Summary: A story slider is a method for assessing and representing a person’s personality where users rate which of two characters they are most like. It’s a device that researchers can use for personalizing an app.
Kyle prepares a confetti cannon…
He’s getting ready for a friend’s birthday party which he’s been eagerly anticipating for weeks. At the event, Kyle converses with ease; going from person to person like an animated pet when guests arrive. He soaks up the attention from others as he smiles wide and sips his mango daiquiri.
So who is Kyle? He’s the quintessential extravert, a person who is the literal and figurative life of the party. By using stories like Kyle’s we can portray personality characteristics which enables us to build apps that send tailored messages.
That’s the idea behind story sliders. You’ll learn what they are, how they work, and why researchers are excited by their potential.
Let’s dive in.
Personality Traits Reveal a Person’s Persuasion Profile
Have you ever wondered how cheesy infomercials convince people to buy?
If so, what exactly is the difference between the person nodding along during the commercial and you? It comes down to persuasion profiles: a set of personal attributes that shape the kinds of messages people are likely to respond to. One person may be convinced by an ad’s appeal to authority, while another balks at the absurdity of using an athlete to sell a Snuggie.
You can see an example of a persuasion profile below:
On the y-axis are Cialdini’s principles of persuasion: A set of techniques for nudging people to take actions, two of which I describe below:
- Scarcity: “People want more of what they can have less of.” In other words, if people perceive that there’s a limited opportunity to take an action then people will be more likely to do it.
- Consensus: “People follow the lead of similar others.” Put another way, we tend to think, feel, and act in step with those who are like us.
A persuasion profile estimates how likely a person is to respond to each of the Cialdini principles. In the example above, the estimated effect of a scarcity appeal is high while an authority one is low. In practice, that would mean an ad telling this person they have only an hour to capitalize on a deal would do better than showcasing a celebrity endorsement.
How can we explain such variability? How can we make sense of persuasion principles, thought to be universal, differing so much in their effectiveness?
One solution would be to invoke the Big-5 personality traits. They are openness, conscientiousness, extraversion, agreeableness, and neuroticism (remember these with the acronym OCEAN). These characteristics describe how people tend to think, feel, and act.
Let’s hone in on conscientiousness. Those high in this trait tend to work hard, stay organized, and follow orders. Thus, we might predict they’d find an authority figure — like an expert in an infomercial — especially persuasive.
In other words, the Big-5 traits inform what people find convincing. Here’s what the means: if we know a user’s level of these traits then we can create persuasive appeals just for them, a surprisingly effective strategy.
Tailored Persuasive Appeals More Effective Than Generic Ones
In the beginning there was the spoken word.
It sparked the oral tradition which facilitated the spread of stories. Those tales were often told by a master around a campfire of listeners eager to hear how a drama would resolve. For thousands of years, that dynamic stayed largely the same.
Even after the dawn of the printing press writers would share their message to audiences that were at best segmented. That is, broken up into groups typically based on age, gender, and interest.
Communication has now changed. With the dawn of the information-age, personalized messaging is now possible. In other words, communications are made just for you, not a crowd. Here are some of the more impressive findings demonstrating the benefits of personalization:
- A widely cited meta-analysis, drawing on data from 58,454 participants, showed that personalized health articles tend to outperform generic ones.
- A thought-provoking study found that XPhone ads tailored to four of the five major personality traits received significantly more clicks.
- One paper published in a prestigious journal revealed ads tailored to the user’s level of openness received 38% more clicks.
In sum, the literature strongly supports the idea that personalized messages are more persuasive. But there’s a problem. Personalization often stops with using a customer’s first name even though much richer tailoring is possible given the right methods.
The User-as-Wizard Method Inspires Personalized Experiences
Picture our friend Kyle, the party animal from earlier. His birthday is coming up and we have three options for gifts: a fountain pen, a calculator, or a bottle of Jack Daniels.
What gift would put the biggest smile on Kyle’s face?
My guess is that you picked the whisky because it jives more with his lively character. The premise of this example illustrates the User-as-Wizard method: A group of participants play the role of the app and report on how they would adapt communications to a user with a given set of characteristics.
Consider how this applies in Kyle’s case. We’re acting as the app by generating recommendations for, say, a personalized gift service. Our first task is to suggest a present for Kyle. We learn he has the user characteristic of extraversion, so we pick whiskey as a gift instead of the more dull alternatives.
Those are the steps one participant would follow in this method. Now if we aggregate recommendations like this across roughly 50 participants then we’ll notice patterns called statistical regularities. These reveal the rules to follow so we can personalize a service; for instance, if the user is extraverted then recommend whiskey.
To get this result, though, we need first to convey that Kyle is extraverted to our participants. Thankfully, we can accomplish this using story sliders: a method for assessing and representing a person’s personality where users rate which of two characters they are most like.
Here’s how it works.
Drawing on the slider above, a participant would report on their personality by stating whether they are more like the conscientious or laid-back character. They then find out that a person, we’ll call them Scott, has a similar personality to them. Finally, the participant plays the role of the system and rates the effectiveness of different persuasive appeals for Scott.
By providing these ratings the user is acting as a wizard that trains the system — that’s why it’s called the user-as-wizard method. To clarify, let’s look at a more practical example.
Story Sliders Help Tailor Skin-Cancer Prevention Messages
Melanoma is one of the top five cancers.
In fact, a doctor will diagnose two-thirds of Australians with skin cancer by the age of 70. If a person who catches the disease is fortunate enough to have it enter remission, they’re still not in the clear. Unfortunately, people who have recovered from skin cancer are 8-15 times more likely to have another malignant growth.
For this reason, those in remission have to be extra careful. Doctors recommend that they check their skin — top to bottom — on a regular basis. To remind users of this preventative measure an app was developed, but the creators were dissatisfied by the rate of compliance.
How do we make these reminders more effective?
That’s exactly what Judith Mastoff and colleagues — leading researchers in human-computer interaction — set out to investigate. They started by creating different reminder messages. These messages varied based on which of Cialdini’s persuasion principles they expressed (see below for an example)
“This is your last opportunity for your monthly skin check. Do not miss out – please check your skin now.”
A scarcity framed persuasive appeal
With the messages in hand, researchers asked participants to complete story sliders for the Big-5 traits. They then rated the persuasiveness of different reminder messages for a person like them. Here are some of their most notable results:
- Participants high in emotional stability were more receptive to all persuasive appeals.
- Those high in conscientiousness we’re more likely to rate an authority framed reminder as persuasive
- Those low in conscientiousness were more likely to find a scarcity framed reminder persuasive.
One explanation for the low conscientiousness finding is that these individuals tend to procrastinate. A message that basically says “Hurry up and check yourself for skin-cancer” is one they’re more likely to respond to.
These results give us invaluable proof of concept. Researchers can use story sliders to create algorithms that send personalized (health) messages. By algorithm, I’m referring to a set of steps, much like a cooking recipe, that produces a result. Here’s an example:
If the user is low in conscientiousness then send a reminder framed in terms of scarcity.
One limitation of this rule is that it fails to take into account other traits and demographic characteristics such as gender. That said, prior research has shown that personalization based on only a single trait can have a significant impact, i.e., increase click-through rates by 38%.
Conclusion
We’ve covered a lot of territory from persuasion profiles to the user-as-wizard method. Let’s recap by going over the major arguments:
- Personality traits point to people’s persuasion profiles. That is, the Cialdini principles (e.g., scarcity) they are apt to respond to.
- Personalized persuasion is a particularly effective tool in marketing and health communications.
- In the user-as-wizard method, participants play the role of the system and provide recommendations for different types of users.
- One reliable way to convey the characteristics of a user is with story sliders.
- If we use story sliders in conjunction with the user-as-wizard method then we can create algorithms that tailor messaging based on a user’s personality.
The sliders are a relatively recent invention — they came out in 2018. Despite their recency, these tools show promise as devices for both measuring and depicting user characteristics.
On that point, this piece focuses on just one set of traits, the Big-5. However, there are other characteristics that researchers can convey including cultural orientation, self-efficacy, and resilience.
In the end, the technique is flexible. Teams can use it to create personalized messages for everything from skin examination to eating fruits and vegetables.
As time slides by we’ll see the many uses for this clever device.
P.S. The confetti cannon was a hit (RIP friend’s rug).
References
Hirsh, J. B., Kang, S. K., & Bodenhausen, G. V. (2012). Personalized persuasion: Tailoring persuasive appeals to recipients’ personality traits. Psychological science, 23(6), 578-581.
Kaptein, M., Markopoulos, P., De Ruyter, B., & Aarts, E. (2015). Personalizing persuasive technologies: Explicit and implicit personalization using persuasion profiles. International Journal of Human-Computer Studies, 77, 38-51.
Matz, S. C., Kosinski, M., Nave, G., & Stillwell, D. J. (2017). Psychological targeting as an effective approach to digital mass persuasion. Proceedings of the National Academy of Sciences, 114(48), 12714–12719. https://doi.org/10.1073/pnas.1710966114
Noar, S. M., Benac, C. N., & Harris, M. S. (2007). Does tailoring matter? Meta-analytic review of tailored print health behavior change interventions. Psychological bulletin, 133(4), 673.
Smith, K. A., Dennis, M., & Masthoff, J. (2016, July). Personalizing reminders to personality for melanoma self-checking. In Proceedings of the 2016 Conference on User Modeling Adaptation and Personalization (pp. 85-93).
Smith, K. A., Dennis, M., Masthoff, J., & Tintarev, N. (2019). A methodology for creating and validating psychological stories for conveying and measuring psychological traits. User Modeling and User-Adapted Interaction, 29(3), 573-618.
SunSmart. (2018). Skin cancer facts & stats – SunSmart. Sunsmart.com.Au. https://www.sunsmart.com.au/skin-cancer/skin-cancer-facts-stats
Thomas, R. J., Masthoff, J., & Oren, N. (2017, April). Adapting healthy eating messages to personality. In International Conference on Persuasive Technology (pp. 119-132). Springer, Cham.