Published July 23, 2012
Social media can reveal much about an individual’s personality but now researchers believe they may be able to spot psychopaths on Twitter.
By analyzing speech patterns, researchers at Florida Atlantic University were able to hone in on the “Dark Triad” of psychopathy, narcissism and Machiavellianism and were able to predict with a "reasonably high degree of accuracy" if a Twitter user was a possible psychopath.
The study is inspired by research released last year by Cornell University, which analyzed the speech patterns of known psychopaths. Researchers found that psychopaths tend to swear more and use emotionally charged words like “hate.” They were also twice as likely to use words relating to physical needs such as food, sex or money. Interestingly, confirmed psychopaths also preferred the past tense and used more conjunctions such as “because,” “since” or “so that” and filled their speech with more “ums,” “uhs” and “blahs.”
In collaboration with the Online Privacy Foundation Kaggle, a site that focuses on data science competitions, the team will present their findings next week at DefCon, one of the world’s largest hacker conventions.
“The FBI could use this to flag potential wrongdoers, but I think it’s much more compelling for psychologists to use to understand large communities of people,” Chris Sumner, of the Online Privacy Foundation, told Forbes.
Researchers tested the model by analyzing over three million tweets from almost 3000 tweeters across 80 countries. Of the 2,927 tweeters, 41 were certifiable psychopaths, meaning they had high ratings for the three main traits. Beyond having their tweets analyzed, participants were asked if they agreed with statements like “Payback needs to be quick and nasty,” “I have been compared to famous people” and “Most people are suckers” in order to measure the “darkness” of their personality.
The results were quite illuminating, according Florida Atlantic University’s Randell Wald. Wald explained to Wired that the results show there are “a number of statistically significant correlations between an individual’s darker personality traits and their Twitter activity.
But while this could be a useful tool in better understanding criminals or even potential criminals, using it to prevent a crime before it happens is probably unrealistic, Sumner says. “Just because someone scores highly doesn't mean they're criminally minded,” he told Forbes.
“Using this to try to spot someone who's going to commit a crime is going to result in catching people who aren't going to.”