Social Media and Fear: A Case Study

It has been an interesting couple of days at my new academic home, the University of Alabama. As is the case with the initial time period around many incidents involving the safety of large numbers of people, the facts are a little unclear. According to school officials and local police, here is what happened: someone posted a threatening comment on a University of Alabama sorority's YouTube video. The threat was specific, referencing a time in the near future and a place on campus, and a threat of harm to large numbers of people. Additionally, at least one student reported, and later retracted, a statement about being attacked off-campus. Very soon after, school officials worked with police and determined that there was no threat to our students' safety beyond the YouTube comment, that it appeared as though no actual person was planning on carrying out any attack. This information did not have the intended effect of calming and reassuring everyone that it was okay to go about our regular business. As misinformation spread, it prompted many of our students and their parents to be fearful, so fearful that the students did not feel comfortable coming to class, which disrupted what, ostensibly, we're here to do: learn stuff in class.

The fears were based in part on real-life events. There was, apparently, a rumor that someone was dressed as the Joker on sorority row, which relates to the shooting in an Aurora, Colorado movie theater in 2012. They were also based on something resembling an urban legend. A student of mine quoted a message he had seen that was circulating, citing it as part of the reason he was electing to stay home: "The name of the person who posted the comment on the sorority video, Arthur Pendragon, is the name of the main character in a book who went and killed a bunch of people on the fall equinox, which is tonight at 9:30...That is an actual person and he calls himself King Arthur because he believes he is the reincarnated King Arthur from hundreds of years ago and holds a celebration every fall equinox."

One question immediately occurred to me as all this was happening (or not happening): what was the role of social media in this occurrence of mass fear (or, if you like, fear contagion?)?

Did social media cause the mass fear?

In trying to answer this question, I try to maintain the stance of a skeptic: just because social media was involved in the event at various stages does not mean that it caused the event, or necessarily caused it to occur in a certain way. Social media could simply be standing in for face-to-face or other existing forms of communication (phone, television, radio, etc.). The underlying psychological mechanisms associated with our tendency to believe scary information that may not be true clearly pre-dates social media. Even the hint of the occurrence of a low-frequency/high-severity event is enough to set parts of our brains into overdrive (as it happens, those parts have been around for a very long time, before humans evolved). Tightly-knit groups of young people in particular were likely always especially susceptible to this kind of rumor. Thirty years ago, before social media, you could call in a bomb threat and freak out a campus. So maybe this isn't anything new.

On the other hand, perhaps some unique attributes of social media interact with those underlying psychological mechanisms in a way that brings about a new outcome that would not have existed without social media. So here is a consideration of how social media might have facilitated mass fear.

Social media as inciter

The initial threat was delivered via social media. Perhaps the person issuing the threat believed he could do so anonymously, achieving the desired goal of sewing fear in the community without getting caught. This is certainly true of lower-scale harm, such as bullying. As of now, you can (unfortunately) use the anonymity provided by social media to harass or bully another person but not get caught in the act. But once the stakes get high enough (e.g., terrorist threats), we suddenly see that even those posting anonymously can be tracked down. The person posting may not have known this to be the case. They may have believed that they could do this (because they hated the school, because they were bored, because they wanted to get out of an exam that day, who knows) and thought they could get away with it.

Social media also provides a kind of remove or abstraction that perhaps makes it easier to commit harmful or disruptive acts without considering the consequences. Even if the social media user can never be truly anonymous, not having to see the face, hear the voice, or be in physical proximity to the individuals they are harming or the lives they are disrupting likely makes it easier to do so.

Social media as effective propagator of misinformation

When you discuss a rumor with someone face-to-face, you can spread misinformation to one other person. When you post about it, you are spreading it to a larger group of people, potentially. I say "potentially" because while the potential audience for any given post is all Internet users (and, if it's picked up by the mainstream media, all users of TVs or radios), in practice, most posts are either ignored or seen by very few people. However, in events of great interest like this, each post becomes part of a whole. When people search for a specific term (e.g., University of Alabama incident), each person's post about the topic gets added to the tally. Ten people posting about a topic doesn't make it seem worth paying attention to, but if a million people post about it, even if it doesn't have much evidence substantiating it, it gets more attention which may lead it to spread more easily. When they are included in search counts and taken out of context, even reasonable social media posts (or de-bunking social media posts that attempt to correct misinformation) can inadvertently contribute to the stoking of mass fear.

Even if relatively few people post misinformation, if those people make up a substantial number of a given community (e.g., most people in your sorority, or most people in your Facebook feed), you're likely to believe that whatever they are talking about is worth paying attention to, regardless of how true it is.

It is also very easy to post misinformation on social media; easier, I would argue, than talking to another person face-to-face. For most of our students, social media is always available. Many of them are in the habit of posting their thoughts and feelings frequently.

It is also easier for misinformation to spread this way. Linking to other information sources is an integral part of the affordances and norms involved in all Internet use. Unfortunately, citing the sources of information (and the sources of those sources, until you get to a primary source) is not. One wonders whether this lack of substantiation will persist after we go through more and more of these episodes. It certainly doesn't have to. Theoretically, I'd imagine you could design a quick and simple way to track the flow of each bit of information from link to link to link, back to an original source. Those bibliographies that were so annoying to format may be more useful than we thought!

Social media as abettor of flawed reasoning

This came to mind as I tried to make sense of the fact that there were multiple alarming events reported across a period of 48 hours. People tend to see patterns, connections, and stories in human behavior even when they are not really there. This is likely especially true of people who are in a state of fear. At some point, people's flawed reasoning would run into counter-evidence that would suggest that the perceived connection among these events isn't really there. It occurs to me that in most cases, that counter-evidence comes to us from authorities: police or officials. Often, the messages from those authorities are mediated in some way, coming to us through mainstream media channels (e.g., television news). Which brings me to the next consideration of the role of social media in stoking mass fear.

Social media as a perceived unfiltered alternative to mass media narratives

I've been thinking more and more about the importance of trust in modern life. Our fates are connected to one another in a million ways we tend to ignore. Most of us assume that our packaged food is safe to eat, that our cars are safe to drive, that our votes will be counted when electing a political candidate. But do we trust authorities to tell us about threats to our safety? Some events have given us reason not to.

Under oppressive regimes, the authorities tells one story through government-controlled mass media while social media tells another story, a story "from the ground". We know that the authorities have an incentive and an ability to deceive, so we might be more likely to believe the social media narrative. The social media narrative also comes not just from one source but from many individuals, so it would seem that it would be less likely to be corrupted or biased. This ignores the fact that those posting on social media may have their own agendas, and though they may be posting uncensored pictures or facts about what is going on in the world, they may be purposely (or unconsciously) ignoring other images or facts. Thus, social media may only appear to be a less biased place to get unvarnished information about the world.

Social media narratives have the advantage of being more difficult to repress or control, but they have the disadvantage of not having a reputation. When the social media mob gets things wrong, the reputations of each individual posting and forwarding messages do not suffer in the way that the reputations of mainstream news organizations (or authorities like the police) do if they get things wrong. Social media can respond quicker to events and spread rumors because there is less of a price to pay if the information turns out not to be true.

This, of course, doesn't matter much if you don't trust information from authorities or the mainstream media. There was likely always some distrust of such official narratives, but there weren't many alternatives besides the odd underground newsletter or the person ranting and raving on the street corner. Social media fills this void. It presents people already predisposed to distrust official narratives with a seemingly trust-worthy ("unvarnished, unbiased") alternative.

In the case of the events (or non-events) at the University of Alabama, I wonder about students' trust in messages from authorities and their trust in messages from social media. Their trust in social media messages may not reflect an ignorance or unawareness of information from reliable sources, but a cynicism about just how reliable those mainstream sources are.

There are many important questions which will be answered in different ways by mainstream and social media sources: How many people really are dying from Ebola? How big of a threat is ISIS, really? Who we believe is, in part, a reflection of who we trust.

Teachable moments

In the end, I hope things settle down quickly so our students, all of our students, will come back to class. I'm already thinking about how we can turn this into a productive discussion about where we get information from, what sources we trust, and how we all might do things differently next time.

Hoaxed

A couple of viral stories, or memes, circulating through the information ecosystem over the past month got me thinking about hoaxes and viral media. They're both stories about interactions between strangers. They have the elements of any good story: conflict, a moral element, a character with whom you can identify. They both incorporate pictures of written documents which have the effect of serving as proof of the events (they make the stories slightly more believable than if there were no pictures at all). One took place on an airplane. The other took place in a restaurant. They both turned out to be hoaxes. Many people, including major news outlets, believed them at first.

My initial thoughts on hoaxes and new media revolved around the premise that hoaxes happen when any information dissemination technology is new. They expose the fact that we have placed too much trust in the new source of information. Think of War of the Worlds Halloween broadcast (radio), the quiz show scandals of the 50's (TV), Lonelygirl15 (online video), Manti T'eo (online dating). Whether intended or not, hoaxes help audiences or users develop a proper skepticism toward information from the new technology.

But this latest rash of hoaxes seem like something else. This just seems to be par for the course in the era of decentralized, virally spreading information. It's easy to create and spread a hoax online and an audience can only be so skeptical. This kind of hoaxing and unmasking happens a lot on Reddit. People pretend to have cancer, or make up a story around a photo, e.g., "my 5-year-old daughter made this" in order to garner greater attention or sympathy. As good as digital media is at spreading lies, it seems equally equipped to unmask them. Crowd-sourced Sherlocks put pieces of evidence together to determine the truth of any of these stories.

The situation raises a couple of questions:

Can an online information consumer really spend the energy being that skeptical about all the information he/she encounters online, all of the stories like this one? The signal-to-noise ratio isn't quite so low that people will stop believing certain sources, but that could happen. But with a site like Gawker, I'm not convinced the average reader, after having read a fake story on the site, would be any less likely to believe or go to the site for information. But perhaps that is because Gawker readers seek a certain kind of information, which leads me to my next question.

Does it really matter if stories like these turn out to be untrue? Does that erase all of the meaning? It matters if a story about an attack on the White House turns out to be fake. But these morality play stories go viral in part because they spark a conversation about morals and behavior. The conversation and the thoughts shared about the topic still seem valid to me. Perhaps there is news and there is gossip. Both serve a function, and with the latter, we're more tolerant of the occasional falsehood.

----------

After listening to an excellent podcast in which Chuck Klosterman, Bill Simmons, and Chris Connelly discussed what we know about the Kennedy assassination 50 years after it occurred, I had another question related to truth and lies in the digital age: are we, in general, developing a more complete understanding of the world around us, or are we somehow moving further away from that? We have better tools for recording, seeing, hearing, and analyzing. We should be better at answering questions about what happened. But those tools developed at the same time as the hoax-spreading machine known as the internet.

There is some theoretically knowable answer to the question of who killed Kennedy and why. But information and knowledge do not move in tandem. William Gibson once said that the future is here, but that it was just unevenly distributed. Maybe knowledge of the world around us is like this. Those with a certain orientation to the technology are moving closer to this knowledge while others move further away.

Post script: There was an interesting article on the Reuters blog about this topic. It features quotes from Jonah Perretti (BuzzFeed) and Nick Denton (Gawker) and is definitely worth reading. Here's a terrific quote from it: "the reasons that people share basically have nothing to do with whether or not the thing being shared is true." The article acknowledges the two types of information circulating - news and info that let's people feeling "fleeting instances of comfort or joy" that are more difficult to verify. But even if you're dealing in the latter, if you repeatedly pass on stories that draw some of their appeal from being real (and let's face it: these little gossipy stories make for sub-par fiction) that turn out to be untrue, you become like the friend who's always telling tall tales and loses all credibility. Not as bad as the news organization that loses credibility, but still, I think people will seek out reliably true "gossip"-type news, shunning sources that get repeatedly duped.

What is Wikipedia now?

I was searching for information on a current event for class discussion: the Sandy Hook Elementary School shootings. Before, I was assigning a reading to another class: the wikipedia entry on Bandura's Bobo Doll study (I didn't feel as though students were ready to read an academic article from the early 60's, and I thought that the wikipedia article did a good job of distilling the essence of the article and doing so in a manner that most undergrads could understand and remember it). I felt a bit guilty about using Wikipedia in these ways. I wasn't holding up Wikipedia as an object of study (how is it created? Who creates it? How trustworthy is it relative to other sources?), but instead using it as if it were a legitimate source of information.

Upon reflecting on these feelings of guilt, I thought about how Wikipedia has probably changed over the past few years, since gaining prominence. Initially, the question was "is Wikipedia as reliable as published sources?" The assumption was that, because it is open to be edited by anyone, it could never be as reliable as published, vetted sources. As far as I know, it is still as open to change as it was back then (though perhaps the structure of correcting errors and detecting bias has been improved), but I think its wrong to assume that openness determines the extent to which any document is trustworthy. The factor that people should be paying attention to is motivation to introduce bias.

Think about other sources of information on the web. Is there some motivation to introduce bias to the information? In most case, there is some motivation. Maybe it is to please corporate shareholders, or to retain a certain audience so that they can get more views within that niche and generate profit that keeps them afloat. With Wikipedia, there are some particular topics that are edited frequently by interested parties, but as I understand it, the system has a way of detecting those and flagging those. But for my purposes, for using it in class as a way of getting the facts straight about Sandy Hook, or learning about Social Learning Theory, doesn't it present a much more even-handed, complete summation of these entities than any other available source on current events?

Maybe it wasn't always this way. There might have been a contingent of people adamant on shaping public perception of some event or entity that systematically altered the Wikipedia entries on those things. That was when there was some public conversation about the trustworthiness of Wikipedia entries. But now that that conversation has dropped out of the news cycle, out of the public consciousness, are those interested parties still as numerous, and as interested? These things have implications for the true trustworthiness of the content on Wikipedia. I am skeptical that people intent on fooling people retained their passion for altering Wikipedia entries, or, in that passion, eclipsed the Herculian efforts of volunteers working for Wikipedia to keep it free of disinformation and bias. But who knows?