Novelists, Inc. Ninc Home Ninc Facebook Ninc Twitter

Ninc Newsletter

More Proof of
the Power of Words


A recent study by Ohio State University researchers indicates that what people read can influence their behavior.

No kidding. <Yawn> We know all about it. Next topic.

Well, maybe we don’t know all about it. The study, which will be published in the Journal of Personality and Social Psychology, delves into the finer points of “experience-taking,” which is the term the researchers coined for that sense of losing yourself, as a reader, in the world occupied by the characters in a novel. “Losing yourself,” it turns out, might be more than just an expression. You might just lose a bad habit. (We’ll ignore the possibility that you might lose a good habit, natch.)

The researchers found that, under certain circumstances, taking on the same emotions, beliefs, thoughts, and internal responses as a character can change a reader’s behavior, at least temporarily. The researchers note that the “circumstances” conducive to losing oneself in a character are those in which readers are able to “forget about themselves…their own self-concept and self-identity while reading.” To test for this, some of the subjects read material while in a cubicle containing a mirror; presented with such an obvious reminder of the real world—their reflection—they did not achieve the “experience-taking effect.” Researcher Geoff Kaufman explained that “The more you’re reminded of your own personal identity, the less likely you’ll be able to take on a character’s identity….You have to be able to take yourself out of the picture, and really lose yourself in the book in order to have this authentic experience of taking on a character’s identity.”

Across the board, the experiments undertaken showed that stories written in the first person with characters resembling the subjects had a greater effect on the subjects’ later behavior than did stories written in the third person or that had characters that did not resemble the subjects.

In one experiment, readers who identified with a character who overcame obstacles, such as long lines, rain, or car problems, to vote in an election were “significantly more likely” to vote in a real election held a few days later than those who read about a character who had no obstacles to voting. The stories involved a student, in one case an Ohio State student and in the other a student from a different university. Sixty-five percent of the subjects (all OSU students) who read first-person stories about the OSU student encountering difficulties in voting reported that they voted in the real election a few days later. However, only 29 percent of the subjects who read the same story featuring a student from a different university reported that they voted.

In another experiment, male heterosexual subjects read one of three versions of a day-in-the-life story about a male character. In one version, the character was heterosexual. In another, the story revealed the character was gay early in the story. The third version revealed that information later in the story. Researchers said that knowing early on that the character was not like them prevented subjects from having the experience-taking response, whereas subjects reading the story in which the character was revealed to be gay late in the story had levels of experience-taking similar to those of the subjects who read about the heterosexual character. Additionally, in post-experiment surveys subjects who read the story with the late-gay revelation showed “significantly more favorable attitudes” toward homosexuals and relied less on stereotypes than did readers of both other versions. Readers of the late-gay story rated the homosexual character as less feminine and less emotional than readers of the early-gay narrative in part because, the researchers surmised, the subjects had accepted the character as being like them. Comparable results were seen in an experiment in which a character was revealed to be African-American early in the story and late in story.   

TOC ->  Page 6 ->