This Psychological Trick Makes Rewards Backfire

In the 1970s, researchers ran a study where children were asked to draw pictures. Some children were told ahead of time that they would receive a reward. Others were not promised anything. After the drawings were finished, the researchers kept watching the children in class for the next few weeks.

The results were surprising, and they matter for game design. Designers often want to motivate players: teach them a new mechanic, encourage them to use a feature, or keep them playing for longer. The popular solution is a goal paired with a reward. Do this, get that.

Quests lead to experience points. Challenges unlock cosmetics. Achievements wrap a goal and a reward into one tidy package. But goals and rewards do not always have the intended effect. In some games, they can do the opposite.

When Klei was building the first prototype of Don't Starve, playtesters had no idea how to begin. They became stuck almost immediately. A few hints helped them get over the first hump, and once they understood the basics, they could experiment, explore, and start having fun.

Klei responded by adding small tutorial-like quests. Survive a certain number of nights. Find a certain number of items. The structure solved the onboarding problem, but it created a larger design problem.

Players focused almost entirely on the quests. Everything outside the task list became noise. They optimized their play in dull ways just to complete the current instruction. They avoided risk because risk could make the quest fail. Then, when the quests ran out, their motivation collapsed.

By structuring the game as a series of explicit tasks, the design accidentally taught players to depend on those tasks for meaning. Klei eventually cut the quest system and solved the onboarding issue with subtler interface hints, such as highlighting important craftable items.

That lesson matters because Don't Starve is about experimentation, exploration, and player-guided discovery. In a game like that, explicit goals can narrow the player's imagination. Instead of asking "what can I try?", the player asks "what does the checklist want?"

Outer Wilds was built around avoiding that problem. The developers deliberately avoided explicit directions about where to go or what to achieve. The miniature solar system works because players are driven by curiosity. They see something strange, chase it, learn a new connection, and set their own next objective.

If a game wants curiosity to be the engine, a task list can pull attention away from the very thing the game is trying to cultivate.

Zach Barth's puzzle games, including Exapunks and Shenzhen I/O, are about designing automated machines. If the machine works, it works. But much of the fun comes from going back and refining the solution: making it smaller, faster, cheaper, or more elegant.

In early commercial games like SpaceChem and Infinifactory, there were achievements that encouraged that kind of optimization. One might ask players to beat a level under a certain number of cycles. Later games dropped those achievements.

The reason is that these games already have a better reward system. They show personal scores, friend comparisons, leaderboards, and histograms that reveal how your solution compares to everyone else's. Those measurements invite the player to set a personal or social goal rather than chase an arbitrary threshold.

A goal you set yourself can be more powerful than a goal the game assigns to you. If the game is about improving your own skill, comparative metrics and personal bests can motivate without turning mastery into a checklist.

Mini Metro shows the same tension from a different angle. The game is about personal growth and high scores, so its developers tried to avoid goal-and-reward structures that could become a means to an end.

The game does have unlockable cities, partly to keep the opening choice set manageable. But that structure can change how some players read the game. They play each city until they hit the required threshold, unlock the next one, and once every city is available, they feel as if the game is complete.

Goals are checklists, and checklists can be exhausted. Measurements of skill are different. Scores, leaderboards, and personal bests do not really end. You can continue improving forever, which helps explain why a game like Tetris can remain compelling for decades.

The psychology behind this is usually described through extrinsic and intrinsic motivation. Extrinsic motivation means doing a task for a reason outside the task itself, usually to get a reward. A job is the obvious example.

Intrinsic motivation means doing something because the activity itself is enjoyable or meaningful. A hobby is the obvious example. Intrinsic motivation is often stronger, and it tends to last longer. People can enjoy a hobby for a lifetime. Extrinsic motivation lasts only as long as the reward keeps coming.

That brings us back to the drawing study. The children had already shown interest in drawing before the study began. They were intrinsically motivated. But when one group was promised a reward for drawing, those children later showed less interest in drawing on their own, and their pictures were judged lower in quality.

This is called the overjustification effect. When an external reward is attached to an activity people already find intrinsically motivating, they can become less interested in the activity itself.

Other studies suggest that rewards can also make people less creative, worse at problem solving, more likely to cheat, and more likely to lose motivation when the rewards stop. They were once happy to do the thing for its own sake. After the reward is added and removed, that original desire can be weaker.

The same idea applies cleanly to games built around exploration, creativity, expression, or growth. These games ask players to set goals, follow curiosity, improve themselves, and make meaning without needing a prize at every step. Add explicit tasks, progress meters, and achievements too aggressively, and motivation can narrow instead of expand.

The risk is not just that a player becomes greedy for prizes. The risk is that the reward changes what the player sees. Creative solutions become distractions from the fastest route to completion. Self-improvement becomes less appealing than hitting the listed threshold. Open-ended play becomes a search for the next marker.

Designers then inherit a new obligation. If players have been trained to depend on explicit goals and rewards, the game must keep producing them. The moment the drip feed ends, those players may feel as if the game has stopped giving them a reason to care.

That is a dangerous tradeoff for any game whose best moments come from player-driven curiosity.

None of this means that goals and rewards are always bad. Some players are not comfortable generating their own direction. For every Minecraft player who happily creates their own fun, another player feels lost without a prompt.

The open-ended detective game Her Story captures this beautifully. One player said that it is up to you to decide when you are satisfied with the information you have found. Another replied: how do I decide when I am satisfied?

That question is funny, but it is also real. Goals and rewards can provide structure, progression, and confidence. The design challenge is to provide that structure without crushing curiosity, creativity, or self-directed mastery.

A safer approach is to use large, overarching goals that players can complete in many ways rather than restrictive step-by-step instructions. Let the goal frame the possibility space instead of prescribing the route through it.

Comparative metrics can also work better than absolute thresholds. Leaderboards, histograms, and personal bests ask players to improve in relation to themselves or others, not merely pass a fixed line. Optional goals can help too, as in Hitman's challenges. Hidden achievements can preserve discovery, as in Outer Wilds.

The key is to avoid making the reward structure louder than the core activity. If the game is about exploration, the player should still be exploring. If it is about optimization, the player should still be optimizing for reasons that feel personally meaningful.

There is one kind of reward that appears less likely to trigger the overjustification effect: the unexpected reward. In the drawing study, a third group of children was not promised anything in advance, but received a reward afterwards as a surprise. Those children later spent the most time drawing, even if only slightly more than the unrewarded group.

Research suggests that rewards can help in intrinsically motivating situations when they are unexpected, reasonably low value, and tied to the performance of the action. In games, Overwatch's Play of the Game is a useful example. It does not grant much, but it gives the featured player a short ego boost by celebrating a standout moment.

Nintendo often uses this idea in exploration. Super Mario Odyssey does not tell players to climb a tricky rooftop with Mario's advanced movement, but it may place a small cache of coins there as a pat on the back. Breath of the Wild fills suspicious nooks with surprises like Korok seeds. The reward encourages exploration because it feels like a discovery, not a wage.

That is the narrow path. Rewards are powerful, but they are not neutral. Used carelessly, they can shrink a player's motivation. Used gently, optionally, comparatively, or unexpectedly, they can support the kind of play the game was already trying to inspire.