Skip to main content


Hey there! This week, we’re talking about the personalization principle and creating learner personas. Specifically, we hope to answer:

  • Does learner interest impact the personalization principle?
  • Can cognitive load interfere with understanding persona profiles?

Personalization for the People

Considering how much learning occurs online, we often reference Mayer’s principles of multimedia learning when talking about how to design effective learning environments. Recently, a group of researchers decided to put the personalization principle to the test - they evaluated whether the learner’s interest in the topic made a difference in learning outcomes related to the personalization principle (Schrader et al., 2018). As a refresher, the personalization principle essentially says that we should use an informal, casual, and conversational voice rather than an overly formal one. I find that Andrew Debell’s breakdown of the principles, and their accompanying images, are helpful when reviewing Mayer’s principles! The idea behind using conversational language is that it’s effectiveness comes from “triggering situational interest” (Schrader et al., 2018). So, what if the learners are already interested? Or, what if they won’t become interested regardless of our best efforts?

In the study, learners were split into two groups regarding language type: conversational (which follows the personalization principle) or formal. The multimedia presentation revolved around Gestalt Laws. Since I’m a big ‘ole psychology nerd, I encourage you to check them out, if you haven’t already heard of them. While you should read up on them because Gestalt Laws give us super cool information on how we perceive the world around us, they are also incredibly helpful for designing learning experiences. Anyhow, learners first completed a pre-test knowledge assessment, as well as a survey on individual interest. They then watched a prerecorded instructional video with slides; again, the only difference was the conversational vs. formal language of the instructor. After the presentation, learners completed a “learning performance test” (Schrader et al., 2018).

Did the personalization principle hold up? Results showed a significant difference - those in the conversational language group had better learning outcomes than those in the formal language group. Further, learner interest actually predicted learning outcomes. Learners with “either a very low (10th percentile) or very high individual interest (90th percentile)” were actually unaffected by the language style of the presentation. Learners with “intermediate individual interest,” those in the 25th, 50th, and/or 75th percentiles, were positively impacted by the conversational language style (Schrader et al., 2018).

When considering these results, conversational language style appears to reign supreme! While the lowest and highest levels of interest weren’t particularly impacted, they weren’t *hurt* by the conversational style. However, considering conversational language did not particularly affect those high- and low-interest learners, other measures may need to be taken for those groups. Thus, the authors suggest interest should be measured to assess whether multimedia principles would be effective and if personalized language needs to be employed (Schrader et al., 2018).

Key Takeaway: When recording instructional videos, be sure to use conversational language rather than formal language (i.e., the personalization principle). Learning outcomes are significantly higher when an instructor uses conversational language.

Read More ($): Schrader, C., Reichelt, M., & Zander, S. (2018). The effect of the personalization principle on multimedia learning: the role of student individual interests as a predictor. Education Technology Research and Development, 66.

Pictures in the Persona Profile?

We’re taking a small leap from personalization over to personas. For an introduction to personas, why they are great for understanding learners, and how they are immensely helpful for instructional design, check out Dr. Julia Huprich’s Medium article, Building Learner Personas for Instructional Design Effectiveness. While persona’s are great for understanding learners, when we provide a persona profile to an end user, we’re also asking them to learn about the persona. Due to this, a recent study sought to understand how we interpret information from persona profiles. Specifically, researchers wanted to know if additional photos beyond one headshot help the end user learn more (Salminen et al., 2019).

Although research on persona profiles is limited, some past work has found benefits to adding images. Specifically, adding images to a persona profile makes the profile more memorable, instills “greater confidence in the persona content,” and encourages the user to empathize (Nieters et al., 2007). If a photo is good for learning about the persona, are more photos better?

Salminen et al. (2019) evaluated pictures in persona profiles across two studies. In the first study, eye-tracking data was collected while end users were shown three different categories of persona profiles: a headshot and text, a headshot with additional contextual photos of the same person and text, and a headshot with additional photos of different people (although with similar characteristics) and text. While all users were shown all categories, presentation order was randomized (Salminen et al., 2019). As users interacted with the persona page, they were instructed to “talk-aloud” and say why they were looking at something specific. Through this “talk-aloud” portion, confusion and informativeness were coded. The second study consisted of qualitative interviews regarding persona profiles with and without context photos (Salminen et al., 2019).

The eye-tracking data showed, as expected, more fixations on the profiles with more pictures. This indicates that personas with multiple photos required more cognitive focus than those with just a headshot (Salminen et al., 2019). While this is an interesting finding, I think that confusion and informativeness are likely more important constructs. Regarding confusion and informativeness, profiles with photos of multiple people appeared to be the most confusing and least informative, while those with contextual photos were less confusing and the most informative (Salminen et al., 2019). The “traditional” persona profile, with just a headshot, was not seen as confusing, but was less informative than the contextual photos profiles. The following graph from the article is a handy visualization of the profile categorization.

(Salminen et al., 2019)

In the quest to understand “the optimal information content for persona profiles,” researchers found contextual photos significantly improved the information that end users got from a profile. Conversely, showing photos of multiple people led to confusion and lowered informativeness (Salminen et al., 2019). Further, interviews regarding persona profiles revealed some “cultural assumptions and simplistic explanations” of the persona (Salminen et al., 2019). As someone with a special interest in bias, I am particularly curious to find where this line of research leads; this finding was just “a starting point toward” the goal of understand the impact of culture on persona perception (Salminen et al., 2019). Although this finding was just a starting point, it does illustrate the need to decrease bias-inducing information in profiles, as well as educating end users on group diversity.

Overall, while choosing the photo(s) for a persona might not seem like a meaningful task, it can impact end users’ “experiences and preconceptions.” The researchers recommend considering the “intended persona use objectives when selecting photos” (Salminen et al., 2019). Since contextual photos bring some heightened level of informativeness, they should be integrated. However, photos of multiple people in a persona profile evoked confusion, so this should be avoided.

Key Takeaway: When creating persona profiles, adding contextual photos (multiple photos of the same person) increases informativeness for the end user over using a singular headshot. However, multiple photos of different people increases confusion and decreases informativeness, so this should be avoided.

Read More ($): Salminen, J., Jung, S., An, J., Kwak, H., NIelsen, L., & Jansen, B. J. (2019). Confusion and information triggered by photos in persona profiles. International Journal of Human-Computer Studies, 129.

"A tradeoff exists among informativeness, confusion, and perceptional bias when increasing the number of information elements in persona profiles, and for determining the optimum number of calls for awareness of how the information is perceived by the end users."
**- Salminen et al. (2019)

Pets of Learning Science Weekly

Reader Kristine K. shared her adorable “recent additions to the household” with us! These cuties can be seen enjoying “their first office fireside event” 😍

Send us your pet pics at

Wondering why we’re including animal photos in a learning science newsletter? It may seem weird, we admit. But we’re banking on the baby schema effect and the “power of Kawaii.” So, send us your cute pet pics -- you’re helping us all learn better!

The LSW Crew

Learning Science Weekly is written and edited by Katie Erhardt, Ph.D.

Have something to share? Want to see something in next week's issue? Send your suggestions: