Expected tangible rewards reduce later free-choice engagement with a rewarded task, while verbal praise does not¶

Claim · All claims
Evidence · 5 studies · 2 quant-synthesis r2, 2 review r2, 1 causal r? · q3–q4 · i1 small

Extrinsic rewards — particularly tangible, expected rewards offered contingent on task engagement or completion — can reduce learners' intrinsic motivation for activities they initially found interesting.

Subclaims¶

q4 i1 A 128-study meta-analysis found that engagement-contingent, completion-contingent, and performance-contingent tangible rewards significantly reduced learners' later free-choice engagement with the rewarded activity, though the pattern was contested by an earlier meta-analysis reaching the opposite overall conclusion. → Deci Koestner Ryan 1999

q4 i? A competing 96-study meta-analysis found no overall undermining effect of reward on free-choice intrinsic motivation, with a significant negative effect confined to expected tangible rewards given merely for doing a task. → Cameron Pierce 1994

q3 i? Paying university students to do puzzle-solving and headline-writing tasks they already found interesting reduced their subsequent free-time engagement with those tasks relative to unpaid students, even though both groups started out equally interested. → Deci 1971

q4 i? The effect is substantially weaker or absent when pay is a flat, hourly rate rather than tied piecemeal to output, and when the task itself is well-defined with a clear standard for high-quality performance. → Cameron and Pierce 1994

q3 i? A synthesis of the reward/intrinsic-motivation literature confirms the undermining effect occurs often enough that extrinsic rewards for already-interesting tasks need to be used selectively and thoughtfully, rather than never or always. → Deci Koestner and Ryan 2001

Evidence¶

Deci Koestner Ryan 1999¶

Deci, E. L., Koestner, R., & Ryan, R. M. (1999). A meta-analytic review of experiments examining the effects of extrinsic rewards on intrinsic motivation. Psychological Bulletin, 125(6), 627–668. doi:10.1037/0033-2909.125.6.627

q4 · meta-analysis · i1 · small to medium effects, d=-0.28 (performance-contingent) to -0.40 (engagement-contingent); the printed range spans the i1–i2 boundary and is binned by its smaller end, so it covers all three contingencies the subclaim names · n=128 studies · quant-synthesis · r2

A meta-analysis of 128 experiments compared free-choice persistence and self-reported interest in a rewarded activity, after the reward was withdrawn, between reward and no-reward groups. Engagement-, completion-, and performance-contingent tangible rewards each significantly undermined free-choice behavior, and engagement- and completion-contingent rewards also undermined self-reported interest; all tangible rewards and all expected rewards undermined free-choice behavior overall. By contrast, positive (verbal) feedback significantly enhanced both free-choice behavior (d = 0.33) and self-reported interest (d = 0.31), and tangible rewards were more detrimental for children than for college students. The authors present this as a defense and extension of the overjustification/self-determination account against contemporaneous critiques, including Cameron & Pierce's own reanalyses.

Cameron Pierce 1994¶

Cameron, J., & Pierce, W. D. (1994). Reinforcement, reward, and intrinsic motivation: A meta-analysis. Review of Educational Research, 64(3), 363–423. doi:10.3102/00346543064003363

q4 · meta-analysis of between-groups experiments (plus a secondary within-subject analysis) · i? · no effect size in the abstract; described as no overall effect and a minimal negative effect for expected tangible rewards · n=96 studies (main analysis), plus 5 studies in a secondary within-subject analysis · quant-synthesis · r?

A hierarchical meta-analysis of 96 between-groups experiments compared rewarded and non-rewarded participants on four measures of intrinsic motivation, with a secondary analysis of 5 within-subject studies. Contrary to the overjustification hypothesis, the authors report that reward does not decrease intrinsic motivation overall, and that verbal praise increases it; a negative effect appears only for expected tangible rewards given simply for performing a task, and only on the free-choice behavior measure (not on self-reported attitudes), with the effect described as minimal. (Read as an abstract; the abstract gives no effect sizes.) The flat-rate and well-defined-task moderators in the subclaims come from the second write-up of this study kept in the comment below; the abstract does not state them, so they rest on that write-up's reading of the full text and are unchecked here. The two write-ups coded the design q4 and q3; it is q4 here, a meta-analysis of 96 experiments, as Deci, Koestner & Ryan's (1999) 128-experiment meta-analysis is coded on this page and on autonomy supports intrinsic motivation.

Deci 1971¶

Deci, E. L. (1971). Effects of externally mediated rewards on intrinsic motivation. Journal of Personality and Social Psychology, 18(1), 105-115. doi:10.1037/h0030644

q3 · peer-reviewed lab experiment · i? · no source text available to check; the entry prints no effect size · n=modest (university sample) · causal · r?

University students worked on interesting puzzle-solving and newspaper-headline-writing tasks across several sessions; one group was paid for the activity and another was not, though both groups reported equal initial interest. Students who had been paid spent measurably less free time on the same activities afterward than students who had never been paid — the original demonstration of the overjustification effect, and the source study behind the "hugging and bridging" caution against over-relying on extrinsic reward systems (see Behaviorism, Reinforcement Theory).

Eisenberger and Cameron 1996¶

Eisenberger, R., & Cameron, J. (1996). Detrimental effects of reward: Reality or myth? American Psychologist, 51(11), 1153-1166. doi:10.1037/0003-066X.51.11.1153

q3 · narrative synthesis of experimental literature · i? · no source text available to check; the entry prints no effect size · n=large (aggregated) · review · r?

Reaching a similar conclusion to Cameron and Pierce (1994) from a partly overlapping literature, this synthesis argues the overjustification effect, while real, has often been overstated as a blanket argument against all extrinsic reward, and identifies boundary conditions (reward contingent on mere participation vs. on producing more of an activity) under which rewards are unlikely to undermine intrinsic motivation.

Deci Koestner and Ryan 2001¶

Deci, E. L., Koestner, R., & Ryan, R. M. (2001). Extrinsic Rewards and Intrinsic Motivation in Education: Reconsidered Once Again. Review of Educational Research, 71(1), 1-27. doi:10.3102/00346543071001001

q3 · review of a meta-analysis (Deci, Koestner & Ryan, 1999) · i? · the abstract prints no effect size; the full text may · n=large (aggregated across studies) · review · r2

Responding directly to Cameron and Pierce's (1994) conclusion that the undermining effect was minimal, this article reviews the results of the authors' 1999 meta-analysis, which it says showed Cameron and Pierce's analysis was seriously flawed and that tangible rewards do have a substantial undermining effect, supporting cognitive evaluation theory. It discusses what that means for rewards in classrooms: they should be used selectively and thoughtfully, not treated as a costless motivational tool.

Studies bearing on this claim¶

Do rewards and praise undermine voluntary engagement? This table gathers the studies behind this claim and the 12 narrower claims that bear on it: 17 distinct studies (5 review, 4 causal, 3 quant-synthesis, 3 theoretical, 1 associational, 1 design), q1–q4, 2 of 17 with an effect size coded. One row per study, codes as each entry records them; no effect is averaged and no verdict is drawn. Built by scripts/synthesize_claim.py from eval/synthesis/; a row that looks wrong is fixed on the claim it comes from.

Discussion¶

Mechanism. The standard explanation draws on self-determination theory: rewards can shift a learner's perceived locus of causality from internal ("I do this because I enjoy it") to external ("I do this to get the reward"), undermining the needs for autonomy and competence that sustain intrinsic motivation. This is closely related to the claim that autonomy supports intrinsic motivation.

Moderators matter. The undermining effect is not uniform across reward types. Tangible, expected, task-contingent rewards (given simply for doing the task) show the strongest undermining; verbal praise and unexpected rewards show weaker or no undermining; and rewards contingent on performance quality or achievement of standards can sometimes maintain or even enhance motivation, plausibly because they convey competence information rather than mere control. Rewards for tasks that are already intrinsically uninteresting pose little risk — there is little intrinsic motivation to undermine — and may be the appropriate tool there.

Boundary conditions and open questions. The effect is typically demonstrated on free-choice persistence and self-reported interest in novel, playful tasks; how strongly it generalizes to sustained engagement in structured academic settings remains debated. Designers should also weigh the risk against behaviorist uses of reinforcement — see behaviorism — since reward systems that successfully shape behavior in the short term may simultaneously erode the interest that sustains behavior once rewards stop.

Design implications. Where learners already show interest, prefer informational feedback over controlling incentives; if rewards are necessary, make them unexpected, intangible, or tied to performance standards rather than mere task completion. Where interest is absent, rewards can bootstrap engagement, but designers should plan for fading them and building autonomy-supportive structures before withdrawal.

Evidence status. Five entries are recorded: two competing meta-analyses (Deci, Koestner & Ryan 1999, 128 experiments; Cameron & Pierce 1994, 96 experiments, read from its abstract), the original laboratory experiment (Deci 1971), and two reviews from either side of the debate (Eisenberger & Cameron 1996; Deci, Koestner & Ryan 2001). Still missing: studies in authentic classroom settings, and full-text effect sizes for Cameron & Pierce (1994).

Merged from “Extrinsic Rewards Undermine Intrinsic Motivation” (extrinsic-rewards-undermine-intrinsic-motivation): Scope and mechanism. The claim is usually explained by the overjustification account: when an external reward supplies a sufficient justification for performing an already-interesting task, learners reattribute their engagement to the reward rather than to their own interest, and voluntary engagement drops once the reward is removed. This sits in tension with the reinforcement tradition of behaviorism, which predicts rewards should increase the rewarded behavior — the contradiction is sharpest for measures of free-choice persistence and self-reported interest rather than for task performance while rewards are still available.

Likely moderators. The effect is generally understood to depend on how the reward is framed and delivered: tangible versus verbal rewards, rewards promised in advance versus given unexpectedly, and rewards contingent merely on task engagement versus on performance quality. Rewards interpreted as informational — signaling competence or recognition — are less likely to undermine interest than rewards experienced as controlling [~M]. The two meta-analyses recorded above document several of these moderators, and they disagree about how much they matter: Deci et al. (1999) find undermining for engagement-, completion- and performance-contingent tangible rewards, while Cameron and Pierce (1994) find no overall effect and a minimal one only for expected tangible rewards given for doing a task. Both find that verbal praise does not undermine intrinsic motivation.

Boundary conditions. For tasks learners find uninteresting to begin with, there is little intrinsic motivation to undermine, and rewards may operate as straightforward incentives [~M]. This connects to the positive counterpart of the claim: supporting learners' needs for autonomy, competence, and relatedness sustains interest — see Autonomy supports intrinsic motivation. and Self-determination theory.

Design implications. On the evidence recorded above, the practical guidance is conditional: avoid promising tangible rewards for activities learners already find interesting; if rewards are used, prefer unexpected, performance- or progress-contingent, informational forms (feedback that signals competence) over expected, engagement-contingent, controlling ones [~M]. Gamification systems built on points and badges should be evaluated for whether they displace the interest they are meant to support — see Autonomy. Where rewards are unavoidable (grading systems, institutional incentives), framing them as feedback on competence rather than as payment for compliance is the safer design choice [~M].

Open questions. Studies in authentic educational settings, rather than laboratory tasks, are still not recorded here.

Merged from “Rewarding an already-intrinsically-motivating activity can reduce future engagement with it” (overjustification-effect-reduces-intrinsic-motivation): The 1990s-2000s literature split into a genuine, still-cited disagreement: Cameron and Pierce (and Eisenberger & Cameron) argue the undermining effect is smaller and more conditional than early studies like Deci (1971) suggested, while Deci, Koestner, and Ryan push back that even a bounded, condition-dependent effect is large and common enough in classrooms to warrant real caution. Both camps agree on the two clearest moderators: rewards delivered as a flat rate for participating (rather than piecemeal for output) and rewards attached to tasks with a well-defined, consistently-expected performance standard are least likely to backfire. This directly informs the constraint most learning-design guidance now gives on Reinforcement Theory and Behaviorism: extrinsic reinforcement is a legitimate design lever, but layering it onto a task a learner is already intrinsically engaged with is exactly the case where it is most likely to reduce, not increase, future engagement.

Learner Variables¶

• Motivation — outcome: instruction changes it
• Interest — outcome: instruction changes it

Related Claims¶

• Autonomy supports intrinsic motivation. — the complementary positive side of the same motivational mechanism
• Self-determination theory — the theoretical framework explaining why controlling rewards undermine interest
• Behaviorism — the contrasting tradition in which reinforcement strengthens behavior; highlights the tension this claim creates
• Autonomy — the learner need most directly threatened by controlling reward structures
• A non-peer-reviewed web essay asserts, citing an earlier source second-hand, that behaviorist reinforcement methods are very effective in creating positive behavior in almost any learning environment — reports the opposite
• Useful cognitive associations acquire secondary-reward value, making information storage an operant event — related
• The article argues extrinsic rewards such as grades can reduce intrinsic motivation and should be phased out — a broader claim this one bears on
• Rewarding only 'good' ideas in a final task produced no incentive effect, but this condition was confounded and excluded from the main analysis — related
• Praise for intelligence, rather than for effort, after success undermines children's motivation and performance after a later failure — related
• Learners invest more effort in tasks they perceive as interesting, personally important, or useful to their future goals. — reports the opposite
• A 1970 teacher self-analysis programme asserts, without data, that teacher responses including verbal rewards can close off pupil thinking — related
• Students who generate their own research question are apt to be more invested and more engaged — related
• Early ability to delay gratification predicts later life outcomes — both concern how external structure interacts with a learner's own self-regulation of motivation
• Traditional salary schedules may not reward effective teaching or give the most effective teachers incentives to work in high-need schools, motivating performance-based compensation — related
• In Pilot District 3, majorities of students agreed that Compact Writing Assignments gave them topic choice and that autonomy and varied expression formats improved engagement and writing quality — related
• Autonomy, meaningful purpose, and interest improve intrinsic motivation and deeper processing — related
• AI tools can enhance short-term task performance while undermining durable learning (the learning-performance paradox) — a narrower finding that bears on this claim
• An analytical chapter identifies reward hacking, engagement optimization over learning, and scaffolding dependency as risks of RL policies with poorly specified rewards — related
• Extrinsic goals are more strongly associated with cheating than intrinsic enjoyment of learning — related
• Student choice is rare in authentic assessment designs while assigned, sometimes graded, self-reflection is more common — related
• Well-designed games can promote persistence in practice and high-quality engagement, and when basic needs of autonomy, competence, and relatedness are met, learning becomes its own reward, reducing the temptation to outsource cognitive effort to LLMs. — related
• Experienced effort-by-need frustration is neither a necessary requirement nor a typical concomitant of high-quality learning — related
• Extrinsic rewards deplete intrinsic motivation for already-motivated activities once the rewards are removed — possibly the same claim (merge candidate)
• Being good role models, taking interest in students' lives outside class, and allowing students choices were the behaviours most consistently associated with positive change in engagement — related
• Behavior modification programs relying on rewards rarely produce lasting changes in attitudes or behavior, and children whose parents frequently use rewards tend to be less generous — related
• Reach — which lever settings remain in force when the student is alone with AI — is the framework's organizing question — related
• Positive reinforcement and praise correlate positively with learning while criticism correlates negatively — related
• Grading practices can diminish interest, encourage easier paths, and reduce quality of thinking in maker education — related
• Extrinsic incentives have mixed results on effort and achievement, working better when immediate and larger — related
• Rewards cause people to lose interest in whatever they were rewarded for doing, undermining intrinsic motivation — a narrower finding that bears on this claim
• Autonomy-supportive teaching contrasts with controlling teaching, and controlled students do poorly compared to autonomous students — related
• People expecting a reward for a task perform worse than those expecting nothing, especially on tasks requiring cognitive sophistication and open-ended thinking — related
• Financial incentives do not decrease intrinsic motivation, though point estimates are too imprecise to detect any effect — reports the opposite
• Most African students enrolled in the SuaCode smartphone coding course for intrinsic rather than extrinsic reasons (68.7% vs 31.3% of coded motivations) — a narrower finding that bears on this claim
• Teachers provided targeted extra support, immediate feedback, and reinforcers to learners with DD — related
• Receiving grades is positively associated with achievement and external motivation compared with receiving no performance feedback (review-reported meta-analysis) — related
• Reward techniques show positive meta-analytic effects for gamification and personalization, but perceptual fluency did not directly affect learning outcomes — related
• External rewards can undermine intrinsic motivation for learners who are already motivated — a broader claim this one bears on
• Most of the 45 reviewed studies report positive gamification effects on motivation and engagement in mathematics; only four report negative effects — related
• Feedback targeting reading self-efficacy had a positive effect on motivation, while error-focused feedback and rewards had mixed effects on long-term interest and engagement — related