Attribution: Gage Skidmore
Misinformation has been big news lately—and emotional misinformation more so. Donald Trump’s claims that thousands in New Jersey cheered the fall of the Twin Towers, that the Mexican government sends “the bad ones over,” and that most white people are killed by African-Americans all seem rooted in fear, alongside a good dose of racial stereotyping. Both President Obama and Republican presidential candidate Mike Huckabee describe Trump as tapping into voter anger. Norman Ornstein of the conservative American Enterprise Institute says the public’s desensitization to inflammatory rhetoric has created the perfect environment for Trump’s bombast.
And while much of this is speculation, there is data to show Trump has a unique angle on outrage. The Pew Research Center found that most of the GOP candidates appeal to angry voters, but Trump gets markedly unfavorable support from right-of-center voters who are not angry but who are content or merely frustrated.
Meanwhile, Caitlin Dewey of the Washington Post recently penned the final installment of her reliable and entertaining column, What Was Fake on the Internet This Week, blaming its demise on a shift in the fake-news ecosystem. Dewey argues that faux news sites have moved beyond celebrity death hoaxes and started to cynically prey on people’s political and ideological biases. They also do a brisk trade in schadenfreude, racial prejudice, and hate.
Whether or not you think that we’ve reached some kind of anger apex, the events of 2015 do raise an important question: what role do anger and other strong emotions play in the spread of misinformation?
Emotion and Information Flow
Research suggests that emotions play a number of important roles in how information spreads, if it is believed, and how we act upon it. First, it’s understood that anger and other strong emotions—both positive and negative—are a powerful factor in helping information go viral. In an analysis of 7,000 New York Times articles,1 Jonah Berger and Katherine Milkman of the University of Pennsylvania found that the high-arousal states of awe, anger, and anxiety were all correlated with high virality, while low-arousal states such as sadness were correlated with low virality. In fact, a one standard-deviation rise in anger level led to a 34 percent increase in the likelihood that an article would appear on the Times’ most-emailed list.
Then, there are the effects that emotions play in our information-seeking. Several researchers have found that anger causes people to spend less time looking for information on a political campaign2,3. Nicholas Valentino (then of the University of Texas, now of the University of Michigan) offers two possible explanations: first, anger may distract people as they try to perform complex cognitive tasks. Second, it may cause more targeted information-seeking, meaning people can reduce search volume while improving search efficiency. Valentino does not explain why anger in particular would cause either outcome.
UNC’s Michael MacKuen likewise found that participants experiencing anger and similar emotions seek less information overall, less information from the opposition, and more information that reinforces their prior beliefs. The author characterizes anger, disgust, contempt, and hatred as components of aversion—an emotional strategy people employ when encountering familiar, negative stimuli. When confronted with such conditions, MacKuen suggests, we fall back on well-worn coping routines. These often involve either ignoring the disturbing information or seeking information to confirm our prior views.
MacKuen’s theory of emotions’ effects on information-seeking.
Anxiety, on the other hand, results from unfamiliar situations. Anxiety signals that our existing habits are not sufficient to deal with the potential threat we now face. It therefore makes sense for anxious people to seek information, even that which challenges their prior views, and MacKuen’s experimental results bear this out, too.
“A standard expectation is that when people get emotional about politics, their positions become entrenched, resolute, and steadfast,” MacKuen notes. But his data seems to support this expectation only for anger, not for anxiety. When anxious, it appears, we can actually suspend the partisan biases that so often drive our reasoning.
Valentino, meanwhile, found more variable results for anxiety: sometimes this emotion suppressed information-seeking, sometimes it promoted it. But the difference between results for anxiety and anger were still significant. “Outrage, perhaps, is more damaging than fear if we hope to foster an informed citizenry,” Valentino concluded.
Information Processing and the “Lamestream Media”
Once we’ve acquired information, emotions also likely influence how we process that data. Specifically, they may do this by moderating or enhancing the power of our group identities. Researchers have repeatedly found that such affiliations play a great role in how we evaluate misinformation. Through the process of motivated reasoning, we are compelled to dismiss information that threatens our identities, and to accept information that solidifies our identities. Corrections that threaten our identities can actually backfire, strengthening our belief in misinformation.4
University of Michigan professor Brian Weeks wanted to investigate how emotions tap into the motivated reasoning process. In his experiment,5 participants were asked to write about topics that made them angry or anxious. They were tested to see if the writing produced the desired emotion, and in most cases it did. (There was also a control group that did not receive the emotional manipulation.) Then they were exposed to misinformation on the same topic, which appeared to originate either from their in-group (the ideology with which they affiliated) or out-group (the opposing ideology).
Weeks concluded that anger increases the likelihood that people will process information in a partisan way, while anxiety decreases the likelihood. Angry participants who received misinformation from their own party ended up holding less accurate beliefs, compared to participants in a neutral emotional state. It appears that the anger intensified these participants’ motivated reasoning, making it easier for them to accept misinformation that confirmed their group identities.
On the other hand, anxious participants who received misinformation from the opposing party also held less accurate beliefs. That is, anxiety appears to have diminished their motivated reasoning, to the point that they were actually more likely to believe misinformation from their opponents.
This may seem strange, but makes sense in light of the previous findings that anxiety provokes information-seeking and the consideration of outsider viewpoints. An anxious person seems to assume that his own beliefs and knowledge are not sufficient to deal with the situation at hand, and becomes more open to persuasion and compromise. Such open-mindedness can be a boon to our democracy, but if the outside information is actually erroneous, the person can end up worse informed than before.
There is another important aspect of anger to consider. It’s worth noting that Trump plays not just on anger about politicians and political issues but also on anger at the mainstream media. Not surprisingly, such anger is bound up with identity protection, too. In a study of 293 undergraduates,6 Florida State’s Laura Arpan and UC Santa Barbara’s Robin Nabi found that identity-threatening news content was more likely than neutral content to provoke anger responses. Anger made the participants more likely to read identity-affirming content, as well as, strangely, identity-threatening content. In the latter case, participants might be engaging in surveillance, trying to learn the extent of potential harm to their in-group. Anger also made subjects more likely to read irrelevant content, perhaps as a way to distance themselves from their emotions.
So What’s the Deal with Trump?
If we accept the balance of findings above, then we may gain some insight into the Trump phenomenon. Trump supporters are no great fans of the mainstream media, if they’re typical of Republican voters, only 27 percent of whom have a fair or great deal of trust in media. Likely Trump’s camp trusts the media even less than that. When the mainstream media undermines Trump’s claims (as fact-checking organizations such as PolitiFact, FactCheck.org, and the Washington Post Fact-Checker do),7 his supporters dig in their heels, as predicted by the theory of motivated reasoning.
Since the fact checks are identity-threatening, they are also likely to provoke an anger response. This further enables acceptance of misinformation. Anger could cause the supporters to seek information that confirms their identity affiliations—or it may shut down information-seeking that poses a threat to their identity. The anger also will likely provoke sharing on social media, helping to galvanize fellow supporters—and the cycle continues. As Weeks argues, “Anger therefore has the potential to create media diets in which people are primarily exposed to like-minded messages, which may further enhance anger directed at political opponents.”
This is by no means a proven process or a complete picture. For example, there is some tension between the dual information-seeking effects proposed above: can people really cleave their information-seeking so neatly between “identity-affirming” and “identity-disturbing”? Yes, there’s a myriad of partisan news sources available to all of us, promising to confirm our preconceived notions. But it’s hard to imagine that we pursue information in such entirely dissimilar states, rather than in the many gradations in between.
And what of fear? If we assume that Trump is tapping into old, familiar anxieties, then MacKuen’s work would support the idea that supporters then reduce their information-seeking. Terrorism, immigration, races other than one’s own—these all seem to be old bogeymen. On the other hand, ISIS in many ways represents a new threat, employing different tactics than other terrorist groups like Al-Qaeda, so it would be interesting to explore how voters’ information-seeking have responded. And it seems that anxiety can both promote and inhibit information seeking, as Valentino found, so overall the picture is a bit murky.
Attribution: littlewonder. Source material: BBC.
Are We So Outraged?
It would also be wise to take a step back and ask ourselves: Is the public anger we observe today really so new or special? Have we reached a zenith of ire? Vox labelled 2015 “the year of fake outrage,” only a year after Slate exhaustively chronicled what angered us in 2014, “the year of outrage.” Jon Ronson recently argued that Twitter has evolved into an arena for electronic shaming, with its stone-throwing citizenry getting froth-at-the-mouth angry 140 characters at a time. But even if his anecdotal assessment of Twitter content is correct, there have always been many corners of the Internet—really, more like atriums, or arenas—for us to air our most base feelings.
Intriguingly, the Pew Research Center recently found that 22 percent of voters say they are angry with the federal government, down from a record 30 percent during the government shutdown two years ago. That’s not a huge decline but it’s certainly not a blossoming of anti-government feeling.
Despite this persistence—in fact, because of it—understanding the role of anger in political communication is undoubtedly an important task. As the Internet brims over with so many versions of the “truth,” anger appears to worsen our ability to recognize and dismiss misinformation. What’s more, while righteous anger can provoke positive social movements, bringing greater equality and helping to lift people out of poverty, anger can also serve as a conduit for racism and xenophobia.
Anger motivates action MacKuen notes: “Commonly, enraged partisans want to go to the streets, send in their financial support, and march to the polls.” And a politician who plays on anger will succeed in motivating those who already support him. On the other hand, MacKuen argues, undecided voters and oppositional voters are more likely to be swayed by an appeal to anxiety. So as we move towards the primaries, perhaps we’ll see Trump tone it down a bit (though I’m not holding my breath).
The solution is not to seek some Platonic ideal of pure rationality, which is neither achievable nor desirable. Anger, managed correctly, can drive ethical action. The key is to focus on the informational content of those anger states and find the most effective ways to fashion and spread correctives.
While the tenacity of motivated reasoning can sometimes drive one to despair, there is some hope in the knowledge that we don’t let motivated reasoning drive our information-processing all the time. Where beliefs are not crucial to identity, corrections are easier to make. And even strongly partisan beliefs are susceptible to a “tipping point”—a certain amount of corrective information that finally breaks through the wall of motivated reasoning.8 So I remain hopeful that the cause of political fact-checking is still worth pursuing. It will just take perseverance, creativity and, when we evaluate arguments, trying to keep our cool.
- Berger, Jonah and Katherine L. Milkman. 2012. What makes online content viral? Journal of Marketing Research 49(2): 192–205.
- Valentino, Nicholas A., Vincent L. Hutchings, Antoine J. Banks, et al. 2008. Is a worried citizen a good citizen? Emotions, political information seeking, and learning via the internet. Political Psychology 29(2): 247–273.
- MacKuen, Michael, Jennifer Wolak, Luke Keele, et al. 2010. Civic engagements: Resolute partisanship or reflective deliberation. American Journal of Political Science 54(2): 440–458.
- Nyhan, Brendan and Jason Reifler. 2010. When corrections fail: The persistence of political misperceptions. Political Behavior 32(2): 303–330.
- Weeks, Brian E. 2015. Emotions, partisanship, and misperceptions: How anger and anxiety moderate the effect of partisan bias on susceptibility to political misinformation. Journal of Communication 65: 699–719.
- Arpan, Laura M. and Robin L. Nabi. 2011. Exploring anger in the hostile media process: Effects on news preferences and source evaluation. Journalism & Mass Communication Quarterly 88(1): 5–22.
- Full disclosure: I am a consultant for the American Press Institute, which manages a project that seeks to increase and improve fact-checking and accountability reporting.
- Redlawsk, David R., Andrew J. W. Civettini, and Karen M. Emmerson. 2010. The affective tipping point: Do motivated reasoners ever “get it”? Political Psychology 31(4): 563–593.