Why Motivation Matters: What Moves People, and What the Research Says About Rewards and Habits

Start hereIn a 1973 experiment, children who expected a reward for an activity they already showed interest in later showed less interest in it than children who got a surprise reward or none. Yet a 1994 review of 96 experiments concluded that, overall, reward does not decrease intrinsic motivation. Which is right?

Motivation is what moves you to act, and the word motive comes from the Latin for moving. It matters because it gets things started. Our reading of the research on rewards and self-control: conditions and habits do much of the work of keeping it going. This page starts with a famous experiment, then the words, the research and the arguments. It ends with five things to try.

011. A simple example: the prize that backfired

In a 1973 field experiment, children who already showed interest in an activity were put into one of three groups.

• One group agreed to do the activity in order to get a reward.
• One got a reward they knew nothing about until they had finished.
• One neither expected nor received a reward.

As predicted, the children who expected the reward later showed less interest in the activity than either of the other groups. Our reading: the trouble was the reward they were working for, not the reward itself.

More than four decades of research and nine meta-analyses have focused on this undermining effect. Section 4 sets out the argument.

022. What the words mean

Motive comes, through Anglo-French and Medieval Latin, from Latin motus, the past participle of movēre, “to move”. Motivation is a later word: Merriam-Webster’s first known use is 1873.

Discipline comes from Latin disciplīna, “teaching, instruction”, from discipulus, “pupil, learner”. The English word arrived already carrying “chastisement” and “system of ordered conduct”. The Latin word could also mean “a custom, habit”.

Our reading: a motive is what moves you, and discipline began as teaching and learning.

033. What motivation is, in the research

In 2000 Richard Ryan and Edward Deci summed up research on the conditions that help or hinder self-motivation. They wrote that the findings had led to three innate psychological needs: competence, autonomy and relatedness.

When the needs are satisfied, they wrote, the result is more self-motivation and better mental health; when they are thwarted, less motivation and well-being.

In plain words: did I choose this, can I do it, and am I connected to anyone in it? Our page on motivation vs discipline explains each need and turns it into steps.

044. Do rewards kill motivation?

Reviews of this research have reached dramatically opposed bottom-line conclusions, as one 1999 comment noted. Here are both sides.

The case that rewards undermine

A 1999 meta-analysis by Edward Deci, Richard Koestner and Richard Ryan covered 128 studies.

• It found that rewards for taking part in a task, finishing it or doing it well significantly undermined what the authors call free-choice intrinsic motivation.
• Tangible rewards tended to do more harm to children than to college students.
• Positive feedback, by contrast, increased both free-choice behaviour and self-reported interest.

The case that the harm is narrow

A 1994 meta-analysis by Judy Cameron and David Pierce covered 96 experiments and concluded that, overall, reward does not decrease intrinsic motivation.

• It found that verbal praise increased intrinsic motivation.
• The only negative effect appeared when expected tangible rewards were given simply for doing a task, and it was minimal.

In 1996 Robert Eisenberger and Cameron argued that harmful effects of reward occur under tightly limited conditions that are easy to avoid.

The exchange

In 1999 the journal that printed the Deci review also printed comments from both sides, and the authors’ reply. The reply was titled “The undermining effect is a reality after all”.

One comment, by Mark Lepper and two colleagues, concluded that the Deci review had captured the literature more accurately than Cameron and Pierce’s. Both comments came from people already in the argument: Lepper was an author of the 1973 experiment, and Pierce and Cameron wrote the 1994 review. The other comment came from Eisenberger, Pierce and Cameron.

In 2001 both sides restated their case. Cameron and two colleagues concluded from a new meta-analysis that rewards do not have widespread negative effects on intrinsic motivation. Deci, Koestner and Ryan wrote that tangible rewards do have a substantial undermining effect.

A later look

A 2014 meta-analysis covering 40 years of research and 212,468 people found intrinsic motivation a medium to strong predictor of performance.

• Its importance to performance remained whether or not incentives were offered.
• But when incentives were tied directly to performance, intrinsic motivation mattered less to performance, a pattern the authors call “crowding out”.
• Intrinsic motivation better predicted the quality of performance, and incentives better predicted the quantity.
• The authors concluded that, for performance, incentives and intrinsic motivation are not necessarily at odds.

Our reading: these are links between measures, not proof of cause.

Partly Rewards kill motivation. Both camps found a negative effect for some expected tangible rewards, and both found that praise or positive feedback raised motivation on average. The 1999 review also found that praise given in a controlling way, such as “just as you should”, may lower it. How far the harm reaches is still argued.

055. Willpower, habits and self-control

Still argued Willpower is a resource that runs out when you use it.

• A 23-lab registered replication found an effect whose confidence interval “encompassed zero”.
• A 36-site study reported that its “preregistered analyses did not find evidence for a depletion effect”, though an exploratory analysis of its full sample found a small significant effect.
• A preregistered 12-lab study found “a small and significant ego depletion effect”.

Our reading is the same as on our page about famous findings and whether they held up. The popular idea that willpower drains like a fuel tank is not supported; whether a tiny effect exists is still open.

So what do people with good self-control rely on? Across six studies with 2,274 people, Brian Galla and Angela Duckworth found that good habits helped account for the link between self-control and better outcomes. The outcomes included homework completion, grade point average and persisting through the first year of college. They wrote that the results suggest beneficial habits, perhaps more so than effortful inhibition, are an important link between self-control and good outcomes.

A 2012 meta-analysis of 102 studies looked closer at one of its three self-control scales. On that scale, trait self-control was more strongly linked to automatic behaviour than to controlled behaviour. It was also more strongly linked to imagined behaviour than to actual behaviour.

Our reading: people who look disciplined may be leaning on habits more than on resisting. Our summaries of The Power of Habit and Atomic Habits look at how habits form.

066. When it is not a motivation problem

Low drive is not always about motivation. The NHS lists depression among the common causes of tiredness, along with not getting enough sleep and stress. It says to see a GP if you have felt tired for a few weeks and are not sure why, or if tiredness affects your daily life.

If that sounds like you, that NHS advice to see a GP is the place to start, not a better goal or a reward chart. This page is not medical advice.

077. How to use it

• When you stall, check the three needs. Did I choose this? Can I see progress? Is anyone with me?
• Be careful about paying yourself for what you already enjoy. Keep rewards for jobs that need a push; that is our reading of the 1973 experiment. Praise and positive feedback raised motivation on average in both camps’ reviews, except praise given in a controlling way.
• Build the habit rather than relying on resisting. Same time, same place. Our self-improvement guide covers if-then plans and habits.
• Don’t read one tired afternoon as proof your willpower is used up. The large replications do not support the fuel-tank idea.
• Use a tally for how much and your reason for how well. Our reading of the 2014 review, where incentives went with quantity and interest with quality.

08Quick answers

09Use this

10Quick quiz

How we checked this · Last checked October 2026

• We read the Merriam-Webster entries in a browser, because the site blocks scripts, and the Latin dictionary entry on Perseus.
• For the research we read abstracts - the 1973 experiment, the 1994, 1999 and 2014 meta-analyses, the 1996 article, the ego-depletion replications and the habit and self-control studies.
• We did not read the full 1999 comments or the reply. We give only what their abstracts and titles state, with authors checked on Crossref.
• Both 1999 comments came from authors already in the argument, and the page says so.
• For the finding on praise given in a controlling way we read the full text of the 1999 review.
• Our ego-depletion wording matches our page on famous findings. In October 2026 we searched for a meta-analysis on rewards newer than 2014 and found none.

How we check every article →

11Sources