UX for Learning: Design Guidelines for the Learner Experience

5. Don’t separate related text and visuals.

In line with the basic design principle of proximity, learning research has shown that people learn better when text relating to an image appears near that image. As obvious as it sounds, designers often violate this guideline. For example, rather than adding a list of labels to a diagram’s caption, create callouts, drawing lines to various parts of the diagram and labeling those parts directly; and rather than providing feedback for an incorrect answer at the bottom of a screen, include it beside the learner’s answer.

Also, include an explanation in the form of a narration whenever possible. If you can’t use narration, provide text for learners to read before playing a video. While you can present these on the same screen, avoid having text appear on a screen at the same time a video or animation is running. That would be too much visual information for learners to take in at once.

6. Use virtual agents to deliver instructional content.

Research shows that people learn better when they perceive social presence. For this reason, using a conversational tone in first or second person increases learning. We are hardwired to pay more active attention when we’re dealing with human beings rather than machines. This is probably why virtual coaches, tutors, characters, or avatars help people learn. Keep in mind that you should use agents for instruction, not entertainment. Agents are good for providing hints, explanations, and demonstrations, but you need to use them sparingly to prevent their becoming irritating.

Interestingly, realistic characters are not more effective instructors than cartoon characters. In fact, an agent does not even need to be visible, as long as learners hear a voice. Also, a voice should be a human voice rather than a machine voice. So don’t blow your budget on a 3D animated avatar with a robotic voice. A simple illustration with a human-voice narration—regardless of whether that human is visible—is enough for learning gains. As a final note, while having a realistic human appearance is not necessary, learners do benefit from agents who exhibit realistic human behavior such as human-like eye gaze and gesturing.

7. Break complex information into chunks and provide pre-training.

Two basic ways of managing complex information are either to break it into segments, or chunks, or to extract difficult terms or key concepts and deal with those first, which is called pre-training. Segmenting means breaking content into parts that deal with just one, two, or three steps or concepts at a time.

Lynda.com provides a wonderful library of training courses on everything that relates to the Web. If you haven’t already seen the site, it’s well worth a visit. Their instructional approach is to provide high-quality videos of best practices or software training. This works well because video is a good medium for demonstrating procedures. They manage complexity, and give users control over the pacing, by breaking each course up into bite-sized concept chunks. This also means people can select and review just the bits they need when they need them.

8. Avoid irrelevant video, audio, stories, and text.

Research shows that adding sidebar stories for the sake of interest hurts learning. It may seem sensible to spice up a lesson on the physics of lightning with a sidebar on the world-record holder for the most lightning hits. However, research has proven that this is not a good idea. The message here is to focus on what is inherently interesting about the content rather than adding extraneous detail—especially if specific learning goals are at stake. Along these lines, adding music or audio to learning experiences for ambiance impairs learning.

9. Set learning in a real-world context to facilitate knowledge transfer.

The closer a learning experience is to a real-world context, the more likely learners will be able to transfer what they’ve learned to the real world. One criticism of video games for learning stems from their low transfer rates—skills people learn in the game world don’t transfer to the real world. One company thought a corporate orientation game set in a sci-fi world would have greater appeal. However, in reality, it had little uptake because learners couldn’t see its relevance. Moreover, learners would be unlikely to transfer concepts they learned in such an environment back to the real world, because there would be no contextual cues embedded in the learning.

If people need to adapt the skills they’re learning to multiple environments and diverse circumstances, include examples set in more than one context. When you expose learners to multiple contexts, they are better able to extract essential principles and transfer them to various real-world scenarios.

10. Provide informative feedback.

Avoid providing feedback that sounds personal or draws attention to the self instead of the task—for example, “good job” or “you got that wrong.” Also, avoid giving feedback that is uninformative like “correct” or “incorrect.” While an instructional designer writes the feedback, as a user interface designer, you need to design space in the interface for rich forms of feedback, remembering to place answers near the questions to which they refer.

11. Choose your tools based on learning activities.

Different types of tools are most effective for different types of learning activities. For example, asynchronous communication tools that allow interaction over time—such as discussion forums—are well suited to learning experiences that require reflection or independent research. Synchronous tools like chat or live online classroom sessions are well suited to experiences that benefit from group synergy and social presence.

12. Know when to give learners control.

Allow learners to control their pace through a learning experience. Give them video and animation controls, let them review previous content, and allow them to exit at any time. However, you should be more careful about giving learners control of the sequence of instruction.

Research shows that novices are not able to effectively design their own path through non-linear content or decide how much practice they require. This impacts how you design both navigation and information architecture. You should allow learners choices regarding what topics to read or activities to do only if they have prior knowledge, in advanced courses, when chunks of content are not interdependent, or when designing a learning experience to be supportive and informational rather than for skill building. Along these lines, research shows that unguided exploratory learning is not effective for achieving specific learning outcomes.

13. Design learning games and simulations in strict alignment with learning goals.

The recent trend in gamification has led to the assumption that adding badges, rewards, and scoreboards to just about anything makes it a game. The main point to take away from games and learning is the notion of extrinsic versus intrinsic motivation. While it’s tempting to think that adding game rewards and goals would increase learners’ motivation and thus help them to learn better, you have to ask what is motivation about? Are people more motivated by learning a topic or by collecting points?

In fact, research has shown that learning does not improve when a learner’s focus is on game goals rather than learning goals. You could have learners practice and teach them facts by placing them in a virtual Jeopardy-style game show, but this would decontextualize the knowledge making it unlikely to transfer. Effective learning games retain contextual cues, align game goals with learning goals, and drive interest with intrinsic motivation. More on that later.

14. Be aware of the expertise reversal effect.

A major caveat to everything I’ve said so far is that the rules break down when it comes to experts. This is known in the research world as the expertise reversal effect. In essence, all of the guidelines I’ve provided are especially important for novices. If you’re teaching biology or management skills to non-experts, everything applies. However, if you’re teaching advanced concepts in biology to a doctor or management strategy to a CEO, these rules may not apply. Such design decisions usually make less impact on such people. Experts are more flexible and can learn just as well from text alone as they can from a combination of text and visuals. Experts don’t need information to be chunked, and they can handle having more control over their instruction.

In general, following design guidelines that benefit non-experts won’t negatively impact experts. However, in a few cases, the rules for designing for novices do actually decrease learning outcomes for experts. Therefore, if you’re designing for an audience of experts, it’s worth reading more about their unique requirements—first, in Clark and Mayer’s book; second, by looking up the research on the expertise reversal effect.

Unleash Intrinsic Motivation

Motivation is key in learning. Therefore, when considering approaches to digital learning, it’s worth giving motivation some extra attention. In addition to the points I made earlier about the pitfalls of gamification and the limits of extrinsic motivation, I want to focus now on how UX professionals can become more creative about uncovering intrinsic motivation.

Too often, intrinsic motivation gets skimmed over in digital learning design, but UX professionals are just the right people to tackle this. It takes empathy, ideation, and a knack for storytelling. One prevalent example is kids and literacy. The market abounds with drill and practice software for spelling and phonics. But imposing extrinsic motivation in the form of points, competition, and speed rounds has obvious limitations. Can’t we do better?

We need to ask ourselves the question: what is inherently motivating about reading and writing? The answer? There are many motivations: storytelling, engagement with characters, self-expression. Kids get wildly excited by Harry Potter. They use social media to tell the stories of their own lives. How can we tap into this kind of motivation, which is so much more powerful than badges and scores, to help teach literacy skills? Might we place spelling words in stores instead of random sentences? Could we ask students to write a letter to Hogwarts rather than just fill in a template? Could we design collaborative, global story-writing activities so all children, even lower-ability writers, could be part of something special? Could we ask famous writers to contribute?

The point is, while the onus of activity design rests with teachers and instructional designers, there may be times when UX designers have an opportunity to contribute to effective learning. In these cases, challenge your team to go beyond extrinsic motivation and tap into the heart of intrinsic motivation—what intrinsically makes your topic exciting.

Going Deeper

As I’m sure you’ve guessed, there is a lot more to designing for learning than what I’ve presented here. The trouble is, much of what is known is buried in academic research. If designing for learning is a significant part of your job, here’s where to turn next to enrich your credentials:

• Read the book. For a complete list of principles, examples, and in-depth explanations of the psychology behind them, read Mayer and Clark’s book E-learning and the Science of Instruction.
• Follow my blog. I have been posting information on LXD on my blog, Learning Interface Design, regularly since 2005.
• Read the research. Keep up with new publications on this research topic by following or joining the Learning Interface Design research-sharing community on Mendeley.

In Conclusion

These guidelines provide a solid start to informing our practice of the design of learning experiences. As digital learning continues to enrich and, in some cases, even displace traditional learning strategies, I anticipate that a rigorous field of research and best practices that focus specifically on the design of user interfaces and experiences for learning will begin to develop.