Why internet 'trolls' are so cruel—and what we can do to stop them

Research shows cyberbullying is caused by mood, environment

According to a recent Pew Research Center survey, four in every ten adults using the internet have experienced harassment online—and even more have seen it happen.

To get to the bottom of why internet "trolls"—those who harass others online—continue to rampantly engage in this behavior, three researchers conducted a study of 667 online participants.

The researchers were Justin Cheng, a doctoral student in computer science at Stanford University, Cristian Danescu-Niculescu-Mizil, an assistant professor of information science at Cornell University, and Michael Bernstein, an assistant professor of computer science at Stanford University.

To conduct the research, Cheng, Danescu-Niculescu-Mizel, and Bernstein gave the 667 participants one of two quizzes—one of which was much harder than the other—and then asked participants to read an article and comment on a discussion thread. The discussion threads differed in that some already had malicious, troll comments written, and others had more neutral comments written.

Using the results of the experiment in combination with an analysis of 16 million comments made on CNN.com, the researchers were able to conclude that trolling is caused more by a person's environment than any inherent trait.

People are more likely to post troll comments when:

  • They are in a bad mood;
  • A discussion already has troll comments on it; and
  • It's late at night or Monday, when people tend to be in a bad mood.

The researchers' findings stray from the typical assumption that, as the researchers describe, "trolling is done by a small, vocal minority of sociopathic individuals."

Instead, they note, "it's not just sociopaths who are to blame... many 'trolls' are just people like ourselves who are having a bad day."

Based on the findings, researchers came up with several tactics for curbing internet trolling, such as:

  • Allowing people to delete posted comments, in case they regret spur-of-the-moment posts;
  • Encouraging positive comments by pinning a post about a website's rules of conduct to the top of a discussion board; and
  • Considering the intentions behind troll posts to distinguish whether the troll actually intended to harm readers rather than simply express an alternate viewpoint.

These tactics can "help separate undesirable individuals from those who just need help communicating their ideas," the researchers explain.

You may think of cyberbullying as more of a K-12 issue, but it's also pervasive in higher ed. According to research from Michigan State University (MSU), 18% to 20% of student respondents reported being cyberbullied at some point, and 70% to 75% of student respondents said they had witnessed cyberbullying.

Popular social media apps struggle to control cyberbullying

Saleem Alhabash, an MSU professor in the Department of Advertising and Public Relations who spearheaded the research, says it is important to redefine the public's perception of cyberbullying.

"What does it mean when someone trolls another person on Twitter?" Alhabash asks. "Is that bullying or is that freedom of speech? The lines are very fine that distinguish someone saying their opinion vs. someone saying their opinion in an aggressive way vs. someone attacking another person" (Bozack, State News, 2/6; Bernstein et al., The Conversation, 3/1).

How to prepare for a social media mess


Next in Today's Briefing

Hackers drown university network in tsunami of seafood searches

Next Briefing

  • Manage Your Events
  • Saved webpages and searches
  • Manage your subscriptions
  • Update personal information
  • Invite a colleague