Anyone who's spent time in a gym knows the drill. If you keep lifting the same weight, week after week, your body adapts and then just... stops. No new soreness, no new strength. Progress requires you to keep asking your muscles to do something slightly harder than what they're already used to.

If that's true for physical growth, why do we so often design learning experiences, courses, and training sessions to feel smooth, easy, and comfortable from start to finish?

What sport teaches us about growth

Strength coaches have a name for the principle above: progressive overload. Muscle growth depends on gradually increasing training stress, whether that's the load, the reps, or the total volume, over time. Too little stress and nothing changes. Too much, too fast, and you get injured instead of stronger. The sweet spot is somewhere in between, and it keeps moving as the person improves.

Motor learning research has its own version of this idea, and it's arguably more useful for thinking about education and workplace training alike. In their Challenge Point Framework, Mark A. Guadagnoli from the University of Nevada and Timothy D. Lee from McMaster University explain that we learn from the information we get when we try something. If a task is too hard, all we learn is that we can't do it. If it's too easy, we learn little about where to improve. The optimal challenge point, they argue, is where a task stretches the learner without overwhelming them.

A simple drill won't show an experienced athlete anything new, but for a novice it will surface clear areas to improve. A punishingly hard drill might reveal plenty for both to work on, but it can leave them feeling overloaded and defeated instead. What we need to design is a drill that gives enough information to guide growth, without performance falling apart.

Good coaches calibrate difficulty deliberately, watching for the point where an athlete is stretched but not broken. Educators, facilitators, and instructional designers are, whether they realize it or not, doing the exact same job. The only difference is the equipment, dumbbells in one case, case studies or simulations in the other.

What the research says about “desirable difficulty vs. being overwhelmed”

This isn't just a sports metaphor dressed up as education advice. Cognitive psychologists Robert and Elizabeth Bjork from University of California, Los Angeles (UCLA) have spent years studying what they call "desirable difficulties," conditions that slow down learning in the moment but actually strengthen long-term retention and transfer. Their work, available through their lab at UCLA, and explained well in this short video, points to something a little counterintuitive: some training conditions feel hard and even seem to hurt performance while they're happening, but they pay off far more in the long run than training that feels easy and comfortable.

A few concrete examples of what this looks like in practice:

• Spacing practice out over time instead of cramming it all together

• Testing yourself on material instead of just re-reading or re-studying it

• Varying the conditions under which you practice, rather than keeping everything identical every time

More recent research backs this up while adding an important qualifier. A study by researchers Wesley Pyke, Johan Lunau, and Amir-Homayoun Javadi from the School of Psychology at the University of Kent, looks at how desirable difficulties theory sits alongside cognitive load theory, and how a challenge only helps when it's matched to what the learner already knows and how complex the task actually is. Push someone too far past that match, whether they're a first-year student or a new hire three weeks into a role, and you're not creating desirable difficulty anymore. You're just overwhelming them.

Which brings us to the real point of all this. Difficulty for its own sake isn't the goal, in a classroom or a training room. Meaningful growth is. The struggle only helps when it's calibrated, not when it's just hard.

Why this matters now

There's a reason this conversation feels more urgent than it might have ten years ago, and it's not limited to schools. We're living through a moment where information is essentially free and instantly available, and AI can now do a lot of the recall-and-retrieve work that used to define being "knowledgeable," in the classroom and on the job. That shift changes what's actually worth building in people, whether you're teaching students or upskilling a workforce.

The World Economic Forum's Future of Jobs Report 2025 gives a clear picture of what employers now consider essential. Analytical thinking tops the list, with roughly seven in ten companies rating it as a core skill. Right alongside it: resilience, flexibility and agility, and creative thinking paired with genuine curiosity and a habit of lifelong learning. (Also read our article on career guidance related to the same report.)

Notice that you can't absorb resilience by reading about it. You can't develop analytical thinking by watching someone else do it well. These capabilities get built the same way a muscle does, through repeated exposure to appropriately difficult exercise, whether that's a challenging assignment in school or a genuinely hard problem handed to someone on a project team, not through passive consumption.

Designing learning for growth

If challenge really is the mechanism behind growth, the practical question becomes: how do you actually design for it, rather than just hoping it happens? This applies just as much to someone building a curriculum as it does to someone designing a corporate training program or a single workshop.

David Merrill's First Principles of Instruction gives a solid framework for checking whether a learning experience is doing its job. A few questions worth asking of any course, workshop, or training program:

• Real-world learning: Are learners working through problems that resemble what they'll actually face outside the learning environment, rather than sanitized textbook versions or generic scenarios that never quite match real work?

• Build on prior knowledge: Does the experience activate what people already know before layering something new on top?

• Recall and application: Are learners asked to retrieve and apply knowledge themselves, instead of just recognizing the right answer when it's shown to them, whether that's solving a problem set or actually running a process rather than watching a demo of it?

• Safe failure and feedback: Is there room to fail, get real feedback, and try again, all without it feeling catastrophic, whether the setting is a classroom or a live project with some room for error built in?

• Reflection and transfer: Does reflection give people a way to fold what they've learned into their actual work or life, rather than leaving it stranded in the training room or forgotten the moment the workshop ends?

If most of those answers are yes, you're probably designing something that stretches people in the right way. If they're mostly no, you likely have something comfortable and forgettable instead, the kind of session people rate highly on a feedback form and then never think about again.

Difficulty can still be fun

Here's the part that's easy to miss: none of this means learning has to feel miserable, and it definitely doesn't mean a workshop needs to feel like a punishment to be effective. Seymour Papert, the MIT researcher behind constructionist learning theory, had a phrase for this that's stuck around for decades: "hard fun." His point wasn't that learning is fun despite being hard. It's that the difficulty is often exactly what makes it fun in the first place.

You see this principle of “hard fun” clearly expressed in modern video game design. Researchers Lazaros Michailidis, Emili Balaguer-Ballester and Xun He from Bournemouth University show that when difficulty continuously scales with player capability, cognitive engagement and focus reach their peak. This is why game developers specialize in designing for what psychologist Mihaly Csikszentmihalyi termed "flow state": that zone of deep immersion when a task’s challenge perfectly matches a person’s skill level. Research on player motivation by Richard M. Ryan, Edward L. Deci, and C. Scott Rigby indicates that one of the reasons video games captivate us is not by providing easy wins, but by giving us a genuine sense of progress through challenges that get steadily harder, but stay within reach.

Anyone who's become absorbed in a genuinely tough problem knows this feeling, whether it showed up in a classroom or a strategy session that actually required some real thinking. It's not painless, but it's not unpleasant either. It's engaging in a way that easy things rarely manage to be. And that distinction matters more than it might seem, because genuine interest is really what keeps a person inside a hard problem long enough for it to change them. Without that pull, people just quit at the first sign of friction, whether they're a student closing a textbook or an employee tuning out halfway through a workshop.

Back to the gym

So back to where we started. Nobody expects to walk into a gym, lift the same light dumbbell for a year, and come out stronger. We understand instinctively that growth asks something of us physically. It's worth asking why we don't extend the same logic to how we design learning everywhere else, in schools, in onboarding programs, and in every workshop we build with the best of intentions.

Maybe the real question isn't whether learning should be comfortable. Maybe it's whether we're brave enough to design for the kind of discomfort that builds something.