We often hear that storytelling holds the power to sway elections, shape health outcomes, and even recruit people into extremist movements.1 Academia, politicians, and businesses alike all attempt to use narratives to their advantage. The catch is, scientifically speaking, there is still a lot of uncertainty about whether communicating a narrative will actually persuade your audience. Despite its widespread use, a key question remains: Does telling a story actually change what people think, feel, or do?
In a 2016 meta-analysis published in Communication Monographs, researchers Kurt Braddock and James Price Dillard examined 74 studies to determine whether exposure to a narrative produces a “story-consistent” change in belief, attitudes, intentions, and behaviors, or not.2
In short, Braddock and Dillard found that exposure to narratives influenced beliefs, attitudes, intentions, and behaviors in ways that aligned with the story’s message. Although the study was not conducted with behavioral scientists specifically in mind, many of its insights are relevant to our field, particularly if narratives can be shown to produce similar effects in applied behavioral interventions.
But it’s important to retain some skepticism and not forget nuance. While the study reports positive overall effects, the role of fictionality was more complex, specifically regarding the different effects of fictional and nonfictional stories.
Background
Who are the researchers?
Kurt Braddock, a professor at American University, specializes in the persuasive tactics used by violent extremist groups for recruitment and radicalization. Along with his academic work, he has advised the U.S. Department of State and the Department of Homeland Security on counterterrorism strategies.3 His co-author, James Price Dillard, is a Distinguished Professor at Penn State whose research focuses on how persuasive messages evoke emotions and how those emotions, in turn, shape attitudes and behavior.4
The two academics were inspired to conduct this research after noticing that the literature on narrative persuasion was fairly inconsistent. Some studies suggested stories could inspire opinions and promote behavior change, while others reported zero effect on behavioral outcomes. Naturally, they set out to answer the contentious question: Can narratives really persuade?
Some quick definitions
For the purposes of the study, it’s important to understand how the researchers define a “narrative.” Braddock and Dillard consider a narrative to be a “cohesive, causally linked sequence of events that takes place in a dynamic world subject to conflict, transformation, and resolution through non-habitual, purposeful actions performed by characters.”6 Put simply, a narrative is a story told in any format (film, literature, or audio) and shows a sequence of events, where characters take purposeful actions to overcome conflict.
As mentioned, the study explored how storytelling shapes four key constructs: beliefs, attitudes, intentions, and behavior. Beliefs are the ideas we accept as true, like “smoking causes cancer.” Attitudes reflect how we feel about something, for example, having a negative view of smoking. Intentions show our plans or motivation to act in a certain way, while behaviors are the actions we actually take, such as quitting smoking.
Methods
To test these hypotheses, Braddock and Dillard conducted a meta-analysis—a method that combines results from multiple studies to identify overall trends or patterns in the research.5 They surveyed all the existing research on the impacts of narratives in an extensive literature review of 6,427 studies. From these, they selected 74 that measured beliefs, attitudes, intentions, or behaviors and compared narrative to non-narrative (control) interventions.
They combined and analyzed these studies to measure how narratives (independent variable) affected outcomes like beliefs and behaviors (dependent variables). To measure the effects, they used Pearson’s r, a statistic that indicates the strength and direction of the relationship between two variables. Overall, Braddock and Dillard found effect sizes ranging from 0.17 to 0.23, which indicate a small but meaningful influence of narratives.
Key Findings
Beliefs
The researchers analyzed data from 37 studies involving 7,376 participants, focusing on how narratives influenced beliefs. They found a small but meaningful positive impact of narratives on beliefs (r′= 0.17).
According to their data, nonfictional narratives significantly affected belief change, while fictional narratives did not show a statistically significant effect. This is consistent with more recent studies. For instance, Dubourg and Baumard, researchers from the École Normale Supérieure, reviewed contemporary literature on the subject and found mixed results regarding fiction’s effect on beliefs. They concluded that, until clearer evidence emerges, fiction doesn’t impact beliefs.7
These results suggest that it could be a good idea to privilege nonfictional narratives when the goal is to reliably influence beliefs, especially in contexts where credibility and trust in the source are important.
Additionally, Braddock and Dillard found no difference in the effects of narratives on beliefs as a function of the medium (audio, text, visual, and theatrical), nor were the positive relationships found between narratives and beliefs explained by the belief type (world beliefs, efficacy beliefs, normative beliefs, and expectations).8
Attitudes
Interestingly, when it came to attitudes, Braddock and Dillard found the opposite. It was determined that fictional narratives are an effective means of inducing attitude change, but nonfictional narratives are not.
Regarding the medium, they found evidence suggesting that video-based and theatrical narratives significantly drive attitude change.9 These results suggest that people should privilege fictional narratives through visual media to influence attitudes.
Intentions
Additionally, they found that narratives change intentions, with no significant difference between the effects caused by fictional or non-fictional narratives, nor did any medium prove superior in influencing intentions.10
Behaviors
In this analysis, the researchers examined how narratives influenced behaviors, based on data from five studies involving 978 participants. They found a moderate positive effect (r′w=.23), meaning that narratives have a measurable impact on behavior. This effect was consistent across studies and is unlikely to be due to chance. Since there was no unexplained variation, there was no need to explore additional factors that might influence the relationship between narratives and behavior.11
All in all
To sum up, Braddock and Dillard found evidence that narratives influence four key indices of persuasion: beliefs, attitudes, intentions, and behaviors. This result is important on its own since narratives are frequently used by policymakers, communicators, non-profits, etc., to advocate for change across a wide range of issues. There is an opportunity for behavioral science, often tied to nudges and messages as the quintessential interventions, to include narratives as a tool for change.
behavior change 101
Start your behavior change journey at the right place
At the intersection of narratives and applied behavioral science
When we connect this research to real-world examples, we see that applied behavioral science and narrative-based change successfully intersect in some forms of “edutainment,” which have already shown proven impact.
A good example is a Randomized Control Trial conducted by researchers from Bocconi, MIT, and the World Bank, in collaboration with MTV. Together, they tested whether a television show, MTV Shuga, could help raise awareness around HIV and promote safer sexual behaviors among youths in Nigeria.
Young viewers were exposed either to MTV Shuga or to a control TV series. Among those exposed to MTV Shuga, the researchers found significant improvements in knowledge and attitudes toward HIV and risky sexual behaviors. For instance, participants watching MTV Shuga were twice as likely to get tested for HIV eight months after the intervention as those who watched the “placebo” TV show. They also found a reduction in STDs among women in the treatment group.12
Based on this evidence and the study by Braddock and Dillard, this is an intervention direction that behavioral science should explore further. The exact narrative techniques and mechanisms behind these changes haven’t been pinned down yet, making it hard to build a clear theory or toolkit for using stories in behavior change.
Limitations
A significant issue was that the researchers didn’t focus on the lasting effects of narrative persuasion. Even if narratives can produce real-time alterations in our beliefs, attitudes, intentions, and behaviors, we don’t know if these changes are sustained in the long run.
Also, since the researchers assumed a direct causal relationship between narratives and change, the paper doesn’t provide much insight into the psychological or sociological mechanisms that drive change. We don’t know whether the effective mechanism is evoking emotions through the stories, showcasing positive social norms, or fostering empathy and identification with key characters. The good news? There’s still room to explore in this area.
Conclusion
So, can storytelling change us? Braddock and Dillard’s study demonstrates that exposure to narratives can shift people’s beliefs, attitudes, intentions, and behaviors to more closely align with those presented in the narratives.
While this may sound like we’ve solidified the efficacy of narratives, more research is needed to understand the underlying cognitive mechanisms and the narrative tools and strategies that enhance or inhibit their strength.13 Nevertheless, the basic relationship and the causal association were explored through this meta-analysis, including 74 papers and thousands of participants.
The results were ambiguous with regard to the question of fictionality versus non-fictionality, but, in the case of belief change, the data suggests that it might be better to use non-fiction-based interventions. Finally, the results support the view that the persuasive power of narrative is less influenced by the medium of presentation, whether that is audio, text, or video, but including visual media may be slightly better for changing attitudes. For behavioral scientists and practitioners, narratives could turn out to be one of our best untapped tools—if the research holds up, they could reshape how we design interventions.
