)), this is possible since the semicolon doesn't exist in Japanese. As well some people use ^ at the end of a line for happiness (from ^_^). And // for.... ughh or you are an idiot (from -_-//). There are other various sentence endings that take part of the emoticon and attach it to the end to refer to different things. And japan has hundreds of different kaomoji(emoticons) unlike the 10 we might use. And so you don't need to ask, there isn't to my knowledge a line ending for sarcasm. I think it'd defeat the purpose of being sarcastic anyways :P

  10. Re:Other uses can't be far by dissy · · Score: 3, Informative

    "connecting cameras to their tongues" WTF?

    The original ScienceNews article from 2001 is now subscriber only:

    http://www.sciencenews.org/view/feature/id/1946/title/The_Seeing_Tongue

    But you can read a copy of it at:

    http://findarticles.com/p/articles/mi_m1200/is_9_160/ai_78681631/