The dungeon is pitch black—until the dungeon master blazes a torch, confirming your worst fears. A Beholder monster lurches at you, its eyeballs wriggling on tentacular stems. As you prepare to wield your Vorpal sword, where do you focus your gaze: at the monster's head or at its tentacle eyes? Such a quandary from the role-playing game Dungeons & Dragons may seem like a meaningless trifle, but it holds within it the answer to a scientific question. In fact, a father-son team has used images of such monsters to show that most people will look to another creature's eyes, no matter where they are located on the body.
"Dungeons & Dragons monsters have eyes all over the place," says Julian Levy, a ninth grader at Lord Byng Secondary School in Vancouver, Canada. Two years ago, Levy's knowledge of the role-playing game led him to a unique solution for solving a basic scientific question: Do people focus their gaze on another person's eyes or on the center of the head, where the eyes just happen to be located?
"We were eating dinner and my dad was talking about how, after publishing a paper about gaze tracking, a reviewer said that you could never prove whether people are looking at the eyes or the center of the face," Levy recalls. So he piped up with an idea, offering Dungeons & Dragons characters as an experimental solution. Because many characters have eyes located on their hands, torso, or other areas of the body, a researcher could track viewers' gazes to see what part of the characters they focus on first.
Levy's father, cognitive scientist Alan Kingstone of the University of British Columbia, Vancouver, loved it. The father-son team got to work, with Kingstone recruiting university students for the experiment and Levy combing the Web for the best examples of D&D beings. He selected 36 photos of Dungeon & Dragon humans, humanoids (nonhumans that still have eyes in the middle of their faces), and monsters (creatures with eyes positioned elsewhere). Levy set up eye-tracker equipment called Eyelink 1000 for 22 student participants, who viewed each of the character photos for 5 seconds.
Kingstone and co-author Tom Foulsham at the University of Essex in Colchester, U.K., analyzed the eye-tracking data, as they report today in Biology Letters. They found that participants first tended to look at the middle of the image, but then tended to fixate on the eyes, regardless of whether the eyes were on the head or elsewhere. "This paper makes the point explicitly that no, these brain areas are really interested in processing the eyes, not the center of the head," Kingstone says. The human brain's preference for eyes may have evolved as a way for people to communicate quickly and quietly and to convey simple information about a person's age, health, and emotions, he hypothesizes.
"At first blush, these sorts of reactions can seem trivial: Okay, so we're slightly more likely to look at the eye region, big deal," says Stephen Shepherd, a neurobiologist at Rockefeller University in New York City who was not involved in the research. "But these mechanisms are likely foundational to behaviors like eye contact and gaze following, which humans and other primates use to threaten one another, to flirt, and to share experiences and attitudes," he says.
In addition to answering questions in basic biology, the study's findings may prove useful for children with autism, who often struggle in making eye contact with others. Their therapy includes training that teaches them that skill. Now, researchers may be able to apply the new investigative technique as a first step for clarifying whether children with autism seek out the eyes or whether they focus solely on the head.
However, the study used only two-dimensional images that do not gaze back at the viewer, whereas real-world eye contact is "a much more sophisticated dance," Kingstone notes. "Because there's just so much more going on with the eyes in real life, this would never cut it for teaching natural-looking behavior."
Though his paper is being published just in time for Halloween, Levy says he decided to dress up as a zombie hunter rather than a many-eyeballed monster. "I think my dad might be willing to dress up as a D&D monster, though, and scare the kids who come to the door," he says.
But Kingstone quickly corrects him. "No, that's not true—sorry."