Tuesday 12 November 2013

Transportation, Persuasion, and the Construction of Disbelief

ANDY: Don’t you just love literature?

BENNY:  Um… I guess. Why?

ANDY: It’s great. I learn so much when I read.

BENNY: You mean from reading educational non-fiction?

ANDY: No – well, I guess that too. But I was referring to reading novels. A great novel has so many insights into life.

BENNY: I think so too. Do you think fiction can also convince people to belief false things about the world, however? You know, like propaganda.

ANDY: Propaganda is one thing, artful literature is another. It’s scientifically proven by a dude named Gerrig: when reading people momentarily suspend their disbelief and in doing so, they can absorb whatever useful lessons they like from the text, while ignoring all the bits that aren’t true. There’s a difference between fact and fiction.

BENNY: That’s really cool. Someone should like, totally write a blog post about this stuff.

* - * - * - * - *

When consuming fiction, readers lower their guard and become more likely to passively accept false statements about the real world such as “mental illness is contagious” and “chocolate makes you lose weight.” The classic metaphor of “suspension of disbelief” must be replaced with an alternate concept, the “construction of disbelief.” The default human reaction to new information is acceptance. Analyzing and determining whether this information is true or not only follows afterward. Disbelief is an active process of rejecting previously accepted information.

The way that we read varies with the genre of what we are reading. In reading non-fiction, people are in “Non-Fiction Mode,” expecting to encounter statements about the real world that may or may not cohere with their prior models of reality. In this state of mind, statements such as “mental illness is contagious” are examined critically and compared with prior knowledge. In reading fiction, however, people expect to delve into an alternate world, rather than to be educated about the real world. As a result, people let their guards down and are more likely to reproduce in-story errors as facts on later tests. Fictional knowledge is incorporated into real-world knowledge during reading and the two streams of knowledge may remain entangled afterward.

In mentally simulating fictional stories, read statements are vaguely stored in memory. Two weeks later, readers will still have that information in their heads but will no longer remember where it came from. The false information can feel even truer after a two-week gestation period than it does immediately after reading.

This is not to say that fiction transforms humans into passive recipients of fictional content, enabling a non-probabilistic hypodermic needle effect. But narrative devices such as suspense eliminate the additional careful attention people apply when they usually encounter information that runs counter to their real-world beliefs. Melanie Green and Timothy Brock identified transportation into a story as the process by which people’s high levels of engagement override their caution. During transportation, readers are especially vulnerable to persuasion from in-story messages. Narrative transportation is an exception to the two standard dual-mode theories of persuasion.

The elaboration-likelihood model of persuasion proposed by Petty and Cacioppo accepts two routes to persuasion: central and peripheral. The central route to persuasion is taken when a reader is highly attentive and motivated by self-interest to analyze the quality of arguments and the strength of the evidence supporting them – and yet gets persuaded anyway. This is what happens when people are persuaded by philosophical arguments, for example. The peripheral route to persuasion is taken when a less attentive reader relies on heuristics and cues as shortcuts for coming to a conclusion. The peripheral route to persuasion requires less strength of argument, but information believed through the central route is generally longer lasting and more resistant to counterarguments.

Slater and Rouner argue that the elaboration-likelihood model of persuasion (ELM) is incomplete because of its inability to account for how, in narrative transportation, information obtained through the peripheral route can be durably integrated into prior knowledge. They propose an “extended ELM” (E-ELM) that accounts for transportation. The E-ELM also considers identification with fictional characters to be a strong pathway to persuasion.

The most obvious way for media content to have mass effects on society is for it to persuade people to new beliefs that stimulate new behaviours. As Foy and Gerrig say, "Those works of literature that most effectively immerse their readers have the greatest potential to do both good and harm." It’s important to understand the ways that benevolent and malevolent media producers can plant information into readers’ heads. Rate of false-vs-true belief absorption might be one of the best criteria for determining whether a given artwork’s effect on society is largely positive or negative.


  1. An important note here may be the implicit/explicit discrepancy (Kahneman).

    Though we may not consciously believe fiction to be true, maybe it's affecting out expectations implicitly... outside our awareness.

    1. Yeah, those categories roughly parallel the peripheral/central distinction. Narrative devices are tailor-made to create flow experiences and two characteristics of flow experiences are (1) a loss of self-consciousness and (2) the forgetting of one's environment and complete submergence in the activity at hand. Both those things probably lead to System 1 taking over the role of correctly storing information.

  2. This is an important piece, Michael! Many of the paragraphs would be suitable as part of a Wikipedia article -- have you considered adding them there for a much wider audience? Or do you need ownership of this material for your thesis?

    I've tended not to read fiction because I feel like there's so much true stuff to read, why would I need to read false things? Most of the benefits of literature -- getting into other people's heads, learning about social norms, etc. -- can be acquired from memoirs, news stories, social interactions, interviews, etc.

    Hypothesis generation is important, and insofar as fiction is a form of doing that, it's valuable. But maybe the human brain is not so good at reading extensive hypotheses without believing them.

    1. I have no problem contributing to Wikipedia but I don't want to risk being accused of plagiarism or self-plagiarism.

      I think reading fiction does more good than harm but I don't think it's net better than reading non-fiction. But it depends what we're reading.

      There are a lot of benefits to reading fiction that one doesn't get from reading academic essays - I cover those in my post, The Psychology of Narrative Fiction. You're probably right that most of the benefits of fiction can also be acquired from memoirs and interviews though. I hadn't thought about that.

    2. Perhaps to avoid self-plagiarism you might have to remove the blog post? I don't know. Anyway, at least it's something to keep in mind going forward. :)