About R. Kelly Garrett

R. Kelly Garrett (PhD, Information, University of Michigan, 2005) is an Assistant Professor in the School of Communication at the Ohio State University. His research interests include the study of online political communication, online news, and the ways in which citizens and activists use new technologies to shape their engagement with contentious political topics. His work has presented at HICSS and CSCW, and has been published in journals such as the Journal of Communication, Human Communication Research, the Journal of Computer-Mediated Communication, Communications of the ACM, and Daedalus.

The promise and peril of real-time corrections

Claims about science and politics made online are often startlingly inaccurate—try searching the Web for vaccines and autism or for Obama’s birthplace—but perhaps technology can help. Systems like Dispute Finder and Hypothes.is, which use web-page annotations to help correct falsehoods at their source, have obvious appeal.

Screenshots

Screenshots of the Dispute Finder interface and Hypothes.is website.

But is flagging falsehoods instantly an effective way to correct misperceptions?  Or might it make readers more defensive about their existing beliefs?

We used an experiment with 574 participants to compare different styles of correction. Some participants read an inaccurate article with embedded corrections from a reputable fact-checking organization.  Others read the same material, but were only provided the correction after completing a 3-minute distractor task. A control group saw no correction until after the study.

The results suggest that we should approach real-time corrections with caution.

  • Instant corrections can be effective: Across all participants, beliefs were slightly more accurate when corrections were presented alongside the misinformation than when the correction was presented later

BeliefAccuracy

Belief accuracy varied by when corrections were presented

  • Instant corrections only work for people who want to believe them:  Individuals inclined to reject the misinformation—for instance, an Obama supporter faced with a rumor about the President’s birthplace—benefit from this correction style.  Those inclined to accept it—those opposed to the President in the example—see almost no effect.  
  • Instant corrections promote distrust: The difference in effect appears to be due to heightened distrust of the correction’s source when the correction is presented alongside the falsehood.

These results represent a significant challenge to efforts to promote accuracy on the Internet, but they do not mean that such efforts are bound to fail.  Understanding the mechanisms that threaten to undermine instant corrections may allow us to create more effective systems going forward.

Want to know more? Please see our full paper (The Promise and Peril of Real-Time Corrections to Political Misperceptions) or visit the project website.

R. Kelly Garrett, Ohio State University
Brian E. Weeks, Ohio State University