Become a fan of Slashdot on Facebook


Forgot your password?
Science Technology

Scientists Build Mind-Reading Computer 108

An anonymous reader writes to tell us that researchers from Carnegie Mellon University have developed what they are calling a "mind reading computer." Using a panel of nine volunteers, the team built a "profile" of 58 test words based on brain scans taken while the volunteers were directed to think about the meaning of each test word. "'If I show you the brain images for two words, the main thing you notice is that they look pretty much alike. If you look at them for a while you might see subtle differences,' explains Tom Mitchell of the Machine Learning Department, which lead the study. 'We believe we have identified a number of the basic building blocks that the brain uses to represent meaning. These building blocks could be used to predict patterns for any concrete noun,' added Mitchell."
This discussion has been archived. No new comments can be posted.

Scientists Build Mind-Reading Computer

Comments Filter:
  • by Intron ( 870560 ) on Monday June 02, 2008 @02:12PM (#23629369)
    The list of words chosen were: funding, grant, tenure, award, patent, contract, ...
  • All you have to do is have a different connotation for the word and it doesn't work. The gays stole rainbows so now if people see a picture of a rainbow, they have a distinctly different reaction to it. Or you could purposely make yourself feel angry or sad or do a complex math problem as you're thinking of the word and it would throw the machine off. To get this to work I'd bet they have to tell you to stay calm and what to think about beforehand, during, and after they try to predict what word you're t
    • Re: (Score:3, Insightful)

      by Facegarden ( 967477 )
      Not necessarily... If there truly are key areas that only deal with actually thinking about a noun, they should be unaffected by other brain processes like emotion, etc. One may be off daydreaming about that summer when they "experimented" with the neighbor boy in college, but the actual word "rainbow" is still sitting somewhere in his mind. ;)

      I would, however, be inclined to believe that our brains are more complex than just having "areas" that have "activity" when certain things happen. Until we can map
    • Re: (Score:1, Troll)

      Most of this technology is going towards research for hands-off computing and assistance for people who are paralised. Actual mind-reading technology is a long way off. And I suspect it will always be easier to do with drugs and good old fashioned bush-sanctioned water boarding (to get information from people unwillingly).
      • Ha! This got modded troll, but the parent post claiming "the gays stole rainbows" got interesting. God damn queers = A-OK Bush sanctioned torture = move along troll. You have to love slashdot.
        • What do you people have against homosexual men? I specifically mentioned 'men' as I suspect you have sufficient quantities of lesbian porn hidden away somewhere. Bloody double standards...
    • by dfedfe ( 980539 )
      I haven't read TFA, but I was just reading the actual scientific article before I came here. Your objection doesn't quite apply for a similar reason to what you mention: the software was separately trained for each of the nine subjects. For each subject you give them a set of different words (it was 60 words, five from each of 12 different categories of concrete nouns) and record an fMRI of each one.

      Now what you do is drop two of those out, train the software on 58 words, and see how well it can guess which
  • Excellent! (Score:5, Funny)

    by jimand ( 517224 ) * on Monday June 02, 2008 @02:18PM (#23629425)
    Now that a computer can read my mind I'm waiting for the mind-reading 'puter that knows to change the mouse focus when I look at a new window. I hate looking at one window while typing in another, especially when posting to /. while I have a window open with an email to my boss. It turns out he's not interested in the goatse link.
    • So you've never been looking at some data in one window while typing the description of the data in another?
    • We don't need mind reading to determine where you're looking... just a webcam and some good software. Also, there are programs that make windows (if that's your thing) change focus just by hovering the mouse over a window, so if you had the webcam track your eyes and control the mouse, you could do it with readily avaliable software and hardware today. -Taylor
    • by TheLink ( 130905 )
      So it's you who has been posting those goatse links! I doubt most of us here interested in goatse links either.

      Lastly, read my mind now...

      Thank you.
    • by zsau ( 266209 )
      My suggestion: Use focus follows mouse. Then you can just unthinkingly move the mouse pointer around to follow your eyes. It takes a short period of adjustment, and then you realise it's the next best thing to a computer reading your mind. If you're using Windows, try True X-Mouse. If you're using the Mac, tough. Macs don't read your mind; you adjust to them.
  • And so it begins.. (Score:4, Interesting)

    by multipartmixed ( 163409 ) on Monday June 02, 2008 @02:22PM (#23629469) Homepage
    ..THIS is the basis for yet-another-trek-related-invention: the Universal Translator.

    I always knew it had to work this way.
    • by the_humeister ( 922869 ) on Monday June 02, 2008 @02:27PM (#23629519)
      That's an interesting idea. Do different words that mean the same in different languages light up the same areas of the brain when a person thinks about it? Would a Spanish person who is told to think of "coche", have a similar brain scan of an English person told to think of "car"?
      • Re: (Score:3, Interesting)

        by zappepcs ( 820751 )
        Doubtful actually, at least in all cases. In English, nova has one or two meanings that may bring different thoughts. In some Spanish speaking countries, they might be thinking 'no go' or some option for various value of go in Spanish.

        Grammarians unite! Only those who understand language will be able to interpret the results of this machine.

        It is quite interesting that there are parts of the brain that light up uniformly (or near it) for some processes. Puts the human brain more in the land of machine with
        • It may not be exactly the same in every language, but there might be some similarities. With enough data and a good training set you can learn the most specific and most general hypothesis. Then, you have a representation of the hypothesis space (from specific to general) you are interested in, i.e to classify the word "car" in all languages.
        • by lbgator ( 1208974 ) <(moc.liamg) (ta) (uolo.semaj)> on Monday June 02, 2008 @04:53PM (#23631365)

          I think the GP is onto something a little different than what the parent interpreted. Language may be an unnecessary step in this experiment.

          If someone is thinking "gee - I would love a hamburger" in English - would their brain scan be the same as a French guy thinking the same? If you started at some basic level (hunger, thirst, anger, love, pain) is there a common denominator in all brain activity? If there is commonality, can we hope to someday eliminate language and have comms come straight from the source?

      • by why-is-it ( 318134 ) on Monday June 02, 2008 @02:50PM (#23629791) Homepage Journal

        Would a Spanish person who is told to think of "coche", have a similar brain scan of an English person told to think of "car"?

        Agreed. I suspect that true mind reading will be impossible because everyone will have different internal representations of concepts and ideas. Even amongst individuals who speak the same language, we should not assume that everyone will have the same representation of "car", even though people may have similar levels of brain activity in the same parts of the brain when they think about one.

        • Re: (Score:3, Interesting)

          by Feanturi ( 99866 )
          And then you put a person who was born blind into the MRI and ask them to think about a car. Now what?
          • by Artuir ( 1226648 )
            If they'd ever been in or around a car before there is much more to it than simply the visual aspect. Sounds, smells, and if it's a v8, the rumble you feel in your chest. There may still be common ground to detect the thought based on those parameters.
        • by Dgawld ( 1251898 )

          Even amongst individuals who speak the same language, we should not assume that everyone will have the same representation of "car"

          I know when i think of a "Car" it is nothing similar to that machine with a plastic frame and four wheels that others might imagine.
        • Re: (Score:3, Interesting)

          This reminds me of a psyche 101 exercise where students were asked to draw a map of the town in which the college resided. Upperclassmen drew far more detailed maps than freshmen. I suspect the upperclassmen would have thought of very different things(past experiences) compared to freshmen also, and that's within a very small subset of people. I doubt very much that this machine could "read your mind" primed with someone else's input, but it could be invaluable in determining how the brain works and what s
          • I don't understand what this exercise is supposed to prove, other than the blindingly obvious fact that upperclassmen have on average lived in that town for much longer and thus know it much better.
      • I don't think so. Nearly all words have several meanings; especially in English. The connections between words would also be different because in addition to relating words semantically we also relate words to each other by their sounds. So, aside from strong cognates, the connected graph of word relations looks quite different in different languages. I think only very basic proto-indo-european-ish words like mother and milk could match across languages.
      • that the word "fanny" will light up two rather different regions of the brain depending on which side of the Atlantic you were born.
      • by imess ( 805488 )
        Not only that. Even if 2 persons studied the same language, experience matter, i.e. the word "car" may create an image of a blue 4-door Civic parking in the garage to one, and a gray Explorer from TV ads to the other. I doubt the details are confined to one or two areas in the brain.
      • The same word in the same language probably lights up slightly different parts of the brain in different people. My understanding is that this method is only good for mind-reading the brain the computer has been trained on.
      • Probably not.

        Otherwise these guys would certainly have noticed and made a big noise about it: []

        "Responses among the eight subjects varied with the person and stimulus. "

        "For example, a single neuron in the left posterior hippocampus of one subject responded to 30 out of 87 images, firing in response to all pictures of actress Jennifer Aniston, but not, or only very weakly, to other famous and non-famous faces, landmarks, animals or objects. The neuron also did not respond t
      • No, there would be no way to develop a universal mind-reading device. Because each person's mind is unique, and each person's understanding of a given word is also unique. When you say car, I might think of a red ferrari while someone else thinks of a white focus. The way my mind processes the word car is based on all my memories of cars, so that's why they have to build a unique profile for me that won't work for anybody else. And to make things even more difficult, if you say car again tomorrow, this
      • Do different words that mean the same in different languages

        The problem with translation is that different words don't mean the same in different languages.

        Some words like 'go' can be found in every language with very similar meanings, but for most words which are used less often there are extra connotations and even contradictory meanings depending on the culture, and also who is speaking. For example in American English, British English and French the word liberal has different meanings.

        Even depending who says it, and in what context, the intended meaning may dif

      • That's an interesting idea. Do different words that mean the same in different languages light up the same areas of the brain when a person thinks about it? Would a Spanish person who is told to think of "coche", have a similar brain scan of an English person told to think of "car"?
        pr0n is a universal word and it lights up the same things for everyone
      • I am not a psychologist (but I'm married to one, so while I'm not qualified I'm not completely uninformed either).

        I think it would at least partially depend on how the languages are learnt. From memory the part of the brain used to store words is different if you learn a second language as an adult compared to as a child.

        So, if you show me the word velo or the word bicycle it is very unlikely that that would use the same part of my brain (as I'm learning French only as an adult) whereas someone who learnt F
  • Wonder... (Score:3, Funny)

    by ShiNoKaze ( 1097629 ) on Monday June 02, 2008 @02:23PM (#23629473)
    I have to wonder all this work we do towards reading minds, what's everyone gonna think when they figure out how much we really do think about sex? Cuz damn.

    Might be fun to watch the expressions on the scientists face as they realize what's going on tho. "That guy was a fluke, the next will about something else I'm sure!"

  • This is why passwords by themselves are fundamentally unsafe. Anyone typing a password "thinks" about the next character they're about to enter just before they type it. If concrete nouns can be can be scanned while the subject is entering the password, things as basic as the letters of the alphabet and the numeric system would be dead ringers for remote password stealing.
    • Re: (Score:2, Funny)

      by Drakonik ( 1193977 )
      Yeah, but I dunno how easy it would be to use social engineering to convince people to sit underneath an FMRI so you could scan their brain while they type in their bank's PIN number.
    • Re: (Score:3, Insightful)

      by Shikaku ( 1129753 )
      I remember my password by keyboard location by my fingers.
      • I remember my password by the keyboard location of my written password list taped to it.

        (all joking aside, that started about the time they started changing all of them every 60 days...)
    • Re: (Score:3, Insightful)

      by CastrTroy ( 595695 )
      That's why you need real 2 factor authentication. Something you know, and something you have works well. So that even if somebody peeks over your shoulder (or into your brain), to figure out the password, they still don't have access.
    • by ceoyoyo ( 59147 ) on Monday June 02, 2008 @02:47PM (#23629753)

      Here, would you please lie down while I slide you into this multi-tonne magnet. Thank you. Now, please lie very still and think about typing in your password, very slowly, one letter at a time. No more than one letter every ten seconds or so! Now please repeat a couple dozen times. Thank you for your cooperation.

      I think it would be easier to just ask.
      • by schon ( 31600 )
        Yes, because technology never improves to make things smaller, faster, and less obtrusive, right?

        That's why all cars only have 8HP, plane trips have a maximum distance of 300 yards, cell phones weigh nearly a kilogram and get 10 minutes of talk time, the smallest computer takes up several buildings and can only do 5000 math operations per second.
        • Re: (Score:3, Interesting)

          by ceoyoyo ( 59147 )
          There are just a few physical limitations on fMRI technology.

          The most important for this topic is that fMRI scanners measure changes in BLOOD FLOW. They do not measure electrical activity. The flow response is delayed by about three to five seconds and has a certain minimum time duration. Therefore the requirement to enter the password r...e...a...l...l...y s...l...o...w...l...y.

          Secondly, in order to get any recognizable imaging signal at all (and if you want to measure letters you're going to need a REA
    • I think it is dangerous in a lot more serious way.

      Basically the thing can read a brain scan, what if they develop the technology to stimulate the brain to produce a predefined brain scan. I.e. implant a thought pattern.

      • by fritsd ( 924429 )
        Saw that on TV, years ago.

        Do you mean like Temporal lobe tickling and spirituality TV program (in Dutch) []?

        The reporter was quite impressed, I remember that. But he did'nt feel it as a spiritual experience.

        I have no idea if what you propose is possible: maybe, people's brains are wired sufficiently differently to make this "implanting a thought pattern" very difficult, unless it's a very crude pattern. Plus, the "please put on this special motorcycle helmet" would give the plan away ;-)

  • by davidwr ( 791652 ) on Monday June 02, 2008 @02:25PM (#23629503) Homepage Journal
    Hold on, just got handed this printout:

    "Thank you, but we already knew you were going to say that.


    Your new mind-reading computer overlords."
  • I, Robot story (Score:2, Informative)

    by Anonymous Coward
    Wasn't there an I, Robot short story by Asimov about a mind-reading computer that lied to people in order to avoid hurting their feelings (because that would "harm" a human)?
    • That was my favorite story, actually. It showcased what an utter bitch Susan Calvin could be, and also how utterly brilliant she is. Woman of my dreams.
      • Re: (Score:3, Insightful)

        by Eternauta3k ( 680157 )
        Utter bitch? That story showed she had feelings!
        • Re: (Score:3, Insightful)

          by Zarf ( 5735 )

          Utter bitch? That story showed she had feelings!

          Women with feelings are bitches, men with feelings are pansies. Women without feelings are elf-tarts and men without feelings are vulcan-cakes.

          FYI. That's geek dating slang in the big geek party scene. And, you're not part of the scene if you are cute and stupid... which is a 'tard-muffin.


          Geek Girl1: Ooh, check out that chem-student what a chiseled IQ... he's a total vulcan-cake.
          Geek Girl2: I scoped him already, he's dating a 'tard-muffin lit-major.
          Geek Girl1: So, like totally, illogical! Why are a

          • by Zarf ( 5735 )
            Uh, you guys know that the previous comment was a total farce right? It is either funny or a failed attempt at funny.
  • Objects and Nouns (Score:2, Interesting)

    by natedubbya ( 645990 )
    This area of research has been growing more popular lately. Last year's big language conference had a keynote speaker address the question of brain waves and word recognition. Most of the progress though is based on nouns because they have a core rooted meaning in everyone's basically visualize a generic version of that object in the world. You say hammer, I think of an actual hammer I've seen. It's not really mind reading because the approach falls apart when you start talking about verbs an
  • What about Pron? (Score:4, Interesting)

    by brunokummel ( 664267 ) on Monday June 02, 2008 @02:28PM (#23629533) Journal
    I know it sounds funny but i would like to see the brain activity for pornografic pictures, since it already known that "bad words" are stored in a different area of the brain than regular words... it would be kind of interesting if "bad images (or nice depending on the person)" got also stored on different areas....
  • by prakslash ( 681585 ) on Monday June 02, 2008 @02:30PM (#23629559)
    Yet again we see a jounalist dumbing down scientific research into tabloid fodder.

    What the CMU scientists have done is some preliminary brain imaging using MRI.

    Here is a better CMU link [] with more details and pictures. The scientists hope that this research to could have applications in the study of autism, disorders of thought such as paranoid schizophrenia, and semantic dementias such as Pick's disease. Not once did they ominously dub their research as "mind reading" as claimed by the submitter.

  • a Claymore trigger?
    • by Zarf ( 5735 ) a Claymore trigger?
      I don't think they have the pattern recognition for that.
  • It's too bad the article doesn't go into any detail about what they are measuring... its possible its reading alpha waves (normally linked to waking periods of relaxtaion or possibly drowsiness), beta waves (normal waking consciousness), gamma waves (perception and consciousness), or who knows, maybe they are monitoring the chemical reactions taking place. I guess its a possiblity that they don't want to let out the key to their study just yet.
    • by yumyum ( 168683 )
      The article said that the computer was using data from MRIs, and that through training it was able to discern patterns in the MRIs.
      • I guess i need to read up on MRI's a bit more, i always thought that they were used for imageing purposes; Such as detecting abnormalities, tumors, etc... without any of the harmful (possible outcomes) of say x-rays, or how CT scans are not as accurate ar determining differences extreamly small areas. I didn't think your brain would psyically change just because you were thinking one thing or another. Monitoring the brain wave activity would seem to be a more successful way to dtermine what people are think
        • Re: (Score:3, Informative)

          IANAD (but I watch a lot of House)

          > I didn't think your brain would psyically change just
          > because you were thinking one thing or another.

          Your brain doesn't, but the blood flow patterns do.

          Just like how your computer doesn't physically change when sitting idle or watching porn, it will use less/more power and different parts of different chips will flow more electrons in different patterns.

          Oh, and you need to google up "functional MRI"

        • Re: (Score:3, Informative)

          by mikael ( 484 )
          The latest technique in MRI is functional MRI (fMRI) . The doctors can watch the oxygen demand levels of the brain change dynamically as a person thinks. The resulting brain scan image superimposes the oxygen demand levels in red-yellow-green-blue scale over a monochrome image of that slice of the brain. Effectively, they see which areas of the brain are in use from second to second.

          In some cases, they have discovered that people in coma's or a persistive vegetable state have been discovered to have been aw
  • by eldavojohn ( 898314 ) * <> on Monday June 02, 2008 @02:50PM (#23629789) Journal
    It's strange, every time a researcher is assigned to go disassemble the prototype, something else comes up right when they come within range of the machine. Yesterday something kept spamming "REDRUM" across the networks broadcast address and causing bandwidth issues. Today several printers in the lab wouldn't stop printing out documents that looked like fake rebates for Newegg ...
  • by icebike ( 68054 ) on Monday June 02, 2008 @02:57PM (#23629881)
    Quoting article:
    "We believe we have identified a number of the basic building blocks that the brain uses to represent meaning. These building blocks could be used to predict patterns for any concrete noun..."

    The implications of building blocks would suggest that the french word for "Desk" (bureau) would elicit the same response as the english word for "Desk", instead of some governmental unit.

    That would be useful, (once we get cheap portable MRI hats).

    However I doubt these building blocks are anywhere near that generic due to the excess emotional baggage that people associate with words. I suppose it might be able to detect the presence of such baggage even if it could not decipher it.

    • Re: (Score:2, Insightful)

      by orateam ( 861461 )
      Of course this will eventually lead to computers to mapping and "determining" patterns that lead to criminal activities. Such as the mind of a pedafile or rapist. Reading your mind to see if you have "BAD THOUGHTS" capable of criminal activity, will lead to the government having the ability to read individuals for criminal minds and arresting for such thoughts and predicted activities.
      • by icebike ( 68054 )
        Ohhhh, riiiight. This will help: []
      • Of course this will eventually lead to computers to mapping and "determining" patterns that lead to criminal activities. Such as the mind of a pedafile or rapist. Reading your mind to see if you have "BAD THOUGHTS" capable of criminal activity, will lead to the government having the ability to read individuals for criminal minds and arresting for such thoughts and predicted activities.

        The fear of having my thought patterns available publicly/seizable by a gov't. entity/requestable by employers is one of the reasons I'd be hesitant to have a personal "mind reader" interface.I'd say it's more likely that those organizations will learn more about you from your ISP than what you associate with the word "tomato" though. At least until you rely on it too much.

    • by Roxton ( 73137 )
      What would be creepy about this mind reading technology is the possibility that it would get out of tune the more you change the way you think. This would encourage you to change back to your old ways of thinking. Essentially, it's a subtle mind-stasis device.

      And the personal responsibility people will be, like, "So what? You can suck it up and spend a week recalibrating if you want to. Who cares if most people don't? It's about choice, not consequences."

  • Phrenology was 19th century "science" of discerning personality types by look at fine detail of skull shape. This shares some aspects with 21st century brain scans:
    (1) They are both based on head geometry. Phrenology just looked at the surface, while MRI looks at volumes.
    (2) They are both derived from empirical measurements, rather than first-principles of why geometry is the way it is. This is not bad if it really works. Although more people would believe it if the underlying mechanisms for the geo
  • by rehtonAesoohC ( 954490 ) on Monday June 02, 2008 @03:07PM (#23629997) Journal
    I forsee some lonely nerd using a video-chat application to try and talk with a woman when all of a sudden, his computer reads his mind and says to him:

    "I'm reading that you're horny, Jim. Here is a selection of your favorite porn- Princess Leia doing an Ewok. Enjoy!"
    Prospective Girlfriend: "You sicko! *exits the video chat*"
    Jim: "Oh well... I guess I'll just enjoy this video. Thanks manputer!"
  • either:


    Derren Brown

    Chriss Angel


    Max Maven
  • Could this be one of those technologies that is already considered 'quaint' by the elite societies by the time the public discovers it?

    Just a thought...
  • So, if they make one that runs WinBlows and it crashes, do you lose all unsaved data? Could you require a reboot, or gasp, a reformat before you will recover from lost files in your NTFS partitions? Of course you would come back with a snazzy swap file and would have an epileptic seizure every time someones actually makes you think.
  • [don't take it seriously]
    If I wanted others to know what I am thinking. I would tell them.
    Plus I am posting in /. there are not many secrets as to what I am thinking.
    [/don't take it seriously]
  • How long before we start seeing this technology embedded in the metal detectors in airports?
  • Not SEX only SEX that SEX, it SEX is SEX possible SEX to SEX use SEX this SEX to SEX type. SEX isn't SEX that SEX wonderful?
  • noun verb preposition adjective noun!!
  • Pah! Just plug your controllers into the other slot.
  • The first military use will be in the MiG-31 Firefox, no?
  • Looks like the brain was programmed with microcode.

    Now the only question is, risc or cisc?

    And how many threads of conciousness in a sane person's mind?

Most people will listen to your unreasonable demands, if you'll consider their unacceptable offer.