Bench Bite: What makes a science story engaging? It depends… on the reader.
TLDR: If you can emotionally immerse an audience, they will be much more motivated to learn about the science behind the story. But to do that, you have to know your audience. Why? A new study published in JCOM suggests that vivid imagery and rich characters don’t significantly impact perceived story quality… at least not directly or for everyone. Audience perceptions are key.
Storytelling is often lauded as the most powerful tool in the science communicator’s toolkit. Some have called it the soul of science communication because stories, arguably a uniquely human format for sharing information, make people care in ways that facts do not.
Many studies have demonstrated the power of narrative in communicating science. A 2014 literature synthesis on science storytelling by Michael Dahlstrom found that science information presented as narrative is easier to understand and remember, more engaging, and more persuasive. Yet not all “stories” are made equal. While many practitioners play with the art of science storytelling until they find something that works, communication researchers may wonder, what are the critical elements needed to make a story successful?
Peter Fischer is one of those researchers. He is a research associate at the Institute of Educational Psychology at Technische Universität (TU) Braunschweig, studying how scientists can communicate their findings more effectively (e.g. through stories). Fisher has been trying to pin down the exact elements necessary for a science story to immerse an audience.
Fisher figured that audiences need to perceive stories as good for them to work. And by "work," he envisioned a story that could engage readers emotionally and ultimately increase their interest in the science topic. “After all, who is immersed by a boring story?” Fisher said in a recent email response to my inquiry about a study he and colleagues recently published in JCOM on science storytelling. But this line of thinking led him to the question, what makes a good story?
“Few studies have addressed this empirically, so I looked into books by storytelling experts [for answers],” Fisher said. “Two elements emerged most frequently: vivid imagery and rich information about the protagonist's motivations and emotions. I set out to analyze these through an experiment.”
What is a Story?
Close your eyes and picture a storyteller. What do you see? I see someone like Mark Twain or one of his characters, perched at the end of a river raft with feet dangling in the flowing water, chatting about some wild adventure to a captive audience. Or an older woman dressed in colorful clothes, her bare feet hugging the dirt around a fire in the early evening, sharing the struggles and lessons of her ancestors. Or an astronaut returned home, trying to recount what they had seen and experienced.
From this exercise, you might be realizing how important vivid characters are to good storytelling. Not static characters, but people (or humanized animals or objects!) who have been on an adventure, faced obstacles, and learned to see the world or themselves differently as a result. It is this ability to learn from someone else’s experiences, to learn how to survive and thrive physically and mentally if you found yourself in a similar situation (without having to go through it yourself), that makes storytelling so powerful. This “flight simulator” or “survival guide” function of stories may be one of the core reasons we seek them out.
But how does a storyteller make a character and their journey relatable, engaging, and instructive? As Fisher points out, master storytellers often emphasize the importance of vivid imagery and rich information about the protagonist. One of the reasons I watch more modern TV series than I do movies these days is that series often take more time to develop their characters. Caring about the characters and feeling transported into their worlds and perspectives keeps me binging a show, while an action-filled movie plot can quickly bore me.
Unexpected Results
In an online experiment, Fisher and colleague Barbara Thies used ChatGPT to generate fictional 600-word stories about a neuroscientist or an astronomer containing related scientific facts. Fisher and Thies then modified these basic stories to add elements of narrative depth that might improve them, including vivid imagery, information about the fictional researcher’s emotions and motivations, or both.
A total of 570 participants read one of the basic or experimental stories and answered questions about their perceptions of the story quality, whether their interest in neuroscience or astronomy had increased, and how emotionally transported they were (e.g., “I want to know more about what happens next to the character.”)
But the study turned out to be more frustrating than expected. Even before collecting data, Fisher and Thies struggled to decide on the exact content of the experimental stories.
“A text-based stimulus involves so many variables,” Fisher said, which translates into many decisions with potential downstream consequences. For example, the authors had to decide whether it was acceptable for text length to differ between conditions, what topics to cover, and whether the protagonists’ gender would matter. Fisher and Thies ended up creating versions with either male or female protagonists (studying neuroscience or astronomy), but neither topic area nor gender made a difference.
“This design process took the most time and energy,” Fisher said.
Then came less-than-promising results. Fisher and Thies confirmed that readers rated stories with added narrative depth as appropriately higher on dimensions of vivid imagery and protagonist emotions and motivations. But these narrative elements had no significant effect on perceived story quality, and thus no effect on story transportation or science interest.
But insignificant results weren’t the last challenge. Fisher later developed long-term health issues following a COVID-19 infection. He was so fatigued that he wasn’t able to meet the initial revision deadline after submitting the manuscript covering this study to JCOM.
But Fisher persevered, in both his efforts to revise his manuscript and to rethink his study’s implications for science storytelling.
“Fortunately, the JCOM editorial board kindly extended my deadline,” Fisher said.
A Change of Perspective
The major finding of Fisher’s study was that experimentally manipulating elements to make a narrative richer – adding vivid imagery or detailed information about the protagonist's motivations and emotions – did not significantly influence audience engagement.
But perception of story quality did.
“Overall perception of story quality was a strong predictor of transportation (the emotional immersion created by a story),” Fisher said. “Transportation, in turn, was directly related to an increased interest in the presented scientific topic.”
In other words, audiences rated a story as higher quality when they felt more emotionally immersed in it, which translated into greater interest in the science it contained. Women and readers with higher education tended to report greater story transportation and topic interest – regardless of whether Fisher and Thies added narrative depth elements to the story.
What’s the lesson here for a science storyteller? Fisher thinks it may be more important to understand and cater to your audience (by creating characters they can relate to) than to simply add more “universal” vivid imagery and character depth.
“Know your audience,” Fisher recommends. “So far, it appears that recipient characteristics dictate emotional immersion much more strongly than the textual characteristics [of the story] themselves.”
Is Story in the Eye of the Beholder?
I know people who are turned off by science fiction and simply don’t find the fantastical “what if” scenarios behind sc-ifi stories engaging. It always boggles my mind, because I’m the complete opposite. If you want to engage me, give me an unlikely hero with fantastical powers fighting to prevent or survive a future apocalypse. The forward-looking, anxiety-prone, eternally optimistic, Virgo-minded justice warrior within me is naturally drawn to these characters.
The problem with storytelling research is that a “story” is quite difficult to define. In one of my favorite books on storytelling, “Wired for Story,” Lisa Cron describes story as a communication format humans are naturally drawn to and able to recognize. You know a good story, just like you know a song or painting that makes you feel something, as soon as you experience it. On the other hand, Cron points out that a good story is much more difficult to define and reliably produce by those unpracticed in the art form. And perhaps, just as in music and art, not everyone finds the same story, objectively “good” or not, enjoyable or relatable.
“When reading the literature on storytelling in science communication, I always felt that ‘story’ was defined rather loosely, describing vastly different formats across studies,” Fisher said. “Yet, these formats were often grouped together in meta-analyses, leading to mixed results.”
Going forward, it may be more important to figure out which science stories “work” for which people rather than identifying universally effective storytelling structures. Of course, there are likely some best practices (watch Kurt Vonnegut talk about the story shapes that underlie almost every engaging story ever told). But as in every type of art, rules are there at the beginning so that the masters can later break them.
Fisher is now studying different types of story formats and how they may serve different audiences and purposes. He has found that most stories used in science communication fall into four categories, which he describes as follows:
Research stories, describe the scientific process in a storytelling format.
Autobiographical researcher stories, share a researcher's personal journey and how they became interested in a topic.
Patient stories, used in health communication studies.
Fictional science stories, describe fictional events related to a scientific topic.
But these may be only the beginning. In our scientific efforts to categorize and define stories, we may miss those that bridge and merge these categories, that bring together diverse characters in the telling of science’s inspirations, methodologies, and impacts.
Storytelling Training for the Science Communicator
I don’t know that I was ever taught how to write a story. I think I learned how to tell stories by reading and listening to lots of master storytellers. That’s one of the big challenges in engaging scientists as storytellers — they are often trained in technical communication practices that subvert any natural storytelling skills they might have gained growing up reading sci-fi as voraciously as I did.
Engaging people’s emotions? Sharing the serendipitous and random insights journey that often underlies hypothesis-making? Developing relatable characters? Those are often a “no” in scientific writing.
In previous storytelling studies, Fisher tried to use real stories told by researchers as experimental stimuli, rather than AI-generated ones. But he struggled because the resulting stories were not always understandable or engaging to a broad audience and were difficult to standardize across experimental conditions. He ultimately turned to ChatGPT for better experimental control. I understand, but I would love to see if reading good stories from real scientists (and knowing they are real) has a stronger effect.
So, how can scientists and science communicators learn to tell more effective, emotionally engaging stories? Beyond practice, practice, practice, I might suggest more conversations with nonscientists about science while paying attention to elements that capture people’s hearts and imaginations.
Read the study: What makes a good story?