Reading Faces

Is that a scowl or just disgust? Facial expressions can be harder to interpret than most of us realize, but help is on the way. Read on

  • By Richard Coniff
  • Smithsonian magazine, January 2004, Subscribe
 

 

Article Tools

 
  • Comments (2)
  • Font
    		•
  • Email
    		•
  • RSS
    		•
  • Print
    		•

    Since facial expressions are part of our biological heritage, shouldn't reading them be second nature? In fact, most of us are dismal at reading faces, particularly those of strangers. Since September 11, it has dawned on much of the world that looking at other people's faces, in airports, in crowded subway cars and elsewhere, is something we need to do. And psychologist Paul Ekman, 69, is the man to teach us how.

    Until Ekman came along, no one had systematically analyzed or measured the emotions that pass across our faces. But in the 1970s, Ekman and a colleague at the University of California at San Francisco developed the Facial Action Coding System, or FACS, which has since become the essential tool in the science of recognizing and interpreting facial expressions. The FACS atlas, now on CD ROM, describes all 43 movements, or "action units," facial muscles can perform, plus all the combinations of action units which can create more than 10,000 possible facial expressions in all.

    Ekman's work has created a weirdly diverse following. The Dalai Lama has helped finance Ekman's classes on developing "emotional balance." Federal counterintelligence agencies routinely hire him to teach the nuances of facial expression for use in interrogating suspected Al Qaeda terrorists. Ekman's FACS atlas has also enabled Toy Story's Buzz Lightyear, and a generation of cartoon characters since then, to arch their eyebrows and otherwise make faces like real people.

    Perhaps most importantly, because FACS breaks facial expressions down into all of their component parts, researchers are finding that computers can also learn to read faces. So the thinking goes, if we could just link airport security cameras to computers programmed with FACS, maybe we could spot would be hijackers by a telltale glower or a microexpression of contempt.

    Probably what's most useful about Ekman's work is that he reminds us that reading faces is an ability latent within us all—never lost, merely forgotten. No government security program, nor any computer network ever likely to be conceived, could possibly match the effectiveness of millions of ordinary people simply making conversation and looking one another in the face.


    Since facial expressions are part of our biological heritage, shouldn't reading them be second nature? In fact, most of us are dismal at reading faces, particularly those of strangers. Since September 11, it has dawned on much of the world that looking at other people's faces, in airports, in crowded subway cars and elsewhere, is something we need to do. And psychologist Paul Ekman, 69, is the man to teach us how.

    Until Ekman came along, no one had systematically analyzed or measured the emotions that pass across our faces. But in the 1970s, Ekman and a colleague at the University of California at San Francisco developed the Facial Action Coding System, or FACS, which has since become the essential tool in the science of recognizing and interpreting facial expressions. The FACS atlas, now on CD ROM, describes all 43 movements, or "action units," facial muscles can perform, plus all the combinations of action units which can create more than 10,000 possible facial expressions in all.

    Ekman's work has created a weirdly diverse following. The Dalai Lama has helped finance Ekman's classes on developing "emotional balance." Federal counterintelligence agencies routinely hire him to teach the nuances of facial expression for use in interrogating suspected Al Qaeda terrorists. Ekman's FACS atlas has also enabled Toy Story's Buzz Lightyear, and a generation of cartoon characters since then, to arch their eyebrows and otherwise make faces like real people.

    Perhaps most importantly, because FACS breaks facial expressions down into all of their component parts, researchers are finding that computers can also learn to read faces. So the thinking goes, if we could just link airport security cameras to computers programmed with FACS, maybe we could spot would be hijackers by a telltale glower or a microexpression of contempt.

    Probably what's most useful about Ekman's work is that he reminds us that reading faces is an ability latent within us all—never lost, merely forgotten. No government security program, nor any computer network ever likely to be conceived, could possibly match the effectiveness of millions of ordinary people simply making conversation and looking one another in the face.

        Subscribe now for more of Smithsonian's coverage on history, science and nature.



    Digg
     
    Comments (2)

    can somebody show me the facial expression for "hey i saw that on Walker, Texas ranger!!"?

    Chuck's face was saying: "I might look a bit slow, but I'll kick your az!!"

    hahaha!!

    Posted by george on August 22,2011 | 04:15 PM

    i work with seiously disturbed teens and am very interested in learning about this work as i feel i would be very good at it and it would help help me

    Posted by debra rhead on August 23,2009 | 01:14 AM

    Post a Comment


    Name: (required)

    Email: (required)

    Comment:

    Comments are moderated, and will not appear until Smithsonian.com has approved them. Smithsonian reserves the right not to post any comments that are unlawful, threatening, offensive, defamatory, invasive of a person's privacy, inappropriate, confidential or proprietary, political messages, product endorsements, or other content that might otherwise violate any laws or policies.



    Advertisement


    Advertisement

    Follow Us






    First Name
    Last Name
    Address 1
    Address 2
    City
    State   Zip
    Email

    Smithsonian Store

    Smithsonian Journeys