Engagement as the Architecture of Learning

A few weeks ago, I gave you my two cents (well, they’re more than two cents, by now) on the future of adult learning, and the first pillar of my theory was the necessity to involve game design. This week, back from Denmark and while in London, I’d like to elaborate on that.

1. Why Games Work: Narrative, Challenge, and Progress

Think about the last time you actually wanted to do something difficult. And I mean, really wanted to. Not because you had to, not because your job depended on it or someone was forcing your hand. You wanted to because the difficulty itself was the draw. I bet you were playing a game, or solving a puzzle, or learning something that felt like play.

Mihály Csikszentmihalyi called this state flow: that moment when the challenge level matches your current skill just perfectly, when you’re so absorbed in what you’re doing that time disappears. It’s not about fun in the frivolous sense. It’s about engagement so deep that the work becomes indistinguishable from pleasure. And it’s the closest thing we have to understanding why humans learn at all.

Well, games are flow machines, in a way: they’re engineered from the ground up to create that match between challenge and skill, and to adjust it constantly as you improve. You start a new game, and it’s generally manageable: hard enough to feel satisfying when you win, easy enough that you don’t quit in frustration. As you get better, the game gets harder. You never hit that wall of impossibility that makes you give up. You never spend hours grinding at something you’ve already mastered just to unlock the next level. The best games know exactly how much pressure to apply, and they adjust in real time. This is their architecture.

As far as I see it, narrative is one of the first interesting load-bearing walls in that architecture, because stories make learning sticky. When you encounter information inside a story — when a character needs something, when stakes exist, when there’s a reason to care about the outcome — your brain lights up differently than when you’re reading a list of facts. The story becomes a scaffold. It gives you a place to hang new information. It answers the question that every learner, conscious or not, is always asking: why should I care about this?

I spent time last year talking to educators who use narrative design in their classrooms, reading their books and watching their videos, and the pattern was always the same: the moment they wrapped a learning objective inside a story, engagement didn’t just improve but transformed. Students who had been passive suddenly became invested. They asked questions. They stayed after class. They went home and kept thinking about the problem because the story had made it theirs.

This is a thing game designers have known for decades: a well-told story doesn’t distract from learning, but it enables it by making the learner care about the outcome. The narrative creates emotional resonance, and emotion is the glue that binds information to memory.

Narrative alone, though, isn’t enough. You need progress and a goal. You need to feel like you’re advancing toward something.

Clear goals and mastery loops are the second wall. In games, you always know what you’re working toward, the goal is visible, specific, and achievable within a reasonable timeframe. You’re trying to reach the next level, or solve the next puzzle, or unlock the next skill. And when you achieve that goal you get clear, immediate feedback. You know you’ve succeeded. Not eventually. Not after a midterm exam three months from now. Right away.

This feedback creates momentum because, psychologically, humans are drawn to progress: we are achievement creatures who want to move forward and, when the steps are clear and the feedback is immediate, we keep moving. This is why people spend hundreds of hours grinding through repetitive tasks in video games, tasks that would bore them to tears in any other context: in the game, those repetitive tasks are framed as progress toward a larger goal, and progress itself is deeply rewarding.

Adult learners need this as much as anyone else, maybe more. An adult professional juggling a job, maybe a family, maybe ageing parents, don’t have the luxury of assuming that effort will pay off eventually: they need to see progress now. They need to know that the time they’re investing is moving them closer to competence, closer to the skill they came to learn. Without that clear progression and immediate feedback, they quit. And they should quit. If a learning system can’t tell you whether you’re actually getting better, why would you keep showing up?

The third wall is challenge calibration. Flow requires that the challenge level stay matched to your skill. Too easy and you’re bored. Too hard and you’re frustrated. But learners improve at different rates, and they start from different places: the moment you try to teach a static lesson to thirty people, you’ve already lost half the room because ten are bored, and ten are lost. Some games solve this by adjusting in real time: they watch what you’re doing, they measure how quickly you’re succeeding, and they turn the dial up or down accordingly. When you fail, good games don’t punish you harshly but assume you’re learning and they might scale back the difficulty for a moment, or offer a hint, or restructure the challenge so you can approach it from a different angle. The goal is to keep you in that flow state: engaged, challenged, but not defeated.

This is extraordinarily difficult to do in a traditional classroom, but it’s possible in a well-designed learning system, especially one that has AI at its back end. More on that later. For now, the principle stands: games work because they balance challenge with skill, and they adjust that balance as learners improve.

2. Game Mechanics in Adult Learning Contexts

Understanding why games work in principle is one thing; translating that into adult professional learning is another.

A quest is a narrative structure that wraps a learning objective inside a goal. Instead of “Complete this module on project management,” the learner might encounter: “You’ve been promoted to lead a new product launch. You have eight weeks to assemble a team, manage the budget, and get to market on time. Each week, you’ll face a new challenge that requires a specific skill.” The content — budgeting, resource allocation, stakeholder management — is embedded inside the quest. But because the learner is chasing a goal that matters (in the narrative), they might be motivated to engage with the content.

Progression systems turn abstract learning into concrete advancement. Rather than accumulating vague knowledge, learners move through defined stages: novice → practitioner → expert. Each stage has specific competencies. You know where you are. You know what you need to do to get to the next stage. You can see, at a glance, how far you’ve come. This is remarkably simple and remarkably powerful. It answers two questions that traditional education often leaves unanswered: Where am I? and Where am I going?

Badges and recognition, when used right, serve a specific function: they externalise achievement. A badge isn’t just a symbol: it’s proof. It says: “You did this. You mastered this skill. You completed this challenge.” For adult learners, especially those who are balancing learning with work and life, visible recognition matters. It reminds them why they’re doing this. And if a badge is portable — if you can share it on LinkedIn, or include it on your resume, or wear it as a credential — it transforms learning from something internal into something that has real-world value: the recognition of your peers.

Leaderboards and social proof are trickier: they can be motivating, as humans are competitive creatures and seeing that you’re advancing faster than your peers can spur you on, but they can also be demoralising or create unhealthy competition. If you’re always at the bottom of the leaderboard, seeing it every day is dispiriting, not motivating. Plus, I think I believe that competition is always unhealthy. The best learning systems make leaderboards opt-in, I think, or they structure them around improvement rather than raw score, so the person who improved the most this week gets highlighted, not just the person with the highest absolute score. The goal is to create healthy social proof — the sense that others are doing this work, and you’re doing it alongside them — without creating a feeling of inadequacy.

For adult professionals, as far as I see it, there’s another element that’s often missing from game design literature but essential here: autonomy. Adults have limited time. They don’t want to be told what to do in the order the designer decided. A good learning game for adults offers multiple paths through the content. You can choose which quest to tackle first, whether to deepen your knowledge in one area or move broadly through several, to work alone or collaborate with peers. This is essential for retention and engagement. Adults who feel like they have control over their learning are more likely to persist.

3. Beyond Gamification: Game Design Thinking vs. Cheap Game-Like Elements

Here’s where we need to be honest about what’s failed in the past decade, because this approach isn’t new. Gamification, as it’s often been practiced, is not game design: it’s the deployment of game-like elements — points, badges, leaderboards — without the underlying architecture. It’s treating these mechanics as decoration you can slap onto any learning experience and expect engagement to magically improve.

It doesn’t work that way.

A company might add points to their employee training program (as they do), where employees get fifty points for completing a module. They see a progress bar, 450 points toward a gift card, and technically it’s gamified. But if the content itself is boring, if there’s no clear narrative why they’re learning it, if the challenges are too easy or too hard, if the feedback is vague. The points and pretty badges don’t fix any of that. They just add busywork. People rush through, collect their points, and forget everything.

The distinction between learning mechanics and gaming mechanics is crucial: learning mechanics are the structures that actually support acquisition of skill or knowledge, such as clear goals, immediate feedback, progressive difficulty, practice with variation, error correction; gaming mechanics are the reward systems: points, badges, competition, unlocks. You can have learning mechanics without gaming mechanics and still have effective learning. But gaming mechanics without learning mechanics is just slot-machine psychology. It might drive short-term engagement, but it doesn’t drive learning. And that’s the Dark Side of game design.

Bad gamification, to clarify this latter point, is often manipulative in ways that undermine genuine engagement: when you design a system where points are the primary currency, you’re creating incentives for people to optimise for points, not for actual learning, and that’s not dissimilar to what we saw a while ago about gaming the KPIs. People will find the easiest way to rack up points, whether or not it advances their competence. The moment they realise what’s happening, motivation collapses. They feel tricked. They’ll game the system, literally, and screw you ’cause you tried to screw them.

Good game design, by contrast, aligns incentives. The things that are rewarded (points, badges, progression) are the same things that indicate genuine learning. When you earn a badge for “expert in data visualisation,” it’s because you’ve completed a series of challenges that required you to actually develop that skill. The badge is not a cosmetic reward, but the tangible signal of something you already know: that you can do the thing.

Why has so much corporate training gone wrong with gamification? Usually, because someone in leadership read a blog post about how games drive engagement, they got excited, and they hired a consultant to “add gamification” to their existing training program. The consultant sprayed some game elements onto content that was still fundamentally unchallenging, unnarrative, and unmotivating. And then when engagement didn’t improve, they blamed gamification instead of blaming the underlying content design.

4. Case Evidence: how Game-Based Learning outperforms Traditional Instruction

The research here is actually quite robust, once you dig past the hype.

A meta-analysis of digital game-based learning in STEM contexts found that “enhanced game design outperformed the basic game version in enhancing students’ STEM academic performance. This empirical finding strongly suggests that more attention should be given to the game design elements for game-based STEM education.” More specifically, when games are designed with learning mechanics in mind — meaning when the difficulty adjusts to the learner, when feedback is immediate and instructive, when the narrative is integrated with the learning objective — students show measurable gains in knowledge retention and skill development. The effect sizes vary depending on the subject and the learner population, but they’re consistent: game-based learning works. And crucially, research shows that “elements related to learning mechanism were more effective in promoting students’ cognitive performance than those related to gaming mechanism,” showing that the real engines of improvement are the pedagogical design choices, not the cosmetic game elements.

But here’s what’s more interesting than the raw numbers: the mechanism. When researchers examine what’s actually happening during game-based learning, they find something we could have predicted but rarely measure: research confirms that “games afford rich opportunities for communication, collaboration, fantasy engagement, problem solving, hypothesis generation, identity development, and reflective thinking.” Engagement is higher. Students stay in flow longer. They encounter difficulty, they try again, they adjust their approach. And crucially, they do their work willingly, even enthusiastically. They’re not white-knuckling their way through required material: they’re drawn in.

The emotional and behavioural dimensions are perhaps even more significant for adult learning. Xun Ge (University of Oklahoma) and Dirk Ifenthaler (University of Mannheim) have identified “three dimensions of engagement: behavioural, cognitive, and emotional,” and game-based systems can tap into all three simultaneously. Adults don’t just care about whether they’re learning: they care about how they’re learning. If a learning experience feels tedious or disrespectful of their time, they’ll bail as they should. Game-based learning, when done well, should signal respect for the learner and give the learner agency. It adjusts to their pace. It acknowledges that they’re improving. These things matter psychologically. They change how learners approach the work.

The numbers bear this out. Research shows that “active learning increases student performance by up to 20%, presenting a significant opportunity for improvement in educational outcomes.” But what matters for adult professionals is application and retention. When learning is embedded in game-like structures with narrative stakes, repeated practice, and clear feedback — the hallmarks of actual game design — people don’t just accumulate knowledge but develop competence. They’ve practised to the point of fluency. They’ve made actual decisions and seen the consequences. They’ve received repeated feedback on their performance. This is the kind of learning that transfers to the real world.

Consider what happens in a collaborative game-based learning environment. Research on game design in education found that “game design engaged student teams in sustained, collaborative efforts to create shared digital artefacts. Their efforts involved a great deal of mutual support and knowledge sharing.” The learning isn’t solitary but social. People learn from each other. Expertise flows in multiple directions. And the act of helping someone else learn actually consolidates your own understanding. This matters for adult learners who are often working in teams, who need to collaborate with peers, who benefit from the shared struggle of solving real problems together.

5. Accessibility & Neurodiversity: when Collaboration becomes overwhelming

The emphasis on collaborative, peer-based learning in game-based environments requires an honest caveat: mandatory social engagement can be deeply detrimental to learners on the autism spectrum and those with related sensory processing differences. Research on autistic adults in learning and workplace contexts reveals a consistent pattern: “sustained demands of masking, interpersonal interactions, changes and other stressors combined with unaccommodating environments” drive what’s known as autistic burnout, a state of profound exhaustion distinct from regular fatigue, characterised by skill loss, heightened sensory sensitivity, and reduced functioning. For neurodivergent learners, particularly those engaging in what researchers call “masking” (suppressing natural traits and communication styles to conform to neurotypical expectations), mandatory group work transforms learning from engagement into survival. The cognitive load of decoding nonverbal cues, managing sensory input in shared spaces, performing appropriate eye contact and tone, and simultaneously absorbing new material creates what one study describes as “sensory debt,” a cumulative overload that doesn’t require extreme environments to be exhausting, only that it be constant and unavoidable.

This means that truly inclusive game-based learning must build in genuine alternatives to collaborative work, not as afterthoughts but as architecturally equal paths through the content. Some learners will thrive in asynchronous knowledge-sharing and peer feedback through text-based or recorded formats. Others will benefit from learning in structured small groups with predictable social rules, or from one-on-one mentorship rather than team-based quests. Some will need sensory-friendly virtual or in-person spaces with noise management, lighting control, and clear communication protocols. Research on autistic learners in education offers a clear standpoint: “successful inclusion is possible if educators commit to ongoing training that uses evidence-based practices, supports students socially, and addresses environmental challenges.” For adult learning systems, this translates to a design principle: build collaboration as an option and a strength-leveraging tool, not as a mandatory path. Offer choice. Offer alternatives. And critically, listen to neurodivergent learners about what actually works for them, recognising that what looks like disengagement is often protective withdrawal from an environment that’s physiologically overwhelming. The power of game-based learning isn’t diminished by making it accessible but amplified by creating multiple valid pathways to mastery, where some learners travel together and others travel solo, but all arrive at competence.

6. The Engagement Gap: from Curiosity to Competence

Finally, there’s another gap that opens up in most traditional learning programs, and it’s where learners fall through.

At the beginning, when something is new, there’s natural curiosity. You’re drawn in. You want to understand. But curiosity alone doesn’t carry you through the hard middle. When the material gets more complex, it’s no longer novel and just turns difficult. And in a learning system that doesn’t actively sustain engagement, this is where people quit.

Active learning — where learners are doing something, not just receiving information — increases performance by roughly 20% compared to passive instruction. That’s not trivial. But active learning alone doesn’t guarantee sustained engagement. You need all the other elements: clear progress, narrative stakes, challenge calibration.

Game-based learning bridges this gap because it sustains engagement across the entire journey from curiosity to competence: the narrative is what keeps you interested, the progression system is what shows you that you’re advancing, the difficulty adjusts so you stay challenged but not overwhelmed. By the time you reach competence — by the time you’ve actually mastered the skill — you’ve been pulled through the hard middle by structures that made the work feel like progress rather than drudgery.

And there’s another element that’s often underestimated: collaboration. The best game-based learning systems aren’t purely competitive or purely solitary, but they create space for learners to work together. One person figures out a solution, they share it, others build on it. Knowledge spreads. Confidence grows. The person who helps others learn actually consolidates their own learning: teaching is one of the most powerful ways to deepen understanding.

This is crucial for adult learners, who are often coming from diverse backgrounds and bringing different expertise, and that’s why I love classes with a mixed background: a software engineer learning about project management has things to teach to a project manager learning about technical constraints. A group of peers learning together, where knowledge flows in multiple directions, is where real collaborative learning happens. Game-based systems that enable this kind of peer learning, that reward knowledge-sharing, that make expertise visible and valued—they tap into something genuinely motivating.

By the time learners reach the end of a well-designed game-based learning experience, they have developed competence by proving their skills repeatedly against increasingly difficult challenges. They’ve received recognition for it. They’ve potentially helped others develop it. The learning has become real, embodied, durable.

This is what we should be chasing in adult education: not engagement for its own sake, but engagement as a means to the kind of deep, durable learning that actually changes how people work.

Game design gives us the architectural principles to build that: the narrative, the progression, the challenge calibration, the feedback, the autonomy are the bones that hold the structure together.

The question now is: how do we scale this? How do we personalise it? How do we make sure it works for every learner, regardless of their background or learning style?

That’s where AI enters the picture. But that’s a conversation for next week.