The assumption that rewarding behaviour increases it is one of the most durable in management. Applied to creative and innovative work, it produces a specific and documented harm. The research across five decades of experimental and neuroimaging studies establishes that contingent, performance-based rewards do not enhance creative output — they reliably reduce the quality and quantity of innovative work, through mechanisms that are now understood at the neurological level.

The foundational experiment

Lepper, Greene and Nisbett’s 1973 study divided preschool children who already showed high intrinsic interest in drawing with felt-tip markers into three conditions: expected reward (told they would receive an award for drawing), unexpected reward (received the same award without prior expectation), and no reward. Two weeks later, when no rewards were available, the researchers observed spontaneous drawing during free-play time. Children in the expected-reward condition showed significantly less spontaneous drawing than those in the other two conditions. An activity they had previously engaged in spontaneously had been measurably undermined by having been rewarded for it.

The mechanism is self-perception: people infer their own motivation by observing their behaviour and the conditions under which it occurs. When an expected reward is present, a person observes themselves doing the activity in a reward context and infers “I am doing this for the reward.” The intrinsic attribution is displaced by the extrinsic one. When the reward is removed, the extrinsic motivation disappears — but the intrinsic motivation has already been restructured away.

128 studies, precise effect sizes

Deci, Koestner and Ryan’s 1999 meta-analysis synthesised 128 experimental studies on the effects of extrinsic rewards on intrinsic motivation. The results confirmed and quantified the overjustification effect across reward types. Engagement-contingent rewards undermined free-choice intrinsic motivation with an effect size of d = -0.40. Completion-contingent rewards produced d = -0.36. Performance-contingent rewards — the most common in organisational settings — produced d = -0.28. All tangible expected rewards undermined intrinsic motivation.

The single exception — and the most practically important finding in the meta-analysis — was positive verbal feedback. This enhanced free-choice intrinsic motivation with an effect size of d = +0.33 and self-reported interest with d = +0.31. The critical moderator was whether the reward felt informational or controlling. When a reward signals felt competence — telling the person they performed well in a way that confirms their capability — it supports intrinsic motivation. When it feels controlling — specifying what the person must do to receive it — it undermines intrinsic motivation. This distinction is the psychological mechanism that determines whether a reward helps or harms.

The neurological mechanism

Murayama and colleagues’ 2010 fMRI study with 28 participants provided the first direct neuroimaging evidence for the neural basis of the undermining effect. One group received performance-contingent monetary rewards for an interesting stopwatch task across one session; a control group received only participation compensation. In a later session, neither group received performance-based rewards. The previously rewarded group showed fewer voluntary engagements with the task during a free-choice period — a direct behavioural replication in adults. The fMRI data explained why: the previously rewarded group showed significantly reduced activation in the anterior striatum, the midbrain, and the lateral prefrontal cortex.

These three regions are the core of the dopaminergic reward and valuation network. The anterior striatum represents subjective task value; the midbrain contains the dopamine neurons that signal reward prediction; the lateral prefrontal cortex integrates value signals with goal-directed cognition. When performance-based extrinsic reward is present, the system encodes the task’s value as externally sourced. When the reward is removed, the internal task value signal is diminished. The innovator who has been trained to work for bonuses and performance reviews has a brain that has encoded their creative work as extrinsically valuable — and whose intrinsic dopamine response to that work has been correspondingly reduced.

What intrinsic motivation produces that extrinsic reward cannot

Amabile’s research programme found consistently that work produced under conditions of intrinsic motivation was rated more creative by independent expert judges than work produced under extrinsic motivation or surveillance conditions. In one study, writers asked to think about extrinsic reasons for writing — publication, impressing readers, making money — before completing a creative writing task produced work rated significantly less creative than those who had thought about intrinsic reasons. The judges were blind to condition. The quality difference was not in technical competence but in originality and creative engagement.

Gruber, Gelman and Ranganath’s 2014 Neuron study established that curiosity-driven learning produces greater hippocampal activity, higher dopamine release in the mesolimbic system, and significantly better long-term memory retention than non-curious learning of equivalent material. Intrinsic curiosity generates a genuine dopamine signal arising from the anticipation of resolution — the same system that extrinsic rewards activate, but triggered by internal wanting rather than external promise. When an extrinsic reward is introduced into an activity the person was already curious about, the Murayama mechanism predicts the reattribution: the valuation system shifts the dopamine signal to the external reward, reducing the extent to which curiosity itself generates the reward response.

The practical implication

Innovation bonus schemes and creativity competition programmes that link payment to specific creative outputs are functionally controlling — they define what counts as valid creative work and make payment contingent on meeting that external standard. This signals to employees that their creative work is being externally managed, reduces autonomy over the creative process, and undermines the intrinsic engagement that produces the most genuinely novel work. Internal innovation competitions reliably produce high volumes of submissions. Their outputs tend toward the incrementally rather than radically innovative — safe, legible, evaluator-palatable ideas rather than the genuinely novel recombinations that disruptive innovation requires. The most innovative ideas are often the ones that cannot yet be articulated clearly enough to win a competition. They exist in the early, exploratory, uncertain stage of development that intrinsic motivation sustains and extrinsic evaluation truncates.

The alternative the research supports is not the removal of all reward but the shift from controlling to informational: feedback that confirms creative competence without specifying what that competence must produce. Atlassian’s ShipIt days — 24 hours of autonomous creative time with no performance requirements, followed by voluntary presentation — exemplify this structure. The end-of-day presentation provided informational feedback on competence in a non-controlling context. The outputs consistently included shipped product features. The mechanism was not the removal of reward but the removal of contingency.

Book worth reading on this

Punished by Rewards by Alfie Kohn is the most direct applied treatment of the reward-undermining research for an organisational audience. Kohn draws extensively on the Deci, Lepper, and Amabile research programmes and applies them across education, parenting, and workplace contexts with a directness that makes the implications impossible to avoid. He is more polemical than the underlying research strictly requires, but the core argument is well-supported and the organisational examples are immediately recognisable.

This article is for educational and informational purposes only. Sources: Lepper, M.R., Greene, D. & Nisbett, R.E. (1973), Undermining Children’s Intrinsic Interest with Extrinsic Reward, Journal of Personality and Social Psychology, 28(1), 129–137. Deci, E.L. (1971), Effects of Externally Mediated Rewards on Intrinsic Motivation, Journal of Personality and Social Psychology, 18(1), 105–115. 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. Murayama, K., Matsumoto, M., Izuma, K. & Matsumoto, K. (2010), Neural Basis of the Undermining Effect of Monetary Reward on Intrinsic Motivation, PNAS, 107(49), 20911–20916. Amabile, T.M. (1985), Motivation and Creativity, Journal of Personality and Social Psychology, 48(2), 393–399. Gruber, M.J., Gelman, B.D. & Ranganath, C. (2014), States of Curiosity Modulate Hippocampus-Dependent Learning via the Dopaminergic Circuit, Neuron, 84(2), 486–496. Deci, E.L. & Ryan, R.M. (1980), The Empirical Exploration of Intrinsic Motivational Processes, Advances in Experimental Social Psychology, 13. Deci, E. (1995), Why We Do What We Do, Penguin. Kohn, A. (1993), Punished by Rewards, Houghton Mifflin.