Catch up on stories from the past week (and beyond) at the Slashdot story archive

 


Forgot your password?
Close
typodupeerror
×
Science

Face Recognition Needs 3 Areas Of Human Brain 151

Posted by CmdrTaco from the two-more-than-quake dept.
sushant_bhatia_progr writes "Nature has an article on the recent discovery that face recognition in humans targets 3 areas of the human brain. Using mugshots of celebrities, Pia Rotshtein at University College London and her colleagues have shown that there are at least three separate areas for processing and recognising faces. One processes the physical features of the face, one decides whether or not the face is known, and a third retrieves information about that person, such as their name. Rothstein's team used a computer to create a series of images in which the countenance of film star Marilyn Monroe gradually morphed into that of former British prime minister Margaret Thatcher, or that of James Bond actor Pierce Brosnan transformed into current prime minister Tony Blair."
This discussion has been archived. No new comments can be posted.

Face Recognition Needs 3 Areas Of Human Brain More

Face Recognition Needs 3 Areas Of Human Brain

Comments Filter:

  • Gotta love it... (Score:5, Funny)

    by StevenHenderson ( 806391 ) <stevehenderson@NOspam.gmail.com> on Tuesday December 14, 2004 @09:51AM (#11081192)
    Using mugshots of celebrities...

    Gotta love having enough celebs with mugshots to run an entire research experiment. :)

  • So how about... (Score:1, Funny)

    by Anonymous Coward
    transforming a picture of GWB into a Monkey?

    • Re:So how about... (Score:3, Funny)

      by Anonymous Coward
      I heard they tried one using Michael Jackson morphing into Diana Ross, but no-one could tell the difference.
    • This is actually very easy to do. Video morphing software is easy to come by, and if you find a picture such as the ones on bushorchim.com, in which the facial features of both subjects are lined up at least somewhat, it becomes trivially easy.

      I have done it BTW, but I don't have the MPEG anymore (there have been drive reformats between then and now.)

  • What a way (Score:2, Funny)

    by Locdonan ( 804414 )
    to kill off men and women's fantasies. Now all I see is Margret Thacher sing happy birthday, Mr. President.

    *shudder*

    I think I just inherited Wil Wheaton's sleeping disorder.

  • mugshots? (Score:1, Funny)

    by H8X55 ( 650339 )
    When was Blair or Thatcher arrested?

    I guess it's not just american politicians that are all crooks!








    it's a friggin' joke!

  • The horror section. (Score:5, Funny)

    by Anonymous Coward on Tuesday December 14, 2004 @09:55AM (#11081233)
    "Rothstein's team used a computer to create a series of images in which the countenance of film star Marilyn Monroe gradually morphed into that of former British prime minister Margaret Thatcher, or that of James Bond actor Pierce Brosnan transformed into current prime minister Tony Blair."

    And the fourth part of the brain. Recognizing the horror of it all.

  • Though all three don't have to be functioning... (Score:3, Informative)

    by Transcendent ( 204992 ) on Tuesday December 14, 2004 @09:56AM (#11081241)
    ...because I have a great memory for faces. I can almost always tell when/where I've seen a specific person...

    ...But I won't remember their name for the life of me...

    • Re:Though all three don't have to be functioning.. (Score:4, Interesting)

      by eMartin ( 210973 ) on Tuesday December 14, 2004 @10:05AM (#11081321)
      Same here, but I found a trick that helps.

      When you meet someone, and they or someone else tell you their name, repeat it back ("oh, I have a cousin named Jill" or "hmm, John's an unusual name"), and there's a very good chance you'll at least remember what you said later on.

      I do something similar with passwords. Normally, they're a jumble of letters and numbers from something around me when I needed to think of them, and usually I can remember what that thing was, so the password then pops into my head.
    • I can't remember names or faces.

      Although luckily I'm half-decent at recognising voices. Still, it's a bit embarrassing to almost walk past a friend without noticing them, and to not be able to recall their name when they do say hello... :-)
      • You think that's embarassing?

        I once met a girl when I was drunk, and when I went to meet her and a friend of hers the next day, I wasn't sure who was who at first.
      • Same here. I have never been in a class at school of which I knew all names/faces even at the end of the year. If I meet people I work with every day at other places then at work, chances are big I won't recognise them. Sometimes even at work I have to think for a minute until I remember a collegue's name. If someone unmasked would rob me by clear daylight I would literally not be able to pick that person from a line-up even five minutes later if my life depended on it.
        It is no fun, but I'm glad to see here

  • TFA (Score:3, Informative)

    by JollyRogerX ( 749524 ) on Tuesday December 14, 2004 @09:56AM (#11081243)

    Celebrity shots probe face recognition

    Helen Pearson

    The brain uses three steps to identify faces.

    The features in this set of images change gradually, yet our brains flip suddenly from seeing Margaret Thatcher to seeing Marilyn Monroe. © Dr Jenny Gimpel/University College London By transforming the features of Margaret Thatcher into those of Marilyn Monroe, researchers have revealed hints about how our brains put a name to a face.

    Neuroscientists already know that certain spots in the brain play a vital role for recognizing a familiar face, even as it changes with age or a new hairstyle. But they have not been clear precisely what each area does.

    Using mugshots of celebrities, Pia Rotshtein at University College London and her colleagues have shown that there are at least three separate areas for processing and recognising faces. One processes the physical features of the face, one decides whether or not the face is known, and a third retrieves information about that person, such as their name.

    Rothstein's team used a computer to create a series of images in which the countenance of film star Marilyn Monroe gradually morphed into that of former British prime minister Margaret Thatcher, or that of James Bond actor Pierce Brosnan transformed into current prime minister Tony Blair.

    Although the physical features gradually change from one face into another, the researchers showed that subjects looking at the images tend to "suddenly flip" from seeing Marilyn to seeing Maggie, explains team member Jon Driver.

    The researchers then showed their subjects three different pairs of images from the array while they were in a functional magnetic resonance imaging (fMRI) brain scanner. The two pictures in one pair were identical; in another pair they had different physical characteristics but were both still recognizable as Maggie; and in the other pair they differed by the same degree in their physical characteristics, yet one was still recognizable as Maggie and the other as Marilyn.

    The study allowed the team to pick out the three areas of the brain that carry out different tasks when someone walks into a room. The first region, a pair of structures at the back of the brain called the inferior occipital gyri, was most active when the physical features, such as eyes and hair, in the two pictures differed. It appears to analyse these physical characteristics.

    A second region, the right fusiform gyrus, located just behind the ears, was most active when one picture showed Maggie and one showed Marilyn. This region appears to distinguish between faces, perhaps by comparing the face to known ones.

    A third area, the anterior temporal cortex, appears to store knowledge connected to the faces. This region was most active when people knew the famous subjects particularly well; less so in those who, for example, were less familiar with the British politicians.

    The study is the first to clearly show these three separate stages of face processing, says psychologist Isabel Gauthier, who studies face and object recognition at Vanderbilt University in Nashville, Tennessee.

    Driver says he now wants to study patients who, through injury or disease, have particular problems recognising people. Some people with prosopagnosia, or face-blindness, may be unable to recognise faces as familiar as their own children. Patients with dementia may struggle to put a name to a household face.

    Driver wants to examine whether he can match up patients' specific problems to different defects within the brain regions identified by the team. He also wants to find out whether some patients could be trained to revamp these failing regions.

    • Re:TFA (Score:3, Interesting)

      by NovaScotian ( 547402 )
      Some years ago (I've long lost the reference) a PhD student in Rhode Island digitized human faces with 256 points and then projected these points through the same points on an androngenous composite of hundreds of faces next to the sample; an "average" face. At some point in a plane beyond the reference face, the points at the ends of the projectors were then re-plotted and joined to form a caracature of the amplified differences between the sample face and the sexless "norm". She showed that her subjects

  • Or maybe... (Score:4, Funny)

    by Zangief ( 461457 ) on Tuesday December 14, 2004 @09:56AM (#11081244) Homepage Journal
    We need 3 parts of the brain to recognize celebrities.

    -One to recognize the face and map it to its info.
    -One to categorize the info as hot girl or not.
    -One to ignore the not-hot-girls.
    • Rothstein's team used a computer to create a series of images in which the countenance of film star Marilyn Monroe gradually morphed into that of former British prime minister Margaret Thatcher


      I bet you'd have some trouble classifying whether it's a hot girl's face or not halfway through the morph!

  • Scary... (Score:4, Funny)

    by Martin Spamer ( 244245 ) on Tuesday December 14, 2004 @09:58AM (#11081255) Homepage Journal

    Marilyn Monroe gradually morphed into that of former British prime minister Margaret Thatcher

    Whoever though of that is one sick scary F***er!!!
  • "One processes the physical features of the face"
    Would that be called the sense of sight, perhaps?

    "one decides whether or not the face is known"
    And this one seems to be visual memory.

    "a third retrieves information about that person, such as their name." And this one we typically call ordinary memory.

    I can't say I know what I'm talking about, but this seems kind of obvious. It sounds like they're saying, "well you see the face, recognize it, and identify it."

  • I dunno... (Score:5, Funny)

    by zenmojodaddy ( 754377 ) on Tuesday December 14, 2004 @10:00AM (#11081274)
    ... it's pretty clear that Tony Blair has been morphing into Thatcher for years.

    Shame he doesn't have her balls, though.
    • So true!
    • I dunno. Thatch was anti-union, Blair (while moving away from labour (union) values) seems to weathercock in whatever direction the current middle class pinko populist opinion is (while being consistantly and secretly evil in a totalitarian way). Thatch was more WYSIWYG, where Blair is more hypocritical.
    • Nah. Not a bit.

      Thatcher was her own person; you knew what she believed in, and what she wanted (whether you agreed with her or not). You knew where you were with her -- you might have hated her and everything she did, but you knew where you were with her. She had that integrity, at least.

      Whereas who knows what Blair believes (if anything)? The only thing he seems to believe in is power. You get the impression he'd say anything at all to keep it. A will of his own? A plan? Integrity? Blair crave

  • Re: (Score:2, Interesting)

    by account_deleted ( 4530225 )
    Comment removed based on user account deletion

  • Brittany or Jessica (Score:4, Funny)

    by GillBates0 ( 664202 ) on Tuesday December 14, 2004 @10:01AM (#11081290) Homepage Journal
    And all the 3 areas of my brain still can't figure out if that's Brittany on Jessica Simpson lip synching on TV.

    Sometimes, "context" can be more telling than just the face. Brittany's are way bigger, IMHO.

  • Classic fMRI experiment (Score:5, Interesting)

    by Hug Life ( 643998 ) on Tuesday December 14, 2004 @10:06AM (#11081328)
    IAAfA (I am an fMRI analyst) Of course, the last highly publicized study that gave us a "face recognition area" of the brain turned out to be a crock. The same haemodynamic response came from birdwatchers seeing birds, or car experts seeing cars. It was a cognitive recognition area, not just a "face recognition area". I wouldn't be suprized if this experiment had the same falicies (article wasn't very precise).
    Modularization: Great for OO programming, crappy for the human brain.
    • What was the response from bird watchers seeing cars?
      Or for that matter Teachers finding spelling mistakes or programmers finding that someones used a non-iso date format.

      As someone who's been programming for about 20ish years programming has become more of an visual/emotional response than something I think about, just like looking at a picture of someone you know. I should imagine that this kind of 'instinct' applies to most people with most tasks that they do frequently and is not 'pre-programmed'.
    • Hug Life, email me, would you? dmd at 3e dot org. I do fMRI at UPenn, and would love to talk to another /.-reading fmri person.
    • Modularization: Great for OO programming, crappy for the human brain.

      IAWMUHTIPORI (I am writing my undergraduate honors thesis in philosophy on related issues) What sort of "modularization" are you referring to? Modularization of peripheral systems (input/output systems, i.e., the senses)? If so, you must realize that you would be in the extreme minority in opposing a modular architecture for these systems (see Jerry Fodor's Modularity of Mind [amazon.com], the standard treatment on peripheral systems modularity wi
      • Fodor's brand of modularity, debates about what PDP can and cannot do -- these have been passe for about 15 years in non-philosphical areas of cognitive science. (And I say this with moist eyes as I am of the older, symbolic tradition.)
        • I am curious as to what you think the conclusions of these debates in the "non-philosophical areas of cognitive science" have been. I could cite numerious articles that have come from people in cognitive science outside of philosophy in the past five years defending positions all across the board, from massive modularity to distributed connectionism and everything in between. Just look at the article from BBS that I mentioned above, particularly the replies, the response to the replies, and the associated c
          • [What follows is one man's quick summary of the history of Fodor's modularity thesis and much else.]

            I would distingush between the modularity of Fodor (1983) and the neuroscience notion of localism.

            Fodor's was a specific thesis about the modularity of sensory/perceptual systems, but also more "central" systems, such as the parsing module. It made specific claims about what it means to be a module, including information encapsulation and penetrability. These were strong claims, which is of course a good th
    • Pardon, but how does that make the research a crock? These areas are associated with expert-level visual analysis of stimuli, of which face recognition is the most prominent example.
      • It's extremely likely we have specialised circuits for processing facial expressions since:
        1. Our ancestors needed it to be able to perceive threat & attraction in peers.
        2. If you see a photo with reversed eyes or mouth, it's recognisable but gives you a very weird feeling.

        There's also a circuit that lets us know when we recognise faces, because the unfortunates without it have been diagnosed with Capgras' Syndrome [23nlpeople.com].
    • I also do fMRI.

      Your critique is too strong. It's true that some have found activations in the "fusiform face area" in reponse to other kinds of visual expertise, but that doesn't mean it isn't involved in face perception. There's good evidence that it plays a role in determining facial identity. I've seen my own fusiform lighting up in reponse to faces but not other objects.

      The results in the story article are not new really, although it is nice to have it all together in one experiment.
      • I also do fMRI.

        Your critique is too strong.

        Agreed.

        It's true that some have found activations in the "fusiform face area" in reponse to other kinds of visual expertise, but that doesn't mean it isn't involved in face perception. There's good evidence that it plays a role in determining facial identity. I've seen my own fusiform lighting up in reponse to faces but not other objects.

        The FFA, if I remember the work of Gauthier, Tarr, and others correctly, is better thought of as the site of visual sha
        • The FFA, if I remember the work of Gauthier, Tarr, and others correctly, is better thought of as the site of visual shape knowledge. Of course, faces are one class of shapes for which we are all experts, and that is why the FFA activates when normal college sophmores participate in face recognition experiments. But you also get FFA activation if you train people to be experts on novel shape classes (e.g., "greebles").

          Yep, but it's still an open question I think. For example, in a recent Nature Neurosci
          • For example, in a recent Nature Neuroscience article Grill-Spector Knouf & Kanwisher found FFA to be face specific, even in car-experts looking at cars.

            Ah, yes. I should have named Kanwisher as well.
    • Of course, the last highly publicized study that gave us a "face recognition area" of the brain turned out to be a crock.

      Yes, the FFA is more than a "face recognition area." But calling that hypothesis a "crock" is too strong. It was a scientific hypothesis warranted by the initial data and provocative enough to bring better experiments. You make its sound like a lie deliberately foisted on the scientific community. The data from these newer experiments have falsified the original hypothesis. Newer, more
    • I see another problem with this experiment. Won't the results be different if it's a person no one knows? What if it were a picture of an acquaintance?

  • "...film star Marilyn Monroe gradually morphed into that of former British prime minister Margaret Thatcher."

    Or as it's known in medicine, 'the anti-Viagra.'

  • Intel and AMD today announced plans to market Triple Core CPUs specifically for the facial recognition industry.

    Security officials today claimed the improvements these triple core chips will bring may actually make their airport scanning devices useful.
  • I know this is outlandish, but I wonder if there'll be a drug that enhances the temporal cortex (I know it's a huge generalization for neuroscience), which according to this article is related to familiarity of the subject being observed. It might one day be a treatment for racism. People generalize and say "All people look the same" but if we were more easily able to recognize the individual instead of merely the racial traits, it could be one more evil that humanity can overcome through science.

    I woul

    • interesting thought experiment; bad practice (Score:4, Insightful)

      by Doc Ruby ( 173196 ) on Tuesday December 14, 2004 @10:40AM (#11081635) Homepage Journal
      The problem with such a racist is not their thoughtcrime, failing to recognize racial differences, but their actions. If they can't (or won't) notice differences among individuals of other races, that's they're problem. When they burn these people's houses down, beat them in nightclubs, refuse to hire them, or do other bad things, it doesn't really matter that their facial recognition is wired wrong.

      When we make thoughts illegal, we're faced with legislating people's minds. Not only politically catastrophic in a free society, but probably medically irresponsible to pretend we are in control of all the results. We have a flawed, but much more successful, history of managing behavior. We should stick to what we know until we've improved it to adequacy, before messing with minds and all the worse consequences at stake.

      • When we make thoughts illegal, we're faced with legislating people's minds.

        Where I come from, "legislating people's minds" is called Titles 17 [copyright.gov] and 35 [gpo.gov], United States Code.

        • Although copyright and patent protection is now de rigeur for intangibles like tambres, business processes and mathematics, the copies stored and transient in the mind are not yet prohibited. Thinking and talking about protected creations is OK. As I articulated in my post, the law is properly involved only when doing something that would damage another person or their property, however intangible that property might be.
  • There's a disorder called prosopagnosia (face-blindness) Where the afflicted cannot identify people's faces. Here's a couple of links to pages written by people who have it. It lends credence to the theory that there are entire portions of the brain dedicated to the recognition of faces.

    http://www.prosopagnosia.com/
    http://home.eart h link.net/~blankface/prosopagnosia .shtml

    Here's the google search
    http://www.google.com/search?hl=en&c2coff= 1&q=pros opagnosia&spell=1
    • I once took a huge dose of special k, and the only major visual hallucination I experienced was not being able to recognise people's faces. I had to look out for my friends based on what they were wearing (it was horrible).

      Anyway, does anybody know if special k affects the same parts of the brain mentioned in TFA?

      saxwell
      • The only Special K that I've heard of is a breakfast cereal in the US. Presumably, the Special K you mention is an informal name for some form of medication/intoxicant.

        Either that, or there's something Kellogs needs to come clean about.
    • Similarly sort of on topic, I'm wondering if this research toward facial recognition will aid any of the ongoing Aspberger's Syndrome and Autism research. For Aspie's, it's not so much the recognition of the face that's the problem as the information the face is conveying (i.e. happiness, sadness, etc.). This could contribute toward that end of the research.

  • Not to be picky but (Score:2, Funny)

    by Anonymous Coward
    The title implies these areas are needed, but really, only that activity is found in these three areas. To show need, they would have to ablate these areas and block recognition (and even that could have some problems). They show sufficiency at best.

    I imagine you could do this in chimps with chimp celebrities, but outside of GW, we may not know who's who of chimp celebrities.

    I didn't RTFA, but this is just a thought.

  • Who needs a fancy computer? (Score:4, Insightful)

    by dkleinsc ( 563838 ) on Tuesday December 14, 2004 @10:47AM (#11081690) Homepage
    the countenance of film star Marilyn Monroe gradually morphed into that of former British prime minister Margaret Thatcher

    You don't need computers for that. You just need to wake up next to someone you don't remember meeting.

    For more information on the subject, listen to the song "9 Coronas".
  • Like all those people with head injuries, syphilis and other problems didn't tell us this years ago.

    Try taking acid, it's a lot cheaper than MRI &co, and will point you in the right direction.

  • Although the physical features gradually change from one face into another, the researchers showed that subjects looking at the images tend to "suddenly flip" from seeing Marilyn to seeing Maggie, explains team member Jon Driver.

    What other possibility was there? Since our circuitry is made to identify the faces, it's not exactly trained to say "60% that and 40% this."

    In related news, subjects were found to "suddenly flip" between saying No and Yes when asked "Did you have enough of that?"

    • What other possibility was there? Since our circuitry is made to identify the faces, it's not exactly trained to say "60% that and 40% this."

      I'm not sure you or anyone else is qualified to make comments about what our "circuitry" is made for.

      Nonetheless, connectionist models suggest there are different neural activation patterns which encode Monroe and Thatcher. Contrary to your statement, a given image may indeed activate 60% of the Monroe network, and 40% of the Thatcher network. These activations

      • I think that, while we're not sure about the ultimate purpose of the circuitry, we have a pretty good idea on how we use face recognition behaviorally, for some milions of years now :-)

        As you correctly say, the image may activate 60% and 40% of corresponding networks. My point was that, behaviorally, subjects were not likely to tell "60% Monroe + 40% Thatcher", no matter what percentage of which netwroks were activated. Simply because the behavior of face recognition is used to identify the person, that

      • Correction: I saw it first in 1996, and here is the link [lib.ru]. It's all in Russian, pictures are bad quality, and of the then-presidential-candidates. However, it makes pretty much the same point :-)

  • Don't you need eyes too?
  • My opinion is that without huge mathematical skils this project will sink in oblivion right from now. I don't know waht they're doing or what they can do, but the truth is that without math all serious projects remains on the begin stage or become in work to sci-fi books. The history had proven it.
  • a friend of mine told me he thought it folly to model computational devices after the structure of the human brain.

    I told him how interesting it was that the brain had short term memory and long term memory just like a computer has ram and hard drives- and how we have eyes which are like a video card and so on...

    he told me it was dumb to compare computers to humans in any sense - being that the processes used by the brain vs computer were so vastly different that any comparison was rendered invalid merely
    • Computers are in many ways complementary to humans, not comparable. A video card is part of an image output system, while our eyes are an image input system. Similarly, the keyboard and mouse don't replace or replicate the functions of our hands - they are what we use with our hands.

      The memory structure and processing paradigm in most computers is also quite different from what we know of the human thought process. The computer uses binary logic to carry out its operations, while the brain, as far as we k

  • Not only has the recognition and category task been done in primates ( http://www.sciencemag.org/cgi/content/abstract/29 1/5502/312 [sciencemag.org]) but all three of these areas have been found in previous fMRI experiments.

    I actually find these results extremely misleading -- there is no way that these three processes occur in complete isolation across these three areas. The recognition and recall task has been shown to rely on hippocampal regions (through lesion studies). A correlative finding is very weak.
  • ..film star Marilyn Monroe gradually morphed into that of former British prime minister Margaret Thatcher, or that of James Bond actor Pierce Brosnan transformed into current prime minister Tony Blair.

    Although the physical features gradually change from one face into another, the researchers showed that subjects looking at the images tend to be "suddenly turned off" from seeing Marilyn turn into Maggie, explains team member Jon Driver.

    "I've never seen an erection go flacid so quickly" explains team member
  • UCL? hey i go.. somewhere near there thats not quite as good and doesnt have a news article :(
  • http://news.bbc.co.uk/1/hi/health/4086319.stm [bbc.co.uk]
    Links to a video of it there, and a few possible technologies
  • Breast recognition requires no brain activity...
  • The Independent [independent.co.uk] yesterday ran an article on a stroke victim who could still perceive facial emotions, even though clinically blind.

    His eyes and nerves were fine, but the visual processing part of his brain had been killed. So signals were coming through, just not ones that you and I associate with sight.

  • Copied from here [nature.com]

    Morphing Marilyn into Maggie dissociates physical and identity face representations in the brain
    Pia Rotshtein, Richard N A Henson, Alessandro Treves, Jon Driver, & Raymond J Dolan

    How the brain represents different aspects of faces remains controversial. Here we presented subjects with stimuli drawn from morph continua between pairs of famous faces. In the paired presentations, a second face could be identical to the first, could share perceived identity but differ physically (30% al
  • I think what they actually have here is proof the recognizing pictures of people you don't actually know in real life requires 3 areas of the brain. Are there any comparative studies that contrast celebritity photos vs. photos of people the subject knows in real life? Or photos of people vs. people who are actually, physically present?

Slashdot Top Deals

Ocean: A body of water occupying about two-thirds of a world made for man -- who has no gills. -- Ambrose Bierce

Close