Kid software has converged on a default kit: experience points, levels, streaks, badges, leaderboards. The studio ships none of it. Edward Deci’s 1971 experiments and the half-century of Self-Determination Theory research that followed found that extrinsic rewards reliably erode the intrinsic motivation they sit on top of. The kids we watch in cohort confirm it. We replaced the reward layer with the artifact: you made a thing, it ships to a URL, your friend can play it. That’s the loop.
What gamification promises.
Walk into any conference on kid software and the talks rhyme. Engagement is up since we added streaks. Return-visits climbed 40 percent after we shipped levels. Parents love the badge sheet because it shows progress. Onboarding completion doubled when we put XP on the welcome screen. The metrics aren’t fake. Gamified apps really do get more sessions, more taps, more time logged. The question is what those sessions are for.
The pitch is that game mechanics borrow the engagement of games and apply it to things that aren’t games. A homework app with a streak chart is supposed to feel more like Duolingo. A reading app with a level-up animation is supposed to feel more like Pokémon. Jane McGonigal’s 2011 book Reality is Broken made the most ambitious version of this argument: games are training environments for the kind of effortful, repeated practice that learning requires, and importing their feedback loops into other domains could make those domains stickier.1 She wasn’t wrong about what games do. She was optimistic about whether the trick transfers.
The version that ends up in kid software is usually the thin one. An XP bar on top. A level number. A streak counter that emails the parent when the kid skips a day. Badges that drop with a small animation. The mechanics are decoupled from the underlying activity. The kid learns to chase the bar, and the bar fills whether or not the activity worked.
What the motivation research actually says.
In 1971, Edward Deci ran a now-famous experiment at Carnegie Mellon. Two groups of college students worked on a puzzle called Soma, which most participants found genuinely fun. One group got paid for each puzzle they solved. The other didn’t. After a few sessions Deci took the money away and gave both groups free time with the puzzles. The unpaid group kept playing. The previously-paid group stopped. They had been doing it for the money. When the money went away, so did the interest.2
Mark Lepper, David Greene, and Richard Nisbett replicated the shape of the finding with preschoolers and felt-tip markers in 1973. Kids who already liked drawing were promised a “Good Player” certificate if they drew. A control group drew without any promised reward. Weeks later, in free play, the certificate group drew less than the no-reward group. The reward had converted intrinsic interest into a transaction, and removing the transaction killed the interest.3 The literature came to call this the overjustification effect.
Deci and Richard Ryan spent the next four decades developing Self-Determination Theory around this and adjacent findings. Their 1985 book and the meta-analyses that followed organized the result: tangible, expected rewards offered for performing an intrinsically interesting activity reliably reduce intrinsic motivation for that activity once the reward is withdrawn.4 The effect is robust across ages, domains, and cultures. It is one of the most-replicated findings in motivation psychology.
Alfie Kohn’s 1993 book Punished by Rewards took the polemical version of the argument to schools and workplaces. Stars on a chart, pizza for reading, employee-of-the-month plaques. Kohn called them “control disguised as praise,” and argued the long-term cost was always the same shape: the activity itself stopped being the point.5 Mitchel Resnick’s Lifelong Kindergarten, two decades later and from the opposite end of the temperament spectrum, lands in the same place. Creative tools for kids should foreground projects, peers, and play. Points and badges crowd those out.6
The argument isn’t that all rewards are bad. Unexpected praise after a hard piece of work, recognition that comes from a person rather than a system, feedback that names something specific the kid did. Those still help. The thing the research warns against is the recurring extrinsic structure layered on top of an activity the kid would otherwise enjoy for its own sake.
The bar fills whether or not the activity worked. The kid learns to chase the bar. On what a streak counter is actually training
What we’ve watched in cohort.
Most kids who arrive in our cohort have come from gamified apps. Duolingo, Prodigy, Khan Kids, Scratch with badges turned on. They show up in week one with a specific expectation. They finish a wizard and look at the screen waiting for something. When nothing pops up, they ask. “Do I get a point for that?” “Is there a level?” “Where’s my badge?” The question is sincere. The reward layer is what they’ve been trained to expect a piece of kid software to give them.
The first session, sometimes the second, can be a little flat for these kids. They’re looking for the bar to fill. The bar is the loop they know. We tell them the artifact is the reward. They half-believe us.
By week two something shifts. The kid has a project on disk that has changed shape across a dozen wizards. They’ve published it to a URL their friend can open. The friend played it and sent back a screenshot of a high score. The kid has forgotten about XP because something realer is happening: the thing they made is being used by someone they know. That’s the moment the gamified-app instinct gets quiet. The artifact is doing the motivational work the badge used to do, and the artifact is more durable, because the artifact actually exists in the world.
We’ve seen the inverse pattern too. A few cohort kids ask us partway through whether we’ll add levels. We say no and explain why. They shrug. Two weeks later when they ship something they care about, they don’t ask again. The question only persists while the reward layer is still doing the heavy lifting. Once the artifact takes over, the question goes away on its own.
What we built instead.
The studio replaces the reward layer with the artifact. Make-test-debug-improve is the loop, and the loop’s payoff is the thing the kid is making becoming the thing they wanted. That sentence sounds soft until you watch it land. A nine-year-old who’s been iterating on a level for an hour, runs a playtest, watches the new mechanic finally do what they intended, and says “oh, that’s it.” No XP popped. No badge dropped. The reward was that the thing worked.
There are places where we use small affordances that look adjacent to gamification but aren’t. The family dashboard shows project counts and deploys. Parents see when a kid shipped. We celebrate a v5 of a project the way you’d celebrate a fifth draft of a novel: by reading it, not by handing out a sticker for sustained effort. The dashboard is information, not reward. It doesn’t advance, doesn’t level up, doesn’t streak.
The harder design question, and the one we get asked most, is what to do about the kid who has trouble starting. Without a streak chart, what makes them come back tomorrow. Our answer is the same answer Resnick gives: a project they care about. The kid comes back because the thing they’re making isn’t finished and they have an idea about what to try next. That motivation is more fragile than a streak in the short term and more durable than one in the long term. Our piece on autonomy and authorship walks through why this trade-off is the right one to make.
The honest counter-argument.
The steel-man case for gamification in kid software isn’t silly. Some skills are genuinely boring before they’re fluent. Multiplication tables. Foreign-language vocabulary. Touch typing. The activity isn’t intrinsically rewarding at the start; intrinsic motivation can’t do the work yet. A streak chart or a points system can carry the kid through the early reps until the activity itself starts to pay. Duolingo’s case is roughly this one, and it’s defensible.
The studio isn’t that kind of activity. Making a thing is intrinsically interesting to most kids on minute one. The motivation is already there. Adding XP on top of an activity the kid already wanted to do is exactly the configuration Deci ran his experiment on. The reward doesn’t amplify the interest. It replaces it, and then takes the interest with it when the reward goes away.
So the design rule we ended up with is narrow. If the activity is intrinsically motivating, don’t layer extrinsic rewards on it. Find a way to make the natural reward of the activity legible to the kid instead. Show the artifact. Show what changed. Show the friend who played it. The motivation literature backs this up across half a century of studies, and the kids we’ve watched move through cohort confirm it session by session.
If you want to see what the artifact-as-reward loop looks like in practice, Theo’s game at god-games.vercel.app is the cleanest demo. Or play through the mini-studios at tellandshow.ai/try to feel the absence of the reward layer for yourself. The thing that’s missing is the thing we deliberately didn’t add.
References
- Jane McGonigal, Reality is Broken: Why Games Make Us Better and How They Can Change the World, Penguin, 2011. The strongest book-length argument for importing game mechanics into non-game domains.
- Edward L. Deci, “Effects of externally mediated rewards on intrinsic motivation,” Journal of Personality and Social Psychology, vol. 18, no. 1, 1971. The Soma puzzle study and the first published demonstration of what was later called the overjustification effect.
- Mark R. Lepper, David Greene & Richard E. Nisbett, “Undermining children’s intrinsic interest with extrinsic reward,” Journal of Personality and Social Psychology, vol. 28, no. 1, 1973. The felt-tip-marker replication with preschoolers.
- Edward L. Deci & Richard M. Ryan, Intrinsic Motivation and Self-Determination in Human Behavior, Plenum, 1985. The foundational Self-Determination Theory book. See also Deci, Koestner & Ryan’s 1999 meta-analysis in Psychological Bulletin.
- Alfie Kohn, Punished by Rewards: The Trouble with Gold Stars, Incentive Plans, A’s, Praise, and Other Bribes, Houghton Mifflin, 1993. The popular-press synthesis applied to schools and workplaces.
- Mitchel Resnick, Lifelong Kindergarten: Cultivating Creativity through Projects, Passion, Peers, and Play, MIT Press, 2017. The four-Ps framework for creative learning environments, developed at media.mit.edu/groups/lifelong-kindergarten.