Debunking the flat Earth is a relatively easy task. (NASA)
“Vaccines are safe and effective,” write researchers Philipp Schmid and Cornelia Betsch in a paper published in Nature Human Behavior. “Humans cause global warming. Evolution theory explains the diversity and change of life.” But large numbers of people do not believe that these statements are true, with devastating effects: progress toward addressing the climate crisis is stultifyingly slow, and the US saw its largest measles outbreak since 2000 in 2019, [among other vaccine-preventable infections since that time].
Getting accurate information across in the face of this science denialism is something of a minefield, as there is evidence that attempts to correct misinformation may backfire, further entrenching the beliefs of science deniers instead. In their paper, Schmid and Betsch present some good news and some bad: rebutting misinformation reduces the ensuing level of science denialism, but not enough to completely counter the effect of the original exposure to misinformation.
Denialism is not skepticism
Schmid and Betsch make a point of emphasizing that science denialism is a universe away from a healthy skepticism. In fact, skepticism of existing results is what drives research to refine and overturn existing paradigms. Denialism, the authors write, is “dysfunctional” skepticism “driven by how the denier would like things to be rather than what he has evidence for.”
Because this denialism springs from motivated reasoning, science advocates are scrambling to understand how to debunk misinformation in a way that motivates their target audience to accept it. Schmid and Betsch focused on strategies to counter misinformation as it is being delivered during a debate, focusing on two possible approaches: correcting misinformation and laying bare the rhetorical techniques that are being used to obfuscate the truth.
For instance, in the case of vaccine denialism, a denier might argue that vaccines are not completely safe. Correcting this misinformation (which Schmid and Betsch call a “topic” rebuttal) could take the form of arguing that vaccines in fact have an excellent safety record. A “technique” rebuttal, on the other hand, would point out that demanding perfect safety is holding vaccines to an impossible standard and that no medication is 100 percent safe.
Arguing does help
Schmid and Betsch gathered some participants and asked them about their attitudes toward vaccines and intention to vaccinate and then played them two different vaccine denialism arguments. One group of the participants then listened to a topic rebuttal delivered by a science advocate, another to a technique rebuttal, and a third group to a combined topic and technique rebuttal. A fourth group had no rebuttal (although they did have a debrief at the end of the experiment). Afterward, participants were asked again about their attitudes and intentions.
Dismayingly, exposure to the denialist arguments had an overall negative impact on attitudes and intentions, regardless of the rebuttals the participants heard. But the rebuttals did successfully mitigate this negative impact. To test the robustness of their results, Schmid and Betsch conducted five replications, testing that their results remained the same in different population groups (students compared to a national sample) and cultures (Germany and the United States). They also tested whether the same rebuttal tactics worked for climate change and whether the presentation—with the debates delivered in audio or written format—made a difference.
The results did vary somewhat. In particular, the experiment that focused on climate change found that neither topic nor technique rebuttals resulted in a significant difference from no rebuttal. But when the results of all six experiments were combined to create a larger, more-powerful data set, the overall picture was that both topic and technique rebuttals worked equivalently well. The researchers also discovered that the combined rebuttals had no additional benefit.
In other words, it's effective to either present audience with accurate facts or describe the rhetorical techniques that had been used to spread misinformation.
Better to cancel than debate
The results, write Schmid and Betsch, suggest that advocates can pick the strategy they’re more comfortable with. Critically, they saw no evidence of a backfire effect and, in fact, tentatively suggest the opposite—that people who were more vulnerable to the misinformation on offer were more likely to benefit from rebuttal.
It’s difficult to know how these results might translate to the long term—attitudes. Intentions aren’t perfect measures of people’s beliefs, and these studies can’t say whether the effects of the rebuttals would wear off over time. Still, rebutting in the context of a debate is just one small segment of what ideally needs to be a “multilayered defense system,” writes Sander van der Linden in a commentary on the research. Research into “cognitive vaccines” suggests that teaching people how to spot misinformation before it occurs holds a lot of promise, and it’s possible that rebuttals could be more effective in an “inoculated” audience, suggest Schmid and Betsch.
But one thing seems clear: it could be better to turn up and debate a denialist than to stay away, a tactic that is sometimes advocated out of fear of legitimizing the denialism. There’s an important exception to this, though: “if the advocate’s refusal to take part in a debate about scientific facts leads to its cancellation,” the researchers write, “this outcome should be preferred.” No amount of rebuttal can make up for exposure to misinformation.
Nature Human Behavior, 2018. DOI: 10.1038/s41562-019-0632-4 (About DOIs).
Cathleen is Ars Technica’s contributing science reporter, covering behavioral sciences and life sciences. She has a Ph.D. in cognitive science.