The common path is real and useful: people learn the knowledge in existing e-learning, and later an application step is added on top so they learn to apply it. A separate guide covers that path in detail, from picking the program to measuring the result (adding an application step). But nothing requires the knowledge and the application to arrive in two separate builds. When you are building something new, you can combine the two, the knowledge and its application, inside one experience. That is what this guide is about, and it comes down to two moves and one rule that keeps them from collapsing into each other.
Deliver the knowledge as a conversation, not a lecture
There are two ways to hand someone the knowledge.
The first is a lecture. One presenter talks at the learner, or a second presenter takes turns doing the same. Either way it is a monologue: the learner listens and nothing in the format asks them to do the work of making sense of the material. A monologue is not the only option, and it is not the most effective one.
The second is a conversation, and the evidence favors it. When Craig and colleagues had learners overhear a tutor and a student work through a topic as a back-and-forth dialogue, those learners recalled more of the material and went on to ask deeper questions than learners who heard the very same content delivered as a monologue (Craig et al., 2000). Muller and colleagues found the same pattern in a study of 364 students: a two-person dialogue version of a lesson, in which a student and a tutor discussed the material, produced larger learning gains than the same material presented as a concise lecture, with an effect size of 0.83 for the dialogue version (Muller et al., 2008). Learners can absorb a great deal simply from watching a good exchange: in one study, students who observed a tutoring dialogue alongside a partner learned about as much as the students actually being tutored one-on-one (Chi, Roy, & Hausmann, 2008). And Mayer's research on multimedia learning names a personalization principle: people understand more when words reach them in a conversational style rather than a formal one, a pattern that held across eleven experiments with a median effect size of 1.11 (Mayer, 2008).
So build the knowledge half as a conversation between two 3D avatar characters, the way two hosts of a podcast talk through a subject. They share their own experience of the material: how they worked it out, how they felt, what they struggled with, and how they ended up using it. They question each other, one asking "but what about this?" and the other answering, and they talk it through rather than reciting it. Visuals come up alongside them to help explain, which helps for its own reason: people learn better from words and pictures together than from words alone, one of Mayer's principles for scenarios (Mayer, 2008). The result is a lively way into the knowledge rather than a presentation the learner sits through. Casting those two characters well is its own craft, and our guide on the roles each character plays works through why a podcast-style pair beats a single presenter.
Then set up a situation and let the learner apply it
Once the conversation has delivered the knowledge, the natural next question is: now how do we actually use this? That is the second move. You set up a realistic situation and the learner learns to apply the knowledge by making decisions inside it and seeing what follows from each one.
That situation can be another part of the same scenario, picking up where the conversation left off, or it can be a completely different scenario built for the purpose. Either works. What matters is that the learner now does something with the knowledge instead of only hearing about it.
Keep the two parts separate
This is the rule that holds the whole thing together, and it is easy to get wrong.
The knowledge component and the application component are separate. The characters discuss the knowledge first. Then, as a distinct move, you set up a situation and the learner applies what was discussed. If the knowledge is baked into the situation, with the characters teaching the material inside the decision flow, the learner is making choices about material they were never actually given. Discuss the knowledge, and then, separately, apply it.
The structure looks like this: one experience, two separate parts.
The knowledge part is a conversation between characters. The application part is a situation the learner acts in.
The two parts sit inside one experience, but they are never blended into a single set of decisions. The characters discuss the knowledge first; the situation comes after.
Why build both together
Putting the two parts in one experience does something neither half does alone. The characters' lived experience of the knowledge is what makes it relevant to the learner: here is the material, here is how these two people actually used it, and that is why it matters to you. Then the learner applies it in a situation, and that makes it personal. What you have worked through yourself is more memorable than what you were only told.
It helps to be clear about what each half is for. The knowledge half exists to give the learner the tools to apply the material. The application half is not an assessment and not a test; it is the learner actually applying the knowledge in a realistic situation. The point is not to check whether they remember the content. The point is to let them learn to apply it while the reasoning is still live.
When you already own the knowledge
Combining the two into one new experience is the right move when you are building from scratch. When you already have knowledge content that works, you do not throw it out to rebuild it this way. You keep it and add a separate application layer on top, which is the approach our guide on adding an application step lays out. The two-part structure is the same either way. The only question is whether you are building the knowledge half now or reusing one you already have.
How this comes together
The most engaging way to build knowledge and application into one experience is to deliver the knowledge as a conversation the learner can follow and then, separately, drop the learner into a situation where they learn to apply it. AliveSim is built to carry both: a podcast-style conversation among 3D avatar characters for the knowledge, and a set-up situation with decisions for the application, kept as two distinct components inside a single experience. For a closer look at how that fits together, AliveSim is the platform that embodies this approach.
References
- Chi, M. T. H., Roy, M., & Hausmann, R. G. M. (2008). Observing tutorial dialogues collaboratively: Insights about human tutoring effectiveness from vicarious learning. Cognitive Science, 32(2), 301–341.
- Craig, S. D., Gholson, B., Ventura, M., Graesser, A. C., & the Tutoring Research Group. (2000). Overhearing dialogues and monologues in virtual tutoring sessions: Effects on questioning and vicarious learning. International Journal of Artificial Intelligence in Education, 11, 242–253.
- Mayer, R. E. (2008). Applying the science of learning: Evidence-based principles for the design of multimedia instruction. American Psychologist, 63(8), 760–769.
- Muller, D. A., Bewes, J., Sharma, M. D., & Reimann, P. (2008). Saying the wrong thing: Improving learning with multimedia by including misconceptions. Journal of Computer Assisted Learning, 24(2), 144–155.
Related questions
Can one experience include both the knowledge and its application?
Yes, when you are building something new. You can deliver the knowledge and then set up a situation where the learner applies it, both inside one experience, as long as you keep them as two separate parts. The knowledge is delivered first, as a conversation between characters. The application comes after, as a separate situation where the learner makes decisions. This is different from the common path, where learners acquire the knowledge in existing e-learning and an application step is added on top later. Both approaches are valid; this one is for the case where you are building the whole thing fresh.
Why deliver the knowledge as a conversation instead of a lecture?
A lecture is a monologue: one person talks and the learner listens. It is passive, and it has been shown to be less effective than a conversation. Learners who overheard a tutor and a student work through a topic as a back-and-forth dialogue recalled more and asked deeper questions than learners who heard the same content as a monologue (Craig et al., 2000). A two-person dialogue version of a lesson produced larger learning gains than the same material as a concise lecture (Muller et al., 2008). And people understand more when words reach them in a conversational style rather than a formal one (Mayer, 2008). Two characters talking through the material like podcast hosts, with visuals alongside, is a more engaging way in than a single presenter talking at the learner.
Why keep the knowledge and its application separate?
Because they do two different jobs, and blending them breaks both. The knowledge component is where characters discuss the material so the learner picks up the tools to use it. The application component is a separate situation where the learner learns to apply those tools by making decisions. If the knowledge is baked into the situation, the learner is making choices about material they were never actually given. Discuss the knowledge first, then, as a separate move, set up a situation and let the learner apply it.
What if I already have knowledge content?
Then keep it. You do not have to rebuild anything. When knowledge content already exists and works, the better move is to add a separate application layer on top of it, which our guide on adding an application step covers in full. Combining the knowledge and the application into one new experience is the approach for when you are building fresh. When you already own the knowledge half, the two-part structure still holds; the knowledge half is simply the content you already have, and the application step is what you add.
Published July 20, 2026 · 7 min read