If Someone Secretly Controlled What You Say, Would Anyone Notice?

BY GREG MILLER

09.29.14 | 6:30 AM

http://www.wired.com/2014/09/cyranoid-experiment

The subject enters a room in which a 12-year-old boy is seated. A 20-minute conversation ensues. The subject quizzes the boy about current events and other topics to get a sense of his intelligence and personality. But the boy is not what he appears to be.

Unbeknownst to the subject, the boy is wearing a radio receiver in his ear, and every word he says is transmitted to him by a 37-year-old university professor sitting in a nearby room. For his age, the boy has surprisingly well-informed opinions about the effects of austerity measures on the European economy. He speaks of his admiration of Dostoyevsky. Yet not a single subject suspects that his words are not his own.

The study, conducted by two social psychologists at the London School of Economics and Political Science, and published earlier this month in The Journal of Social Psychology, raises some fascinating psychological and philosophical questions, and the researchers hope it will open new directions of study.

“Beyond the physical, we like to believe that there’s some element in all of us that’s a permanent part of our nature,” said co-author Kevin Corti. We like to think we can recognize that element in other people, and they can recognize it in us. But these findings suggest we are easily fooled. In future research, Corti and his co-author, Alex Gillespie, plan to repeat the experiment with people who already know each other.

“If you were forced to have a conversation with your spouse or your boss or your best friend through the body of another person, how would that change the interaction?” Corti asks.

And, here’s an unsettling thought: could they even tell it was you?

Although Corti and Gillespie are the first to publish a scientific paper on this effect, the idea came from Stanley Milgram, the psychologist best known for his infamous obedience experiments, in which people delivered what they thought were painful electric shocks to a person in another room after being instructed to do so by an authoritative figure in a lab coat. (In reality, the person in the other room was an actor who pretended to get shocked; even so, the experiment would never pass a modern ethical review). Toward the end of his career in 1984, Milgram began a talk to a meeting of the American Psychological Association in Toronto with a strange and somewhat ominous-sounding announcement.

“Since 1977 I have been conducting research on cyranoids,” Milgram said. He quickly explained: “Cyranoids are people who do not speak thoughts originating in their own central nervous system: Rather, the words that they speak originate in the mind of another person who transmits these words to the cyranoid by means of a radio transmitter.” The term was inspired by the French play Cyrano de Bergerac, in which a brilliant but ugly man woos his beloved through love letters signed with the name of a handsome nobleman.

What drew Milgram to this line of research isn’t entirely clear. Gillespie suspects he may have seen it as a followup to his obedience experiments. The people who continued to deliver shocks essentially ceded control of their own behavior to the guy in the lab coat, he says. “They disengaged the moral part of their brain somehow and become cyranoids,” Gillespie said. “I suspect that’s how he came up with it, but that’s just speculation.”

Milgram also may have been inspired by fantasy and science fiction, Corti suggests. In his speech to the APA, Milgram alluded to the man behind the curtain in The Wizard of Oz and asked the audience to imagine a world in which the thoughts of one person come out of another person’s mouth. “What would be the main sociopsychological consequences of such a world?” Milgram asked. (If the world depicted in the 2009 movie Surrogates is any indication, the consequences would not be good).

In 1979, Milgram asked the National Science Foundation for $200,000 to investigate the cyranoid phenomenon, but they rejected his grant application, and he never published any scientific papers on the topic. Cyranoids became a largely forgotten part of the legacy of one of the 20th century’s most famous psychologists.

Artist Robb Mitchell plays an obvious cyranoid at an art gallery in Dundee, Scotland in 2007.  Robb Mitchell

But in recent years, several performance artists have revived the concept. “I thought it would be fun to test this idea of cyranoids in the wild,” said Robb Mitchell, an artist and professor of social interaction design at the University of Southern Denmark. In 2007, Mitchell organized an exhibit at an art gallery in Dundee, Scotland. He wore a silly hat adorned with a camera and other gadgets as he wandered through the gallery talking to visitors. “We tried to make it really obvious and not too scary for people,” he said. Mitchell wore a radio receiver in his ear, and visitors could take turns controlling his speech from a nearby room.

Even people who knew him didn’t catch on, despite the fact that he acted like he’d never met them and offered odd responses to their questions. “That was a huge surprise,” Mitchell said. “They couldn’t get past the idea they were talking to me.”

In an even more bizarre exhibit, artists posted flyers around South Bank in London proclaiming that a unicorn would appear at a certain time. Nobody actually thought a unicorn was going to appear, but people gathered around anyway to see what was up, says Gillespie, who helped organize the performance. At the designated time, a group of about 10 people started speaking in unison: “We are the unicorn.” Unbeknownst to the crowd, they were cyranoids under the control of one of the artists. People in the audience initially thought it was all scripted, but the cyranoids answered in unison to spontaneous questions posed by people in the crowd.

“People started getting really freaked out,” Gillespie said.

For Corti, cyranoids seemed like a promising tool for studying human social interactions. His research interests include how people function within organizations, and in particular how someone’s outward appearance affects their influence within the organization. Cyranoids are an excellent way to investigate whether it’s what someone says or what they look like that matters more, he says.

The new study is just an initial step toward this kind of research. He and Gillespie just tried to replicate the experiments Milgram described in his talk at the APA meeting. “We had no idea if people were going to fall for this illusion, or whether there was something unique about what Milgram did,” Corti said.

Their findings suggest people fall for it pretty hard. The cyranoid illusion worked just as well when the 12-year-old boy and the professor, played by Gillespie, switched roles. Subjects thought the man seemed a bit dim for an adult living in Britain—he botched a question about Margaret Thatcher and was unable to list the country’s most recent prime ministers—but they gave no indication they suspected his answers weren’t his own.

In another version of the experiment, subjects interacted with a male grad student through the body of a female, or vice versa. Again, nobody caught on.

On one hand, maybe that’s not so surprising. Our brains didn’t evolve to deal with people speaking through the body of someone else, notes Jeremy Bailenson, who directs the Virtual Human Interaction Lab at Stanford. “Our brains are wired to treat something that looks and acts like a person as an individual person.”

But that’s increasingly not the case, Bailenson says. “What’s changed since times of Milgram is that this identity replacement has become the norm for online interactions.” From online games to online dating sites, people act through virtual versions of themselves (or assumed virtual identities) more and more.

This is another area Corti and Gillespie want to explore in future cyranoid research. One experiment, for example, might look at whether people can tell when the person in front of them is being fed lines from a chatbot. It’s a twist on the Turing test that even the dubious Eugene Goostman might be able to pass.

If you want to get really deep about it, we are all cyranoids, Corti says. “We all say things we hear other people say,” he said. “Jon Stewart is just an amalgam of ghost writers. The stuff I told you earlier about Milgram is based on what other people said at a conference. Where in all this are original thoughts?”

Around this point in our interview, something strange happened. The connection got a bit janky and Corti’s speech became garbled and almost robot-like. As I strained to hear what he was saying, a paranoid thought crossed my mind: Who was I actually talking to? I hung up and called again. “I’m back,” Corti said. And as far as I know, he was.