Danielle Polage
In 2016, the website PolitiFact named fake news “the lie of the year.”1 There were so many fake news stories circulating online that it was hard for PolitiFact to pick just one. Fake news has been blamed for everything from disbelief in global warming and #Pizzagate to Brexit and swinging the 2016 U.S. presidential election.
Although spin and evasion in politics are almost expected, fake news has been described as more pernicious than typical political bias, as it is intended to mislead the public in systematic ways. If voters base their political views on false information, fake news is a real threat to democracy since the democratic process depends on informed voters. This chapter presents research demonstrating that people do believe fake news and that information coming from credible sources—as long as it does not contradict existing knowledge—will become more believable as a result of repetition. The chapter concludes with potential ways to counteract fake news and avenues of future research.
According to statistics from a research study conducted by Andrew Guess, Brendan Nyhan, and Jason Reifler, approximately one in four Americans visited a fake news website between October 7 and November 14, 2016, the critical last weeks of the presidential campaign.2 Through their analysis of web traffic from a nationally representative sample, they find that the most likely place where people accessed fake news was Facebook. It is estimated that fake news was shared millions of times through this platform. Clearly, people are exposed to fake news, so it is important to understand whether exposure to fake news will increase belief in it.
In one of the first studies conducted on belief in fake news, I presented forty-four undergraduate university students with fake and true news stories.3 Participants were asked to read these stories and then provide titles for them. An example of a fake news story included in the study claims that tomato paste causes male infertility. A sample true story describes the many health benefits of drinking green tea. Half of the participants were exposed to one set of true and false news stories during the first session of the experiment; the other group read a different group of true and false stories. Five weeks later, participants rated the likelihood of both old and new stories. Participants rated the false stories they had previously read as more truthful and more plausible than did participants who had not read those stories. Simply reading the stories one time made them seem more likely to be true. Participants also believed that they had heard about the repeated fake news stories from a source other than the experiment, even though it was not possible, since the fake stories had been experimenter-generated. This increased belief is a result of simply exposing half of the participants to each set of stories a single time. It is likely that participants were using familiarity as a basis for truth. If a story seems familiar, participants assume it is more likely to be true.
Many studies demonstrate that simply repeating a statement makes it seem truer. This illusory-truth effect was first observed by Lynn Hasher, David Goldstein, and Thomas Toppino and has just recently been applied to contemporary fake news.4 Gordon Pennycook, Tyrone D. Cannon, and David G. Rand used fake news headlines that had actually been posted on Facebook to demonstrate that a single exposure to the fake news increases its perceived truth.5 In general, when presented with information, people tend to believe it. This so-called truth bias demonstrates that our default is to believe what we are told. The truth is the most efficient way of communication, and most people tell the truth most of the time.6 Thus, it is our default to tell the truth and to expect to hear the truth from others. We usually lack the time and means necessary to determine the veracity of everything we hear. Therefore, we default to believing that others are generally being honest, unless there is a reason not to believe them. If we trust the source of the information, we will typically believe the information as long as it does not contradict information that we know to be true.
If new information can be rejected simply based on its contradiction to accepted truths, the statement will not be accepted, even if it is repeated. When Pennycook, Cannon, and Rand presented participants with statements that were clearly false, such as “the earth is a perfect square,” participants did not increase their belief in those statements even when viewing them for the second time.7 It seems that, when new information contradicts knowledge that we already have, we use our knowledge, and not the feeling of familiarity, to assess truthfulness. It is important to note that the incompatibility between the old and new information must be apparent and that the previous knowledge must render the novel statements implausible. If the contradiction to fact is not apparent, information that is known to be false might still increase in believability with repetition. For example, Lisa K. Fazio, Nadia M. Brashier, Keith B. Payne, and Elizabeth J. Marsh demonstrate that participants rated statements, such as “a date is a dried plum,” as more likely to be true after they are repeated, even though it was assumed that participants knew that a prune, not a date, is a dried plum.8 The authors suggest that the statement may have been similar enough to the truth to go undetected. They argue that a statement, such as “a grapefruit is a dried plum,” would likely have made the contradiction more obvious and should not result in that statement being believed. On one hand, these data suggest that it is possible to believe false information, even when it contradicts known information, as long as the contradiction is not noticed. On the other hand, information that clearly violates known facts is unlikely to be believed, even if it is repeated.
One potential way that news consumers might counteract belief in fake news is by increasing their knowledge base. However, news, almost by definition, involves the presentation of information that is relatively unknown to the general public. Fake news, therefore, is not likely to be debunked simply by reliance on contradictory knowledge, because the public often lacks the relevant information to dispute the claims. Take, for example, the birther conspiracy, which alleges that Barack Obama is not a natural-born citizen of the United States and was therefore ineligible to be president. Democrats find it shocking that some Republicans do not believe that Obama was not born in the United States. Yet it is unlikely that the location of Obama’s birth would be counteracted by direct knowledge of members of the general population. The statement itself is not really preposterous; the United States is made up of individuals who have immigrated to the country from all over the world. To process the veracity of the statement, however, one needs to have information about which the public lacks access, namely, a legal birth certificate. Yet even after Obama’s birth certificate was produced, a Newsweek story cited data from a YouGov poll in 2017 showing that more than half of Republicans surveyed still believed that Obama was born in Kenya.9 Why would that be? They likely did not believe the source of the information to be credible.
As mentioned previously, people will believe information to be true if it is repeated, if it does not contradict previously stored knowledge, and if the source has not been discredited. Even if the public had access to that original document, the average person would not have the expertise necessary to determine whether it was a legitimate copy. To believe that Obama was born in the United States requires a person to believe the experts who analyzed the documents. Thus, without any knowledge to contradict the claim, the credibility of the source becomes an important component of whether or not the information will be accepted. So how do we decide what is a trustworthy source? It turns out that people determine the credibility of the source based on whether or not the source espouses beliefs that support people’s preexisting opinions.
Dan M. Kahan, Hank Jenkins-Smith, and Donald Braman conducted a study looking at how perceived credibility of a so-called expert can change depending on whose side of a debate the expert is on.10 If the participants share the same views on a topic with the expert, participants perceive the expert as credible. However, if the expert does not validate the participants’ position, the expert is discounted. Participants in this study were given the credentials of a scientist with a Ph.D. in climate studies. The expert’s degree was from Harvard University, and he held a current professorship at the Massachusetts Institute of Technology. Both groups of participants received the same description of the scientist, but half of the participants were told that the expert said that it was undoubtable that humans were causing climate change, whereas the other half of participants were told that the expert said it was premature to claim that humans were causing global warming. Even though the credentials of the scientist were impressive and held constant across conditions, people only agreed that the expert was a credible source on climate change when he agreed with their opinions of the cause of climate change. Very conservative, committed Republicans tended to accept the scientist as a credible expert when he claimed that it was premature to claim that global warming was caused by human activity. Very liberal, committed Democrats dismissed his opinion unless he claimed that human activity caused global warming. These data suggest that the expert’s perceived credibility is not a reflection of his education and experience, but rather a reflection of whether or not his opinion matches the opinions of his audience.
Elizabeth E. Housholder and Heather L. LaMarre show that perceived credibility of a political figure also depends on the perceived similarity between the politicians and their constituents.11 The authors of this study created fake Facebook pages for fictitious senatorial candidates. Their Facebook pages included information that expressed views on a variety of issues. Half of the participants saw a candidate who held traditional Republican views, and half viewed the page of a candidate holding traditional Democratic views. Participants rated the credibility of the two candidates in terms of honesty and competence in addition to other factors. Participants also rated the credibility of the information posted on the fictitious candidates’ Facebook pages. Finally, participants indicated how similar they thought their beliefs were with the politicians’ beliefs. The results showed that the more similarity that viewers perceive between themselves and the candidate, the more credible they think the candidate is and the more credible they perceive the information provided by the candidates. This study supports the idea that source-credibility judgments are based on whether the source espouses information with which the individual agrees.
These studies suggest that viewers determine the credibility of a source based on whether the source supports views similar to the viewer’s own. If news sources do not align with a viewer’s preconceived beliefs, they will be viewed as less credible. Interestingly, mainstream news is losing credibility, and partisan-biased coverage of news stories may be one of the reasons why. In 1976, 72 percent of respondents to a Gallup poll said that they trusted the news, whereas in 2016, that number sank to 32 percent.12 Although a recent meta-analysis indicates that both liberals and conservatives show a similar partisan bias to accept information that confirms their beliefs,13 the Gallup poll shows a higher distrust of mainstream media among Republicans. Fifty-one percent of Democrats and 14 percent of Republicans expressed “a fair amount” or “a great deal” of trust in mass media as a news source.14 This decline in trust in mainstream media could also be a reason that belief in fake news is growing. The number of fake news stories targeting Republicans is higher than those targeting Democrats. It is possible that Republicans’ skepticism of mainstream media might lead them to seek alternative news that they feel is less biased against their positions.
Briony Swire, Adam J. Berinsky, Stephan Lewandowsky, and Ullrich K. H. Ecker show that people evaluate the veracity of information based on who says it.15 Participants were presented factual and false campaign statements that Donald Trump made prior to the 2016 presidential election. Those statements were either attributed to Trump or presented without an identified source. Participants were then given an explanation as to whether the statements were true or false. Participants were asked either immediately or after a one-week delay to rerate the statements in addition to rating their feelings toward the candidate. As expected, Republican supporters of Trump were more likely to believe both true and false information when it was attributed to Trump. Democrats were less likely to believe both true and false statements when they were attributed to Trump. Republicans who did not support Trump were less likely to accept facts spoken by Trump as true, but were not more likely to accept his false statements. In summary, the results show that participants’ opinions of Trump colored their perceptions of the accuracy of his statements. Respondents who supported Trump believed what he said, whereas those who did not support Trump did not believe him, even when what he was saying was true.
The default with sources believed to be credible is to expect to hear the truth from them. This can lead to increasingly polarized views if people are more likely to believe people who agree with them. Eytan Bakshy, Solomon Messing, and Lada A. Adamic find that the majority of people’s friends on Facebook share their political views: 80 percent for liberals and 82 percent for conservatives.16 Sources that are perceived to be reliable are also less likely to be fact-checked. This dynamic creates an echo chamber in which people’s own views are shared back with them, validated, and repeated.
So, how can fake news be counteracted? Since fake news will be believed if it does not contradict previously stored knowledge and does not come from a discredited source, the first line of defense against belief in false information is a deeper knowledge base. One suggestion is to increase general knowledge in a nonbiased way before people are exposed to fake news. The more people know about a topic, the less likely they are to believe false information about it.
Another possibility is to use source credibility as a way to counteract fake news. Those with perceived low credibility are less likely to be believed, whereas increased source credibility increases the likelihood that presented information will be believed. Source credibility is not necessarily based on the source’s experience or credentials, but is based on a perceived match between the opinions of the listener and the source. Fake news is typically targeted toward groups that are most likely to believe it. If a source that seems credible to members of the target group discounts the fake news, it might be possible to decrease the members’ belief in the fake news. For example, Berinsky looked at attempts to counteract rumors surrounding Obama’s Affordable Care Act.17 One such rumor was that people would be required to consult “death panels” when making decisions about end-of-life care. Berinsky’s study found that when a Republican involved in drafting the relevant legislature debunked the death panel story, all respondents, regardless of partisanship, were more likely to reject the rumor. It was assumed that a Republican who supported a Democratic plan would be perceived as a credible source. This study suggests that politicians can help reduce belief in fake news if they are willing to make statements based on truth, regardless of their personal political interests.
The same logic could apply to friends who are discussing issues, both in person and online. Facebook users, for example, could “like” posts that are truth-based, whether or not they support the posts. They could be open to challenging others on political facts, not just emotional rhetoric. Social media users could avoid echo chambers by surrounding themselves with intelligent and well-informed individuals who do not share their point of view. In addition, although difficult, social media users should reconsider unfriending those who disagree with them. This chapter discusses how the credibility of strangers is based on shared beliefs and opinions; however, it is likely that the credibility of a familiar person is based on other factors, such as shared experiences and interests. Exposure to repeated information from friends and family may increase receptivity to facts that challenge previously held beliefs, but only as long as the source is perceived to be credible. Trust is part of credibility. Everyone should attempt to maintain credibility in conversations by presenting facts with cordiality and respect and avoiding shame tactics and name-calling. Friends on social media look to each other as valued sources of information; therefore, individuals should not be a source of fake news. They should not “like” or repost something without reading and verifying it, even if it validates their opinion. Twitter users should not retweet stories when they have only glanced at the headline. All users of social media should report suspected fake news and ask that it be removed. Social media consumers cannot avoid fake news, but everyone has a responsibility to help prevent the spread of it.
In summary, it is hard to counteract fake news that does not contradict information known to be true, that comes from a source considered to be reliable, and that espouses claims that the person already believes. Fake news will continue to be posted online; the goal, then, should be to decrease the spread of fake news. Future research should continue to explore the limits of belief in fake news and to more clearly examine the effects of belief in fake news. Does belief in fake news translate into changes in behavior? Do people believe fake news that contradicts previous knowledge if it comes from a perceived credible source? The presence of fake news is unlikely to decrease; however, it is everyone’s job to fact-check and limit the spread of fake news before it is read by others.
1. Angie Drobnic Holan, “Lie of the Year: Fake News,” PolitiFact, December 13, 2016, http://www.politifact.com/truth-o-meter/article/2016/dec/13/2016-lie-year-fake-news/.
2. Andrew Guess, Brendan Nyhan, and Jason Reifler, “Selective Exposure to Misinformation: Evidence from the Consumption of Fake News during the 2016 U.S. Presidential Campaign” (working paper, January 9, 2018), https://www.dartmouth.edu/~nyhan/fake-news-2016.pdf.
3. Danielle Polage, “Making Up History: False Memories of Fake News Stories,” Europe’s Journal of Psychology 8, no. 2 (2012): 245–250.
4. Lynn Hasher, David Goldstein, and Thomas Toppino, “Frequency and the Conference of Referential Validity,” Journal of Verbal Learning and Verbal Behavior 16, no. 1 (1997): 107–112.
5. Gordon Pennycook, Tyrone D. Cannon, and David G. Rand, “Prior Exposure Increases Perceived Accuracy of Fake News,” Journal of Experimental Psychology: General 147, no. 12 (2018): 1865–1880, http://dx.doi.org/10.1037/xge0000465.
6. D. T. Gilbert, “How Mental Systems Believe,” American Psychologist 46 (1991): 107–119; Timothy R. Levine, “Truth-Default Theory (TDT): A Theory of Human Deception and Deception Detection,” Journal of Language and Social Psychology 33, no. 4 (2014): 378–392.
7. Pennycook, Cannon, and Rand, “Prior Exposure.”
8. Lisa K. Fazio, Nadia M. Brashier, Keith B. Payne, and Elizabeth J. Marsh, “Knowledge Does Not Protect against Illusory Truth,” Journal of Experimental Psychology: General 144, no. 5 (2015): 993–1002.
9. Julia Glum, “Some Republicans Still Think Obama Was Born in Kenya as Trump Resurrects Birther Conspiracy Theory,” Newsweek, December 11, 2017, http://www.newsweek.com/trump-birther-obama-poll-republicans-kenya-744195.
10. Dan M. Kahan, Hank Jenkins-Smith, and Donald Braman, “Cultural Cognition of Scientific Consensus,” Journal of Risk Research 14, no. 2 (2011): 147–174.
11. Elizabeth E. Housholder and Heather L. LaMarre, “Facebook Politics: Toward a Process Model for Achieving Political Source Credibility through Social Media,” Journal of Information Technology and Politics 11, no 4 (2011): 368–382.
12. Art Swift, “Americans’ Trust in Mass Media Sinks to New Low,” Gallup, September 14, 2016, http://www.gallup.com/poll/195542/americans-trust-mass-media-sinks-new-low.aspx.
13. Peter H. Ditto, Brittany Liu, Cory Clark, Sean P. Wojcik, Eric Chen, Rebecca Grady, and Joanne Zinger, “At Least Bias Is Bipartisan: A Meta-Analytic Comparison of Partisan Bias in Liberals and Conservatives,” Perspectives on Psychological Science 14, no. 2 (2019): 273–291, https://doi.org/10.1177/1745691617746796.
14. Swift, “Americans’ Trust in Mass Media.”
15. Briony Swire, Adam J. Berinsky, Stephan Lewandowsky, and Ullrich K. H. Ecker, “Processing Political Misinformation: Comprehending the Trump Phenomenon,” Royal Society Open Science 4, no. 3 (2017), https://doi.org/10.1098/rsos.160802.
16. Eytan Bakshy, Solomon Messing, and Lada A. Adamic, “Exposure to Ideologically Diverse News and Opinion on Facebook,” Science 348, no. 6239 (2015): 1130–1132.
17. Adam J. Berinsky, “Rumors and Health Care Reform: Experiments in Political Misinformation,” British Journal of Political Science 47, no. 2 (2017): 241–262.