Happily ever after?

You know that feeling when you're so deep into a book or a drama boxset that it feels like you've actually entered the fictional world and feel the emotions of the characters? It's a powerful effect of storytelling that psychologists call transportation.

There's obviously a whole spectrum of transportation – from being fully immersed in a whole 'cinematic universe' to almost imperceptible changes in your opinions as you read a short article or watch a social media clip. Researchers find that stories are not only immersive, but can help people understand and remember information, such as a study in the United States on helping doctors remember prescribing guidelines. 

In fact, the mind-altering effects of narrative can be so strong, it's not surprising that there is a whole field of research (and marketing practice!) in 'narrative persuasion'. And that has been a topic for this magazine over the years and recently, with several pieces in the September issue including Marcus Munafo on whether a good story gets in the way of the truth.

But does the power of stories have a darker side?

The story-drug

Studies repeatedly show that embedding an argument (such as the reasons to buy a particular product) inside a story structure can make weak cases more persuasive: it's as if our brains engage with the storyline at the expense of weighing up and interrogating the facts. Narratives can even encourage us to believe misinformation, because a wrong fact within a story can make it seem to 'all make sense', whilst evading our mental scrutiny. And the more emotion and empathy generated by the narrative, the stronger the story-drug is.

I know these things myself: I've switched off dramas where I can't find a character to empathise with, and as a documentary-maker I have used the power of music, images, and personal stories to convey uplifting triumphs and tearful tragedies.

But then I discovered that there are some times – such as when I try to communicate facts and scientific evidence to help people make their own decisions – when the downsides of narratives might outweigh their benefits.

For example, imagine you're at the doctor's and facing a decision about health treatment. The doctor cannot (legally or ethically) persuade you to take one treatment over another. You can take their advice, but they have to be able to give you the risks and benefits of your options in a non-persuasive way. There are lots of information leaflets designed to do this, and one common way to make them more digestible is to include 'patient stories' – anonymised quotes telling a story of someone's personal experience with the treatment. Naturally, these are even more engaging if they come with a photograph, and some background about that person… to turn it into a narrative, with a character that people can identify and empathise with…

You can see the problem. The more you identify and empathise with a particular character and their story, the more persuaded you could be about what they are saying. If this character seems like you, and had a positive experience of a treatment, some research (such as this study from Korea led by Youllee Kim, and a paper titled "Hey, that could be me": The role of similarity in narrative persuasion from Joëlle Ooms and colleagues in the Netherlands) suggests that you might be more likely to think positively about that treatment – and vice versa if they are telling of a negative experience.

When we communicate factual information to those who have decisions to make, we need them to engage with it, to remember the facts – but we also want them to be using all their critical faculties, and not to experience suspension of disbelief and strong emotions.

The journey of discovery

So does that mean that storytelling is thrown out of the researcher's or evidence communicator's toolbox? Together with my research team at Cambridge University, we decided to see if we could construct a narrative that harnessed the benefits of storytelling (being engaging and helping people remember the facts) without being persuasive – testing it against a bullet-pointed list of the same information.

There are, of course, many ways to structure a story. One of those that seemed most promising was the 'journey of discovery, or 'process narrative'. In this, the central protagonist – the writer – takes the reader with them on a story about uncovering the evidence on which to base a choice, rather than experiencing the outcome of a choice personally. This allows the story naturally to include both the evidence for and against making the choice, leaving the ending open to the reader's own decision.

We set the experiment up online, inviting over 1,300 people across the UK to take part. They were asked to read information about a fictional new drug being assessed for licencing, which might help sufferers of a rare (fictional) disease but which carried the very rare side effect of blindness. Each person was given only one version to read: either the narrative version or the bullet-point version.

Once people had read the information, they were tested on how well they remembered the facts in it, asked how engaging, persuasive and immersive they found it, and asked whether they would licence the drug if they got to make the decision.

The result? 

People found the story version more engaging and immersive, and remembered the information slightly better, whilst not rating it as more persuasive. So far, so good – it seems that, used carefully, we can reap the benefits of narrative even when we want to avoid persuasion.

The trick, though, is in being careful. People who read the narrative version were less likely to say they'd licence the new drug and rated its risks as higher. 

Looking more carefully at why, I think the blame lies with a phrasing difference between the two versions. In the bullet points we used a typically official phrase about side effects: "risk of blindness" (followed in brackets by the percentage chance of blindness). When writing the narrative version, though, we said side effects "included blindness" (with the same percentage chance in brackets). Even though we gave the same percentage, I think the narrative version was more explicit. Some people left us comments about not knowing whether 'risk of blindness' meant that some people actually went blind. We can't be sure that this is the reason people reading the narrative version thought the drug was riskier, but it would make sense.

Walking a tightrope

It's a salient lesson. When you're trying not to persuade people, you are walking a tightrope. Every choice of word could affect your audience's emotions or their understanding, and perhaps their decision – just as it does when you are crafting an advertising campaign. For those of us who try to produce information to support other peoples' decision making, empirical research like this is vital to help us understand potential pitfalls.

It's early days in research into the effects of narrative, and it can be a hard thing to study (as we found out!). Overall, though, it seems that story structure is a flexible enough tool to suit information-transfer as well as persuasion. As researchers we are probably only just stumbling across things writers have long capitalised on: that we can open people's minds through stories, to invite them on a journey to discover evidence and reflect on it for themselves, not to merely follow our story blindly to a conclusion we have chosen for them.

• Dr Alex Freeman led the Winton Centre for Risk & Evidence Communication at the University of Cambridge, and is now a crossbencher in Parliament. [email protected]