The Overjustification Effect: Why Rewards Can Quietly Kill the Habits You Love
Adding external rewards to habits you already enjoy can destroy intrinsic motivation. Deci and Lepper's research reveals which habits to protect — and which rewards to avoid.
Rewarding yourself for good habits seems like a no-brainer. You exercise, you get a treat. You meditate, you get a sticker on the calendar. More incentives, more behavior — simple economics.
Except psychology has known for fifty years that this can backfire in a specific and damaging way. When you add external rewards to behaviors you already find intrinsically meaningful, you can inadvertently teach your brain that the reward — not the activity — is the point. Remove the reward, and the motivation collapses.
This is the overjustification effect. Understanding it is essential for anyone building habits they intend to keep for life.
The Landmark Study
In 1973, researchers Mark Lepper, David Greene, and Richard Nisbett at Stanford published a study in Journal of Personality and Social Psychology that changed how psychologists think about motivation.
They studied children who already loved drawing — children who chose art during free time with no prompting. The researchers divided them into three groups: one group received an expected reward (a certificate they knew they’d get for drawing), one received an unexpected reward (a surprise certificate after drawing), and one received no reward.
Two weeks later, all the certificates were gone. The researchers measured how much each group chose to draw during free time.
The results were stark: children who had received the expected reward drew significantly less than before the study — less than the no-reward group, and less than children who received an unexpected reward. They had learned, via the reward, that drawing was something you did to get a certificate. When the certificate disappeared, so did the reason.
The unexpected-reward group showed no drop. The key variable wasn’t the reward itself — it was whether the reward was anticipated, which caused children to explain their behavior to themselves using the reward as justification.
This effect has been replicated across hundreds of studies. A 1999 meta-analysis by Deci, Koestner, and Ryan in Psychological Bulletin, reviewing 128 experiments on the effects of extrinsic rewards on intrinsic motivation, confirmed that tangible, expected rewards reliably undermine intrinsic motivation for interesting activities.
(See: Deci, Koestner & Ryan 1999 meta-analysis, Psychological Bulletin)
Why the Brain Makes This Mistake
The mechanism is cognitive attribution. When you ask yourself “why am I doing this?”, your brain scans for a plausible explanation. If an obvious external reward is present — money, a prize, recognition — your brain gravitates to it. The internal reason (curiosity, joy, values) gets crowded out.
Edward Deci, whose self-determination theory provides the theoretical foundation for this research, describes it as a shift from autonomous regulation to controlled regulation. Behavior that felt self-directed starts to feel compelled by outside forces. And behavior that feels compelled, without the force, stops.
This is particularly insidious for habits you’ve carefully chosen — the morning reading practice, the journaling, the Saturday run. These are behaviors where intrinsic motivation is available, if you don’t accidentally overwrite it with the wrong incentive structure.
Which Rewards Are Safe?
Not all rewards damage intrinsic motivation. The research points to a useful distinction.
Rewards that inform vs. rewards that control. Deci’s framework distinguishes between rewards that carry informational value (“you completed this challenging thing — evidence of capability”) and rewards that feel like control (“you did this to get that”). Verbal praise, for example, often has an informational quality: “That was a strong run” signals competence. A cash bonus for exercise tends to feel controlling — and more reliably undermines motivation.
Unexpected rewards. As Lepper’s original study showed, surprise rewards don’t trigger the same attributional process because they weren’t anticipated. The behavior couldn’t have been done for the reward. An occasional “I’m proud of you for sticking with this” from a friend, landing unexpectedly, reinforces the behavior without framing it as a transaction.
Completion-contingent vs. engagement-contingent. Rewards tied to showing up (“I get to listen to this podcast only while walking”) tend to preserve intrinsic motivation better than rewards tied to outcomes (“I earn $5 every time I exercise”). The former supports the behavior; the latter prices it.
The Habit-Specific Question
The critical question for each habit is: Do I already find this intrinsically interesting or meaningful?
If the answer is yes — you genuinely enjoy the activity, or it connects to a value you hold — proceed with extreme caution around expected, tangible rewards. Protect the intrinsic motivation you already have. The goal is to let the activity be its own reward, and to design context that makes the activity easier to access (time, cue, environment) without monetizing or gamifying the behavior itself.
If the answer is no — you find the habit neutral or mildly aversive but strategically valuable — external rewards are much safer. The activity isn’t intrinsically interesting, so there’s no underlying motivation to undermine. Rewards can bootstrap repetition until the habit becomes automatic and rewarding in its own right through competence growth and identity reinforcement.
Flossing doesn’t carry much intrinsic joy for most people. Rewarding it with a small ritual afterward poses little risk. Rewarding yourself financially for a journaling practice you love is a different matter.
Designing Habit Rewards in Becoming
In Becoming, the habit completion signal — the check, the streak counter, the accumulated evidence — is designed to feel informational rather than controlling. You’re not earning points. You’re observing your own consistency.
This distinction matters because the kind of self-monitoring that builds identity is different from the kind that creates a transaction. “I completed 22 of the last 30 days” is a competence signal. “I earn a reward at 30 days” is a contingency. The former feeds self-determination theory’s need for competence — a core component of intrinsic motivation. The latter can corrupt it.
When you design your habit’s cue and context in Becoming, think of environmental design and identity framing as the real rewards. Removing friction makes the behavior easier to access. Naming the identity statement — “I am someone who prioritizes their mental health” — makes the behavior feel self-expressive rather than externally demanded.
Frequently Asked Questions
Can I ever use rewards to build habits?
Yes — but with care. Reserve explicit rewards for habits with low intrinsic value. For habits you care about for their own sake, use informational feedback (tracking progress, noting capability growth) rather than transactional incentives. And always ask: if this reward disappeared tomorrow, would I still do the habit?
What about streaks? Aren’t those just rewards?
Streaks are a useful form of informational feedback when they reflect genuine consistency — they show you what you’re capable of. The risk is when breaking a streak becomes catastrophic, which can flip motivation from intrinsic to loss-averse. Treat streaks as evidence, not as the point.
I’ve been rewarding my journaling with coffee for a year. Is the damage done?
Not necessarily. Attribution is a cognitive process, and it can be updated. Gradually shift the framing: notice what you get from the journaling itself — clarity, emotional release, a record of thinking. Let that become the salient motivation. The reward can fade naturally rather than being abruptly removed.
The Bottom Line
The habits most worth protecting are the ones you already love. Build systems that support them, reduce the friction around them, connect them to your identity — but think carefully before you put a price on them.
Intrinsic motivation is delicate and extraordinarily powerful. Guard it with the same intentionality you’d give to any other resource that’s difficult to regenerate once spent.