All topics › Behavior discipline emotional regulation
How should parents use praise and rewards to encourage good behaviour?
How should you use praise and rewards to encourage good behaviour — and can you overdo it?
Describe what your child did well rather than labelling who they are: "you worked hard on that puzzle" beats "you're so clever", and "thank you for putting the blocks away" beats a generic "good job" [3][4][5]. Tangible rewards — stickers, treats, prizes — can backfire when they're promised in advance for things your child already enjoys doing, but they're a pragmatic option for boring tasks nobody would do otherwise, with uncertain long-term effects once the rewards stop [1][2]. Most of this comes from small lab studies, so treat it as sensible guidance rather than proven law.
The most famous experiment is over 25 years old and still shapes the advice. Mueller and Dweck gave 10- to 12-year-olds puzzles, then praised them for intelligence ("you must be smart") or for effort ("you must have worked hard"). After a subsequent failure, the intelligence-praised children gave up sooner, enjoyed the task less, blamed their ability, and performed worse than the effort-praised children [3].
The pattern holds outside the lab, at least once. Researchers videotaped about 50 families at home when their children were 1, 2, and 3, and followed up at ages 7–8. Parents who used more process praise — praising effort, strategies, and actions — had children with more resilient, "growth" attitudes toward challenges five years later (r = .35, p = .01). Person praise ("you're so smart") and the total amount of praise didn't predict anything [4]. Two honest caveats: it was one small observational study, and children who persist more may also elicit more process praise — the arrow could point both ways [4].
A review of the wider praise literature converges on the same rule of thumb: praise works best when it feels informational ("here's what you did well") rather than controlling, and when it targets process rather than fixed traits — especially when the child later hits difficulty [5]. A 2021 revisit of the praise literature two decades on affirmed the same conceptual variables: praise enhances motivation and perseverance when it implies success comes from controllable, malleable forces (strategy, effort), promotes autonomy, builds a resilient sense of competence, and provides specific, accurate information about performance [8].
Probably not in the way parents worry about. In the longitudinal family study, the total amount of praise parents gave did not predict children's later motivational frameworks at all — what mattered was the proportion that was process praise [4]. The reviews agree: the risk isn't volume, it's type and contingency — praise that is controlling, inflated, or focused on fixed traits, especially when the child then meets difficulty [3][5]. So narrating your toddler's block-building with enthusiastic specifics isn't spoiling anything. The risky version is "you're so clever" delivered right before the hard bit.
The classic 1973 study: preschoolers who loved drawing were promised an award for drawing; afterwards, they chose to draw less in free play than children who got a surprise award or none at all [1]. The mechanism proposed: the child reinterprets their own behaviour — "I did it for the prize" — and the original enjoyment fades.
A meta-analysis of 128 studies refined this considerably. The composite across all reward types was a small undermining of later free-choice engagement (d = −0.24, k = 101), driven by expected, tangible rewards — tangible rewards alone undermined free-choice engagement (d = −0.34), and engagement-contingent rewards most of all (d = −0.40) [2]. But verbal rewards (praise) tended to enhance motivation (free-choice d = 0.33), and unexpected rewards didn't undermine it. Praise and payment are not the same thing. An earlier meta-analysis (Cameron & Pierce, 1994) concluded the undermining effect was limited and easily prevented in applied settings — but even its authors agreed that tangible, expected rewards on already-interesting tasks are the risky combination, which is exactly where Deci et al. found the effect [2][6].
The practical translation: avoid promising prizes for things your child already loves doing. Do use rewards pragmatically for things nobody enjoys — toilet training, boring chores, the dull repetitions of practice. That's where extrinsic motivation earns its keep, and there's no intrinsic enjoyment to undermine [1][2].
The parent-training programmes with trial evidence don't just tell parents to "praise more". They train a specific skill: frequent, immediate, specific praise during child-led play — "you're building that tower so carefully" — practiced in-session with coaching, then extended into daily life [7]. The point is to shift the attention ratio so that good behaviour gets far more airtime than misbehaviour, making praise the background music of the day rather than a special-occasion speech [5][7]. If you take one practical thing from this topic, it's that: narrate the good stuff, out loud, in the moment.
Here's the awkward truth about the strongest real-world evidence: in the parent-training programmes with trial data (Triple P, Incredible Years), praise and "differential attention" — noticing and rewarding the behaviour you want — are core components, and programmes that teach them get larger effects [7]. But they're never tested alone. So we know the package works; we can't say how much is the praise [7].
The same goes for sticker charts: enormously popular, recommended by health visitors everywhere, and essentially untested as a standalone strategy in typical families. No good randomised trial of home sticker charts was located for this topic. They're a reasonable application of reward principles — small, immediate, visual, faded over time — but "widely used" is not "proven" [7].
Praise is the cheapest parenting tool there is — free, unlimited, and available at 3am. That's worth saying, because the alternatives (reward systems, charts, prize boxes) cost money, time, and mental bandwidth, and they create a second job: tracking, negotiating, and eventually fading the system. If you find yourself haggling with a four-year-old over how many stickers a tidy bedroom is worth, the system is now costing you more than it's saving.
There's a subtler parent-side trap in the person-praise research. "You're so clever" feels warm and loving in the moment — it often says more about the parent's pride than the child's effort. Shifting to process praise ("you kept trying even when it was hard") doesn't just change the child's framework; it changes what you notice. Parents in the longitudinal study weren't trained to do this — the ones who naturally praised process had children who handled challenge better years later [4].
One more honest gap: nobody has studied what praise and reward strategies do to parental stress, nagging cycles, or family finances. If sticker charts reduce your evening battles, that benefit is real even if no trial measured it — and if they increase the battles, ditching them is equally rational.
Effect sizes and qualitative findings only — no per-1,000 data exist for these outcomes.
| Strategy | Finding | Evidence rating |
|---|---|---|
| Process/effort praise (vs intelligence praise) | Better persistence, enjoyment, and performance after failure in lab studies [3]; predicts resilient attitudes 5 years later in one longitudinal study [4] | C |
| Specific/descriptive praise (vs generic) | Recommended by reviews and programmes; direct experimental evidence thin [5][7] | C |
| Expected tangible rewards for already-enjoyed activities | Undermine later free-choice engagement — larger for tangible, expected, engagement-contingent rewards (composite d = −0.24 across all reward types; tangible-only d = −0.34) [1][2] | B |
| Verbal praise / positive feedback | Tends to enhance intrinsic motivation; not interchangeable with tangible rewards [2][5] | B |
| Rewards for boring/unenjoyable tasks | No intrinsic motivation to undermine; pragmatically useful [1][2] | C (expert consensus built on the lab findings) |
| Sticker charts at home | Widely used; no good standalone trials located | D (gap) |
| Differential attention ("catch them being good") | Core component of effective programmes; never tested alone [7] | C (bundled) |