What Rewards Really Tell Us About Motivation
Can a reward make someone less interested in the work? Decades of contested research suggest that what a reward communicates may matter as much as what it offers.
A reward can recognise effort, encourage a task to be completed, or acknowledge a difficult achievement. Yet its meaning is not exhausted by its value. The person receiving it may also learn something about the expectations surrounding the work, the degree of control they are believed to have and whether their effort is being noticed. These are important distinctions for anyone who leads a team, teaches a class or supports a young athlete.
The evidence does not support a simple verdict for or against rewards. The effects depend on what is rewarded, how the reward is offered and which outcome counts as motivation. That is why the argument matters beyond a familiar debate about bonuses and prizes: a reward is both an incentive and a form of communication, and people may respond to both.
What the experiments found
The best-known evidence is a large meta-analysis of laboratory experiments by Edward Deci, Richard Koestner and Richard Ryan. Expected tangible rewards reduced how much people later chose, of their own accord, to keep doing an interesting task once the reward was removed. This happened whether the reward was tied to engaging in the task, completing it or performing well, with the effect strongest for rewards tied simply to engagement and weakest for rewards tied to performance. Unexpected rewards, and rewards not tied to the task, did not show the effect. The effect appeared mainly in what people later chose to do; their self-reported interest changed little.
Positive verbal feedback pointed the other way, increasing later free-choice engagement among college students. Among children the effect was small and not statistically significant, and tangible rewards were more harmful for them than for college students.
Not everyone read the evidence the same way. Judy Cameron and David Pierce, analysing a partly different set of studies in earlier work, concluded that the detrimental effects of reward were narrow and largely avoidable, confined in their analysis to expected tangible rewards given simply for doing a task. The two groups disagreed for years over which studies to include, how to treat unusual results and which experimental conditions best represented everyday life. The honest summary is that the undermining effect is real but conditional, with its practical size outside the laboratory still uncertain.
Control or information
The explanation most often offered comes from cognitive evaluation theory, part of the self-determination theory developed by Deci and Ryan. A reward experienced as controlling, a signal that someone else is directing behaviour and will judge compliance, is predicted to weaken intrinsic motivation. A reward experienced as informational, telling someone they have done well, is predicted to support it. The direct test of that distinction within the meta-analysis rests on a fairly small set of studies, though they point in the predicted direction, and both the meta-analysis and the theory come from the theory's founders. Marylène Gagné and Deci later set out how the broader theory applies to work, proposing that people sustain higher-quality motivation when their environment supports their autonomy, competence and connection with others.
This is the core of the article's argument, and it offers an account of why the same bonus can land so differently. Two people receiving identical sums may draw opposite conclusions: one that their judgement is trusted and their work noticed, the other that they are being steered and watched.
Quantity, quality and what the work needs
A later meta-analysis by Christopher Cerasoli, Jessica Nicklin and Michael Ford adds a practical refinement. Across four decades of research, intrinsic motivation and external incentives both predicted performance, but differently. Intrinsic motivation was the stronger predictor of the quality of performance, incentives the stronger predictor of its quantity, and intrinsic motivation mattered less for performance when incentives were directly tied to performance than when they were only indirectly tied. These are associations, mostly from correlational data, and the quality–quantity distinction describes how performance was measured rather than types of task.
One inference, which none of these studies tests directly, is that the message of a reward matters most where quality matters most. For simple, measurable, high-volume work, well-structured incentives may suit the task. For work that depends on judgement, care or persistence through difficulty, incentives tied tightly to a narrow measure may make intrinsic motivation a weaker guide to quality.
Reading the message before sending it
For leaders and coaches, the practical questions follow from the argument rather than from any tested formula. Is this reward likely to be received as recognition or as surveillance? Is it tied to the quality we actually need, or to a proxy that is easier to count? Does the feedback that accompanies it inform people about their progress, or simply grade them?
Parents face a version of the same question, although the evidence for children is limited. Paying a child for every page read could shift attention from the reading to the payment; noticing and talking about what the child read sends a different message about why reading is worth doing. The laboratory findings for children concern short tasks, not years of family life, so this remains an inference rather than a result.
A reward, in the end, is a sentence addressed to the person who receives it. The research does not tell anyone exactly what that sentence should say. It does suggest that people hear it, and that what they hear can matter alongside what they are given.
ALTIUS Note
The undermining effect remains debated in its scope, much of the evidence comes from short laboratory experiments, and results differ for children. Findings on incentives at work are largely correlational. The control-versus-information framing comes from cognitive evaluation theory; the practical questions offered here are ALTIUS inferences rather than tested interventions.
Research References
- Cameron, J., & Pierce, W. D. (1994). Reinforcement, reward, and intrinsic motivation: A meta-analysis. Review of Educational Research, 64(3), 363–423. https://doi.org/10.3102/00346543064003363
- Cerasoli, C. P., Nicklin, J. M., & Ford, M. T. (2014). Intrinsic motivation and extrinsic incentives jointly predict performance: A 40-year meta-analysis. Psychological Bulletin, 140(4), 980–1008. https://doi.org/10.1037/a0035661
- 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. https://doi.org/10.1037/0033-2909.125.6.627
- Gagné, M., & Deci, E. L. (2005). Self-determination theory and work motivation. Journal of Organizational Behavior, 26(4), 331–362. https://doi.org/10.1002/job.322
Related from ALTIUS
Keep reading ALTIUS
Subscribe for new essays and analysis, or sign in if you already have an account.