Policemen, psychologists, and poker players can all tell you the best ways to spot a liar. But detecting lies and cons over email is a relatively new challenge. A recent study looked into it.
The signs of electronic fibbing. I'm Bob Hirshon and this is Science Update.
How can you spot a liar online? Jeff Hancock, a professor of communication at Cornell University is trying to find out. He and his colleagues analyzed transcripts from email conversations.
They found that compared to truth-tellers, liars used more pronouns, like he, she, and they; more sense words, like see, hear, feel, and touch; and a third more words overall.
Which is somewhat surprising because you might think that people who are lying are trying to avoid being caught. But it makes sense in another respect in that the liar may be trying to enhance the story, make it sound more credible.
Interestingly, he says that people being lied to tended to ask more questions—even when they didn't suspect anything. It's hoped that the research will lead to computer software that can spot online con artists and other predators. I'm Bob Hirshon for AAAS, the Science Society.
Making Sense of the Research
In order to learn how to spot a liar, you first have to learn whether, and how, people behave differently when they're lying. That's what Hancock and his team set out to do.
They assigned volunteers to participate in email conversations about themselves, focusing on five different specific topics. One volunteer from each pair was asked to lie about two of the topics, and given a few minutes to prepare and examples of the kinds of lies to tell. By designing his experiment in this way, Hancock could compare non-deceptive and deceptive conversations between the same two people.
The results were somewhat surprising, since other studies have found that liars tend to use fewer words, not more. However, those studies looked at "deceptive monologues"—in other words, long stories that only one person told. Under these circumstances, it makes sense that a liar would use fewer words than a truth-teller, since a liar has to make everything up rather than simply remember it. Also, these studies did not look at online lying. Hancock suspects that it's easier to use more words in an electronic lie because you have time to think before you type.
Finally, you might think people lie more on email than in other forms of communication, but Hancock's previous work suggests this isn't the case. In fact, he's found that people lie much more over the phone, and even in person, than on email or instant messaging. He suspects that the automatic written record that computers create helps to deter electronic fibbing.
Now try and answer these questions:
- According to Hancock's study, how do false emails (or instant messages) differ from true emails?
- How do people behave when they're being lied to over email?
- Could a person use this information to design a computer program to catch online liars? What obstacles would stand in the way of the program working?
- If you suspected someone was lying to you over email, how might you use the knowledge from this study to try and expose the lie?
This paper and many others by Jeff Hancock are available on his Web page.
Natural Born Liars, a feature by the Australian Broadcasting Corporation, covers the evolution of human lying as well as technologies for spotting possible liars.
Science-based books on the psychology of lying include Charles V. Ford's Lies! Lies! Lies!: The Psychology of Deceit (American Psychiatric Press, 1999) and Aldert Vrij's Detecting Lies and Deceit: The Psychology of Lying and the Implications for Professional Practice (John Wiley & Sons, 2000).