A prize can spoil the play

In the early 1970s, researchers asked children to draw. One group was told a prize was coming. They drew, they got the prize, and later, when nobody was paying, they drew less than the children who had never been offered anything. The reward had not sweetened the activity. It had replaced it. The drawing became the chore you do for the prize.

Games relearn this every time a bar, a daily, or a currency sits on top of a verb that was already fun. Players optimise the prize and tolerate the verb. If the verb was the game, you have trained them to hurry through it. Rewards are not evil. They are loud. Use them to teach a verb that is not yet fun, and take them away, or shrink them, once the verb can stand up. Leave them on a verb that was already the pleasure and you will wake up running a store.

A reward changes what a player is paying attention to, and that change is not always the one you designed for. The clearest evidence comes from a study in the 1970s, in which researchers asked children to draw pictures. One group was told in advance that a reward was coming. Another group was promised nothing at all. The researchers then watched what those children did in class over the following weeks.

What they measured has direct consequences for anyone building a game. You usually have something you want the player to do: learn a mechanic, try a system they have been ignoring, come back tomorrow. The standard fix is to pair an instruction with a prize. Complete this, collect that.

So quests pay out experience points, challenges hand over cosmetics, and an achievement bundles the instruction and the prize into a single line of text. The bundle is tidy. It is also not neutral, and in certain games it works against the design.

Spelling out tasks teaches players that meaning arrives from the task list, which is a hard lesson to unteach. Klei ran into this on the first prototype of Don't Starve. Testers opened the game and froze, with no idea what the opening move was. A few hints carried them past that first wall, and once the basics clicked they were free to poke at things, wander, and enjoy themselves.

The studio's answer was a set of light tutorial quests. Live through so many nights. Collect so many items. Onboarding got easier and a bigger problem took its place.

Attention narrowed onto the list. Anything the list did not mention read as background. Players ground out dull, optimal lines because dull and optimal is what finishes an objective, and they steered clear of risk because a risk can cost you the objective. Then the quests ran dry and so did the reason to keep playing.

Framing the game as a sequence of stated tasks had trained players to lean on those tasks. Klei pulled the quest system out and fixed onboarding a quieter way instead, by drawing the eye to the craftable items that matter through the interface itself.

When curiosity is the engine, an instruction competes with it rather than feeding it. Don't Starve lives on experimentation, exploration, and discoveries the player makes on their own terms. Stated goals shrink that space. The question in the player's head stops being "what happens if I try this?" and becomes "what is the checklist asking for?"

Outer Wilds was designed around dodging exactly that trap. Its developers held back from telling you where to head or what counts as progress. The tiny solar system holds together because curiosity carries the player through it: something looks off, you go and look, you learn how a piece connects, and you pick your own next destination.

A task list, dropped into a game like that, aims attention at the list and away from the impulse the whole design exists to feed.

A goal a player sets for themselves tends to pull harder than one the game assigns. Zach Barth's puzzle games, Exapunks and Shenzhen I/O among them, ask you to design automated machines. A working machine is a pass. The real appeal is coming back to it and making it tighter, quicker, cheaper, or simply more elegant.

His earlier commercial games, SpaceChem and Infinifactory, nudged that refinement along with achievements, some of them asking you to clear a level inside a set number of cycles. The later games left those achievements out.

They did so because a better system was already in place. Those games surface your own scores, comparisons against friends, leaderboards, and histograms showing where your solution sits against everyone else's. A measurement like that hands the player a target of their own choosing, personal or social, instead of an arbitrary number picked by the designer.

That distinction matters most in games about getting better at something. Comparative figures and personal bests can drive a player hard without reducing mastery to a list of boxes.

Mini Metro runs into the same tension from another direction. It is built on personal improvement and high scores, so its developers steered away from goal-and-reward structures that might turn the play into a means to an end.

Cities do unlock, in part so the menu is not overwhelming on day one. That structure quietly reframes the game for some players. They work a city until the unlock threshold falls, move on, and when the last city opens up they read the game as finished.

A goal is a box, and boxes run out. A measurement of skill behaves differently. Scores, leaderboards, and personal bests have no last entry. Improvement is always available, which is a large part of why a game like Tetris still holds people decades on.

Psychology splits this into extrinsic and intrinsic motivation. Extrinsic means you do the thing for something outside the thing, generally a payout. Work is the everyday case.

Intrinsic means the activity is worth doing in itself, because it is enjoyable or it means something to you. A hobby is the everyday case. It is usually the stronger of the two, and it has far more staying power. Hobbies survive decades. Extrinsic motivation survives as long as the payouts keep landing.

Which returns us to the children and their drawings. All of them liked drawing before the study started, so the intrinsic motivation was there already. The group promised a reward for drawing went on to draw less of their own accord afterward, and the pictures they made during the study were rated lower in quality.

That result has a name: the overjustification effect. Bolt an external reward onto something a person already enjoys for its own sake, and their interest in the thing itself can fall away.

Further research points the same direction. Rewards have been linked to less creativity, weaker problem solving, more cheating, and a sharper drop in motivation once the rewards stop. The activity was its own reason to begin with. Add a prize and take it away, and that original pull can come back diminished.

Transplant this into games about exploring, creating, expressing yourself, or growing, and it fits without adjustment. Such games count on players choosing their own goals, chasing what interests them, improving, and finding meaning with no prize attached to each step. Layer on tasks, progress bars, and achievements too heavily and motivation contracts rather than widens.

The danger is less that players turn greedy and more that the reward edits what they notice. An inventive solution starts to look like a detour off the quickest completion route. Getting better at the game loses to clearing the stated number. Open play turns into a hunt for the next marker on the map.

That leaves the designer holding a new bill. Players taught to run on stated goals and prizes have to be kept supplied with them. Cut the feed and some of them will conclude the game has stopped offering a reason to keep going.

For any game whose best moments come from a player following their own nose, that is a poor trade.

Goals and rewards are not a design flaw, and plenty of players want them. Producing your own direction is a skill, and not everyone enjoys it. For each Minecraft player who invents their own project without prompting, someone else opens the game and feels adrift.

Her Story, an open-ended detective game, put the split on display. One player described it as being up to you to judge when you have found enough. Another player answered: how am I supposed to know when I have found enough?

The reply is funny and also fair. Structure, a sense of progress, and confidence are real things that goals supply. The craft is in supplying them without stamping out curiosity, creativity, or the pleasure of getting good at something on your own.

Broad goals that admit many solutions do far less damage than tight step-by-step orders. Let a goal mark out the territory rather than draw the path across it.

Relative measures also tend to sit better than fixed bars. Leaderboards, histograms, and personal bests ask a player to improve against themselves or against other people, which is an open-ended request rather than a line to step over once. Making goals optional helps as well, the way Hitman's challenges do. Keeping achievements hidden protects the surprise, the way Outer Wilds does.

Whatever the shape, the reward structure should stay quieter than the activity it sits on. An exploration game should still have you exploring. An optimization game should still have you optimizing for a reason that means something to you.

One kind of reward seems to slip past the overjustification effect: the one nobody saw coming. The drawing study ran a third group, promised nothing beforehand but given a reward afterward out of the blue. Those children went on to spend the most time drawing, though only a little more than the group that got nothing.

The research suggests rewards can sit comfortably alongside intrinsic motivation when they arrive unannounced, stay modest, and follow directly from something the person did. Overwatch's Play of the Game is a clean example. The prize is close to nothing in material terms, and it still hands the featured player a moment of pride for a standout play.

Nintendo leans on the same idea to encourage exploration. Super Mario Odyssey never instructs you to work Mario's advanced moves into scaling an awkward rooftop, but there may be a handful of coins waiting up there for the player who tries. Breath of the Wild tucks Korok seeds into the nooks that look worth investigating. It reads as a discovery rather than a paycheck, so it feeds the exploring instead of replacing it.

That is the thin line to walk. Rewards are powerful and they are never neutral. Handled bluntly they can shrink what a player wants from your game. Kept gentle, optional, comparative, or unannounced, they back up the play the game was built to produce. You find out which one you built by getting it running and playing it yourself.

In the Flockbay app, build the enjoyable action with no currency. Play it. If you would do it again, do not add a prize yet. If you would not, the action is the problem, and a prize will only hide it for a session.

An AI game maker will offer currencies early. A start-here template is small enough to feel the action naked.