Persuading the unpersuadable

Science blogging: The essential guide - Christie Wilcox, Bethany Brookshire, Jason G. Goldman 2016

Persuading the unpersuadable

Deniers, Cynics, and Trolls

Melanie Tannenbaum

A major difference between science blogging and traditional science writing is the presence of a “comments” section in blogs. In this chapter, Melanie Tannenbaum, psychologist and blogger at Scientific American Blog Network, offers advice on how to maintain a civil and productive comments section, and discusses psychology-based strategies for persuading—and calming—your more quarrelsome audience members.

Ah, the comments section—either the most loved or most hated part of a blog, depending on whom you ask. Comments can offer opportunities for a fulfilling discussion, or they can be cesspools of spam, trolls, and science deniers. In this chapter, I will suggest some ways to maintain a respectful comments section, and point to some strategies from the annals of psychological science that may help to get your more disagreeable readers on board. Science bloggers, after all, ought to take a scientific approach to managing their comment streams.

Respectful commenting

One of the biggest ways that the people around us exert influence on our behavior is through the use of “norms,” messages about what constitutes acceptable and appropriate behavior in a given setting. There are two main types of norms: descriptive and prescriptive. Descriptive norms simply describe the way that things are, whereas prescriptive norms offer a mandate about how things should be. For example, if I said that most college students go to class wearing jeans and sweatshirts, that would be a descriptive norm. If I said that you should wear jeans and a sweatshirt in order to fit in, that would be prescriptive.

Descriptive norms can be incredibly powerful. For example, in a classic study, Robert Cialdini and colleagues manipulated the signs that were displayed in Arizona’s Petrified Forest National Park, a site plagued by tourists who would take fragments of petrified wood home as souvenirs.1 In situations like this, the first inclination of well-meaning environmentalists might be to set a strong prescriptive norm, perhaps by writing on the sign something like, “Many past visitors have removed the petrified wood from the park, changing the state of the Petrified Forest. This is bad, don’t do this.” The idea here would be to invoke a sense of shame and severity before asking visitors to refrain from taking the wood.

But read that prescriptive message once again. That message is not just telling you that you shouldn’t take the wood—it’s also telling you that most other people do. In fact, people were actually more likely to steal wood from the forest when they saw the sign telling them how many people tend to do so. But when the researchers made the message read “The vast majority of past visitors have left the petrified wood in the park, helping to preserve the natural state of the Petrified Forest,” the thievery plummeted.

We don’t care so much about what we should do. We care about what other people do. And then we really, really care about not being different.

When bloggers think about how to moderate comments sections, they often instinctively try to set prescriptive norms: commenters should be respectful, you should use appropriate language, you cannot level personal insults at the author or at other commenters. Yet based on everything that we know about descriptive norms, it seems much more likely that the best way to encourage good behavior in a comments section is instead to model what good behavior looks like. This can involve stringently moderating comments sections so that disrespectful ones do not even appear, or including a note in the comment policy about how the author appreciates that “most people” contribute thoughtful, respectful comments.

It is also important to consider that the people with whom you are arguing in your comments are not the only ones affected by behavior in your comments section. As many others have noted, when you argue with a “troll” in your comments section, you are not just trying to persuade that person—you are also working to correct misinformation and persuade all of the silent readers observing the interaction.2 This is no minor task, and your power here—and the potential negative power of your trolls, if left unaddressed—should not be understated.

Research has shown that hearing the same opinion stated over and over can lead people to overestimate the prevalence, popularity, and even accuracy of that statement, even if the repetitions are all actually coming from the same source.3 Not only that, messages received from unreliable, untrustworthy sources can become more persuasive over time, as people forget where they heard the information (thereby forgetting to “discount” it as untrustworthy) and simply remember the content of the comments themselves.4 It is very important, then, to model good behavior, correct misinformation, and address nasty commenters.

Nip trolls in the bud: The science of persuasion

One of the most fundamental models of persuasion research is Richard Petty’s and John Cacioppo’s Elaboration Likelihood Model, which states that people with varying levels of motivation or ability to attend to a message will take different routes to persuasion.5

Someone with a high level of motivation and/or ability—who finds the topic personally relevant, is knowledgeable in the domain being discussed, or is learning about the topic because she or he is responsible for some important relevant outcome—is likely to take what’s known as the “central” route to persuasion. People taking this central route are more likely to pay attention to the arguments made within the message itself, and so will be persuaded by strong, high-quality arguments and dissuaded by weak arguments. Writers tend to focus on these readers, and on strong, higher-quality arguments, when trying to convince readers of their point of view.

By contrast, someone with a low level of motivation or ability—someone who is tired, distracted, or clicking between several Internet tabs—would be more likely to take the “peripheral” route. This involves persuasion through “heuristics,” or automatic mental shortcuts that we often use to make decisions when we don’t want to think too hard. Some classic examples include appealing to consensus (“four out of five doctors recommend this product”), using attractive or famous message sources, or simply appearing to have a large number of long arguments.6

Image

A basic outline of the Elaboration Likelihood Model, adapted from Richard E. Petty and John T. Cacioppo, “The Elaboration Likelihood Model of Persuasion,” in Advances in Experimental Social Psychology, vol. 19 (Waltham, Mass.: Academic Press, 1986), 123—205.

Broadly, what this means is that the same persuasive strategies will not work with every audience. We must acknowledge that many readers—especially on the Internet—are not particularly motivated to pay close attention to what we are writing, and they are often tired, distracted, multitasking, or otherwise splitting their attentional resources. This makes them more likely to be persuaded via the “peripheral” route. That is, if they see a message coming from a well-known celebrity, if it looks like there are a lot of arguments (but they don’t necessarily read those arguments to see if they’re actually good), or if they think that others tend to agree with the point of view being presented, they are more likely to accept the message itself, even if the actual arguments being presented are not particularly convincing.

You can use this model to your advantage in one of two ways. First, if you have high-quality arguments and you really do want your audience to pay attention to them, you must try and do everything you can to increase your audience’s motivation and ability to pay attention to your message. And second, understand that many of your readers will likely be unmotivated and/or largely inattentive, and will opt for “peripheral” rather than “central” route strategies. As counterintuitive as it may seem for an inattentive audience, writing a long piece (that places your main argument in the headline or the lede) may actually work.

“Foot in the door” and cognitive dissonance

Any good salesperson will tell you: if you’re trying to get someone to agree to something big, start small and gradually raise the stakes. Once you get someone to agree to a small, easy request, he or she will be more likely to continue agreeing to successively larger requests. If you know that you’re trying to convey a message that will be difficult to swallow, start small—don’t plop the big message right there in the lede.

Why does this strategy work? Primarily because once people have started a pattern of “agreeing,” it’s hard to shift gears and start disagreeing—it’s much easier to simply continue down the already established path of agreement. According to psychological research on cognitive dissonance, people hate seeming like “flip-floppers,” whether that means acting differently from how they have acted in the past or behaving in ways that conflict with existing attitudes.

Is it ridiculous that changing your mind in the face of evidence could be interpreted as a weakness? Yup. But unfortunately, it’s a hallmark of how our stubborn brains tend to process information. It’s just easier for people to adopt a new opinion without actually “changing” their beliefs or sense of right and wrong.

Knowing this as a blogger means realizing that a better strategy than trying to change readers’ minds or beliefs outright may be to find ways to make your point of view congruent with their existing convictions. Let’s say that you are against the death penalty, and you are trying to convince someone else to adopt your opinion. If your opponent is arguing that the death penalty is right because it’s “morally wrong” to make the taxpayer suffer the burden of paying for a criminal’s life in prison, pointing out that the death penalty costs more is actually congruent with that conviction. (Calling attention to the Eighth Amendment and appealing to your opponent’s sense of right and wrong by saying that the death penalty constitutes “torture,” by contrast, would likely not be successful.) But if your opponent is inclined to care about the issue because he or she thinks that it is immoral to keep someone in prison for life because it’s more humane to kill him quickly and swiftly—and that is why he or she supports the death penalty—then pointing out how executions themselves can be considered a form of torture (and your arguments for why this is the case) could very well be an effective strategy for persuading your opponent to shift his or her view. Any approach that makes it easier for someone to change his mind without actually appearing like he’s really “changing his mind” will be more successful.

There is no one hard-and-fast way to guarantee that you will persuade every reader. And sadly, there will always be trolls. But by perpetuating descriptive (rather than prescriptive) norms about good commenter behavior, you can shift the tone of your comments section in a friendlier direction. In addition, by targeting the best types of arguments for your audience, easing any disagreeable readers into agreeing with you, and allowing readers to change their minds without really “changing” their minds at all, you can hopefully nip some of these bad behaviors in the bud.

Moreover, by invoking the aforementioned power of norm-setting, you can build a community that values active commenting, so you don’t always have to be the one answering all of the comments on your own. And about communities—take a look at all of the contributors in this book. The science blogging community is an active, communal, and supportive one. It can sometimes be helpful psychologically to switch from an independent mindset to an inter-dependent one, especially if that will help you appreciate and learn how to rely on one of the best resources that you’ve got—your fellow bloggers. So stop. Breathe. Step away from the Internet. Hopefully with some of these tricks in mind, engaging with your cantankerous commenters does not have to be so awful—it might even become a new, fun kind of challenge. Just don’t let the trolls get you down.

Melanie Tannenbaum is an award-winning social psychology instructor who taught at the University of Nevada, Reno. She blogs at PsySociety, which is part of the Scientific American Blog Network. Her writing has also been featured in Scientific American, In-Mind Magazine, and The Open Lab Anthology: The Best Science Writing Online, 2012.

Melanie is based in the San Francisco Bay area, California. Find her at her website, http://www.melanietannenbaum.com, or follow her on Twitter, @melanietbaum.

Notes

1. Robert B. Cialdini et al., “Managing Social Norms for Persuasive Impact,” Social Influence 1, no. 1 (2006): 3—15.

2. David Shiffman, “Dawn Take You All: Bilbo Baggins’ Approach Is Better than ’Don’t Feed the Trolls,’” Southern Fried Science, May 21, 2014, http://www.southernfriedscience.com/?p=16389.

3. Kimberlee Weaver et al., “Inferring the Popularity of an Opinion from Its Familiarity: A Repetitive Voice Can Sound Like a Chorus,” Journal of Personality and Social Psychology 92, no. 5 (2007): 821—833.

4. G. Tarcan Kumkale, “The Sleeper Effect in Persuasion: A Meta-Analytic Review,” Psychological Bulletin 130, no. 1 (2004): 143—172.

5. Richard E. Petty and John T. Cacioppo, “The Elaboration Likelihood Model of Persuasion,” Advances in Experimental Social Psychology 19 (1986): 123—205.

6. Understandably, the terms “central” and “peripheral” can be confusing, especially since in the diagram the routes themselves are both linear, not “centrally” or “peripherally” located in some way. These terms refer to the parts of the message that the recipient is paying attention to, not the position of the route itself. Participants taking the “central route” are paying attention to “central,” fundamental aspects of the message; participants taking the “peripheral route” are paying attention to more tangential aspects.