Can experts save the public from error?
If the "problem of misinformation" refers to human fallibility, there is little reason to think credentialed experts can solve it.
Jennifer Allen and David Rand recently published an interesting article in Scientific American. It begins by reporting the scientific consensus that exposure to online fake news is rare and concentrated among a narrow fringe of active social media users with extreme worldviews. It then raises a puzzle: if misinformation, so defined, is rare and mostly symptomatic of pre-existing beliefs, what explains the prevalence of public misperceptions, such as widespread election denial or inaccurate beliefs about vaccines? “The key,” Allen and Rand suggest, “is that narrowly defined ‘fake news’-style misinformation is only a very small part of what causes misbelief.”
Given this, they recommend “expanding the definition of misinformation” to include "anything that leads people to be misinformed,” encompassing not just outright fake news but false and misleading communication from political elites, pundits, and mainstream media outlets.
Using this new definition, academics must “look beyond narrow sets of previously debunked claims and study the roots of public misbelief more broadly”, mainstream media must be more reflective about its role in spreading misinformation, and social media companies must implement more policies to protect the public from misleading content beyond fake news.
They conclude:
“Combating misbelief is much more complicated—and politically and ethically fraught—than reducing the spread of explicitly false content. But this challenge must be bested if we want to solve the “misinformation” problem.”
Although I agree with much of this, I see things differently in two ways.
First, I think the “problem of misinformation” they identify is much more expansive and complex than they acknowledge. As they frame things, it simply boils down to the problem of human fallibility.
Second, I am much more pessimistic about whether credentialed experts like misinformation researchers are well-positioned to solve this problem.
Misinformation is not “anything that leads people to be misinformed”
First, once misinformation researchers move away from a focus on clear-cut fabrications like fake news, how should they define “misinformation”? Allen and Rand suggest “anything that leads people to be misinformed” (i.e., to endorse “misperceptions”).
One problem with this is that many factors that lead to misperceptions are distinct from misinformation. For example, tribalism, motivated reasoning, and institutional distrust drive misperceptions. Although they might make people more receptive to “misinformation” in a broad sense, they are not the same thing.
More importantly, for matters relevant to public opinion—vaccines, economics, elites’ behaviour, election integrity, climate change, etc.—the truth is not the default state that people can only be pushed away from by exposure to misinformation.
In all these domains, people’s strong intuitions are often systematically misaligned with the truth. When combined with the fact that many people are ignorant and distrust reliable information sources on such topics, misperceptions can arise without exposure to any misinformation. For example, vaccine hesitancy is often driven by the fact that many people find vaccination extremely counterintuitive and do not trust their doctor or the medical establishment. Misinformation need not play any role in this.
For these reasons, the study of misinformation differs from the study of misperceptions. Not only is the concept of misinformation (misleading communication) distinct from misperceptions (inaccurate beliefs) but misperceptions are not always caused by exposure to misinformation.
The many sources of misleading content
This suggests that misinformation should be defined not as anything that leads to misperceptions but as any communication that leads to them. However, once understood this way, misinformation researchers would have to broaden their focus far beyond the kind of content that Allen and Rand identify.
A vast range of communicated content shapes people’s beliefs, including misbeliefs. Consider Childish Gambino’s “This is America” or Oliver Anthony’s “Rich Men North Of Richmond”, two massively influential songs in recent years.
Do they convey an evidence-based analysis of their subject matters likely to maximise the accuracy of their audience’s beliefs? Of course not. Would anyone feel comfortable characterising such content—or any other simplistic and misleading pop songs that bear on matters of public opinion—as “misinformation”? I doubt it.
Or consider Hollywood movies. “Don’t Look Up” was a massively popular 2021 movie nominated for Best Picture at the Academy Awards. It was explicitly presented as an allegory of climate change. Considered as such, it is highly misleading, completely misrepresenting the expert consensus on the threats posed by climate change, which are nothing like a comet threatening human extinction. Highly simplistic and misleading cinema like this is ubiquitous. Would such films also come to qualify as misinformation?
These examples barely scratch the surface of potentially misleading content. Many popular novels, video games, religious texts and sermons, and informal conversations at pubs, cafes, around the water cooler, and at the family dinner table would also have to be included.
Misleading communication is pervasive in all such contexts. Once one expands the meaning of “misinformation” to encompass all forms of misleading content, one therefore ends up in weird places.
Social science is misleading
Moreover, once the term “misinformation” moves beyond clear-cut fake news to include misleading communication more broadly, it must also encompass an enormous amount of the content produced by credentialed experts.
Recently, I tweeted what social psychologist and misinformation researcher Jay van Bavel called a “simply wrong hot take” that was “insane”: namely, that one reason for being sceptical that social psychologists can establish a reliable science of misinformation is that few parts of society have been associated with more influential misinformation in recent decades than social psychology.
Keep reading with a 7-day free trial
Subscribe to Conspicuous Cognition to keep reading this post and get 7 days of free access to the full post archives.