Cailin O’Connor and James Owen Weatherall—

Is that article fake news?  Is this meme trustworthy?  Is that scientific claim reliable?  Over the last few years, it has become clear that most of us are not very good at answering such questions.  For this reason, social media has contributed to a growing crisis in false belief.  Part of our difficulty stems from the fact that we all depend heavily on social information—information we learn from others—in deciding what to believe.  If you believe that tuna can carry high levels of mercury, for example, it almost certainly isn’t because you measured the mercury levels in tuna.  Someone else did, and they shared it with some of their colleagues, who shared it with others, until finally you heard about it far down the line.  Our ability to share knowledge and experiences in this way is essential to our ability to learn about our world.  But it means that we need to trust others to share reliable information.  This trust exposes us to a danger, because often what we learn from others is wrong.  Worse, it is increasingly the case that those who wish to influence public belief are exploiting the ways we share information in our social lives, especially on social media, to manipulate us.

There are ways to protect oneself against false belief in the age of social media.  Here we have drawn on the science of social knowledge, and what we know about misinformation online, to create some guidelines for doing so.

1. Trust the source, not the sharer.  As social learners, humans employ heuristics in deciding whom to trust.  We tend to trust people we identify with and know.  A recent study found that in deciding what to trust and share on social media, individuals were more attentive to the sharer than to the original source of an article.  This is a mistake.  Reputable news sources have fact checkers and strong incentives to report facts accurately; they also have editorial practices that allow them to correct their own errors.  They are, for the most part, trustworthy.  Confused about a current event?  See what the Wall Street Journal, the New York Times, or the Washington Post says.

2. Remember that your reaction to an event isn’t the only one.  In response to a political or social event, you might find yourself surrounded by a storm of outrage, or a warm glow of approval on social media.  Researchers have found that networks of retweets and interaction about moral content on Twitter are highly segregated by political affiliation.  These researchers also found that people are generally more likely to share emotional content.  For this reason, social media is inadvertently selecting for the content that most drives polarization.  Be wary when friends share highly emotional moral content, and remember that elsewhere in the social network, other perspectives are likely being shared and you are not seeing them. 

3. Fight confirmation bias.  People tend to trust evidence that confirms beliefs they already hold and ignore evidence that pushes against these beliefs.  If you find yourself only trusting and sharing things that you already believed, you may be falling into the confirmation bias trap.  Along these lines, be wary of articles that report on a controversial topic, but where it is entirely unclear why anyone would hold the other position in the controversy.  Such articles are designed to get clicks and shares by appealing to confirmation biases.

4. Watch out for surprising scientific findings.  In general, people have a bias towards novelty.  We are fascinated by things that are surprising or new.  This translates into likes, click-throughs, and shares on social media.  And this means that journalists are incentivized to cover the surprising and novel, including in coverage about science. But in science, surprising findings are also often wrong or misleading.  Not every study reflects a true effect, and some studies fail to replicate.   Studies that fail to replicate, though, are more likely to be reported on, and more like to be shared on social media, presumably because they are more surprising.  This unfortunately means that if you’ve heard about a scientific finding on social media, it is more likely to be false than one you haven’t heard of.

5. Read and share science journalism that covers a whole literature, not a single study.  One solution is to read, trust, and share scientific articles that report results from an entire literature, rather than focusing on a single study.  Because scientific evidence is probabilistic, any individual study can be misleading.  But an entire body of evidence, gathered by many scientists, replicated, and critiqued within a scientific community is less likely to mislead.  Ignore sensationalizing articles about one study.  (And no, wine isn’t better than exercise for your health.)

6. Remember, Russian agents are out there. We are unfortunately in a media environment where we are regularly brought into contact with content created and spread by foreign actors trying to manipulate public beliefs.  These agents are extremely savvy about what will be shared and liked.  One major goal seems to be to polarize and divide the US electorate and to erode trust in the US democracy.  For this reason, it is not safe to assume content created by sources you have never heard of is safe or reliable—even (or especially) if it tends to support beliefs or positions you already accept.  Russian agents pose as actors on both sides of the political divide in an attempt to create trust with social media users.  To stymie them, avoid sharing unsourced content, including political memes.  And avoid content that is overly emotional, one-sided, and divisive.  And last, don’t assume you know what Russian online propaganda looks like.  Over time, their disinformation campaign will continue to evolve as US users become more savvy.  (For instance, the Guardians.ai research team have identified Twitter bots that seem designed to avoid Twitter’s detection algorithms.) Cultivate a skeptical attitude towards social media content, and use verified sources to check scientific and political facts before trusting, liking, and sharing.

Cailin O’Connor is associate professor of logic and philosophy of science at the University of California, Irvine. James Owen Weatherall is professor of logic and philosophy of science at the University of California, Irvine, and author of the New York Times best-seller The Physics of Wall Street. Both are members of the Institute for Mathematical Behavioral Science.

Further Reading: