Increasing Suspense by Playing with Structure

We make plenty of choices about the way we structure a story. Do we tell the story in linear order? How much do we give away? How does that affect the reading experience? A recent study had some interesting things to say about this.

Before we get into the actual experiment, let's talk about Structural Affect Theory. It's the idea that emotional reactions to a story are influenced by the story’s structure.

For example, most stories include an initiating event (a butler poisons the Lord’s wine) and an outcome (the Lord dies).

Varying the placement of these pieces in the narrative results in different emotional responses. For example, you could increase suspense by adding intervening scenes between the initiating event and the outcome.  For example, adding a scene where an innocent bystander almost drinks the poisoned wine would introduce the possibility of collateral damage and increase the suspense.

Or, if you want to surprise the reader, you could do away with the initiating event entirely-- the Lord drinks the wine and drops dead out of the blue.

Got it?  Now on to the specific experiment.

Researchers Anneke de Graaf and Lettica Hustinx wanted to know whether a suspenseful story would be better at drawing the reader into the story.  Also, they wanted to know whether a suspenseful story would convince readers to hold more beliefs consistent with the story.

In the experiment, the researchers used a story called Dance of the Spirits by Ton van der Lee about a European man who immigrated to Mali. The story consisted of four parts (listed here in chronological order).

1. Tony sits on the roof of his house in Mali and contemplates the difficulties he had in Africa.
2. Tony contracts malaria
3. Tony goes to a Western hospital, where he is not cured.
4. Tony goes to an African healer, who finally cures his malaria.

The researchers created two versions of the story. The suspense version had the order described above. The non-suspense version had parts in the following order.

2. Tony contracts malaria
4. Tony goes to an African healer, who cures his malaria.
1. Tony sits on the roof of his house in Mali and contemplates the difficulties he had in Africa.
3. Tony goes to a Western doctor., where he is not cured.

In this version, readers find out what cures Tony right away, and the story is therefore less suspenseful. After that, parts one and three are told as flashbacks. The wording of the stories was the same except for a few transitional phrases and tense changes in the flashbacks.

Participants read one of the versions and then answered questions about how compelling the story was. They also answered questions about beliefs related to the story.

The researchers found that participants in the suspense condition were more emotionally engaged They were more likely to agree with statements such as "During reading, I felt tension about how the story would end." They also reported being less distracted by other things while reading the story.

And what about suspense’s effect on beliefs? Turns out that participants who read the suspense structure were more likely to report beliefs consistent with the story. For example those participants were less likely to believe that western hospitals are successful in curing malaria.

There's one caveat to be aware of here, and it's that the experiment only used one story. The effect definitely needs to be tested with other stories as well. But I thought this framework was a interesting way to think about story structure and suspense.

Readers, what do you think? How important is suspense for keeping you hooked?

P.S.  Thanks to Raymond Mar from On Fiction for sending me this article.

Hope you enjoyed this post! To get regular updates from this blog, use the subscription options on the sidebar.

Also, two of my author friends recently released debut novels.  Katie Cotugno just launched How to Love, and Christa Desir released Fault Line.  Check them out below.

Before: Reena Montero has loved Sawyer LeGrande for as long as she can remember. But he's never noticed that Reena even exists . . . until one day, impossibly, he does. Reena and Sawyer fall in messy, complicated love. But then Sawyer disappears without a word, leaving a devastated—and pregnant—Reena behind.
After: Almost three years have passed, and there's a new love in Reena's life: her daughter. Reena's gotten used to life without Sawyer, but just as suddenly as he disappeared, he turns up again. Reena wants nothing to do with him, though she'd be lying if she said his being back wasn't stirring something in her.
After everything that's happened, can Reena really let herself love Sawyer LeGrande again?

Ben could date anyone he wants, but he only has eyes for the new girl—sarcastic, free-spirited Ani. Luckily for Ben, Ani wants him, too. She’s everything Ben could ever imagine. Everything he could ever want.
But that all changes after the party. The one Ben misses. The one Ani goes to alone.

Now Ani isn’t the girl she used to be, and Ben can’t sort out the truth from the lies. What really happened, and who is to blame?

Ben wants to help Ani, but the more she pushes him away, the more he wonders if there’s anything he can do to save the girl he loves in this powerful, gut-wrenching debut novel.

A. de Graaf, & L. Hustinx (2011). The effect of story structure on emotion, transportation, and persuasion Information Design Journal DOI: 10.1075/idj.19.2.05gra


  1. Interesting! When I saw your tweet, I assumed it was going to be a non-linear structure. That can be good for suspense, too, of course. I'd also like to see other orders of the story events, such as starting with scene 3, then 1 and 2, finally ending with 4, etc. Putting the resolution of the story's conflict (4) anywhere but at the end is sure to undercut suspense, so it would be interesting to see how playing with the other elements affects the suspense condition.

    1. Good point, Jordan. I think the order matters a lot, and the skill in weaving non-linear structures back together.

  2. I'm thinking about playing with structure a little in my next book. I just read V.E. Schwab's Vicious, which moves around in time, and I thought it might be fun not to write linearly. I think there are plenty of good ways to play with structure and increase suspense, unlike the non-suspense version of that story.

    1. Hey, I'm experimenting with a non-linear structure now too! It's fun, and I end up thinking less in terms of linear causility and more in terms of information to reveal.

  3. I'm inclined to think they study was silly. Any moderately prolific reader could have told them this.

    1. Jo - This study seems silly because the results line up with intuition, but that's not always the case. For example, I also blogged an article a while back that found that spoilers did *not* increase overall enjoyment of a story. Definitely a nonintuitive result, so it's always good to test! Trying to reconcile this study with the spoiler study would also be interesting.

  4. Lee Child said that a good story sets up a question in the reader's mind in the opening pages that the reader must read to the end to find the answer. Mysteries are based on this premise, but even classics like The Great Gatsby do the same thing. In Gatsby, the reader wants to know who Gatsby is from the opening to the ending. Some form of suspense is what drives all good books.

    1. Very true, Seeley. Suspense requirements vary from genre to genre and reader to reader though. I find that I need a fairly high amount of suspense to stay interested in a story.