How to Fade Out Rewards Without Losing the Habit

By Noah Sato

A new habit rarely survives on willpower alone. Most people who successfully start flossing, journaling, or taking a daily walk lean on some kind of reward in the first few weeks: a checkmark on a calendar, a points app, a treat after a workout, or a small cash prize for hitting a streak. Rewards lower the activation energy of a new behavior. They give the brain a reason to show up on the days when the habit itself doesn’t yet feel good.

The trouble starts later. Research on motivation has repeatedly found that when an external reward is removed after being closely tied to a task, interest in that task can drop below where it started, even for activities people once did for fun. This is the risk built into every sticker chart, streak bonus, and app badge: the reward that got you moving can quietly become the only reason you keep moving, and if it disappears, the habit disappears with it.

This guide lays out a step-by-step protocol for fading rewards deliberately, in an order and on a timeline that protects the habit instead of collapsing it. It draws on research from self-determination theory, habit-formation studies, and behavioral economics to explain why gradual withdrawal works better than either an abrupt cutoff or an indefinite reward schedule.

The Role of Extrinsic Rewards in Early Habit Formation

Extrinsic rewards are effective at the start of a new behavior for a simple reason: a brand-new habit has no track record of its own payoff. Someone who has never meditated does not yet know what a calmer afternoon feels like. Someone who has never kept a food log does not yet feel the quiet satisfaction of an accurate week. Until that internal payoff becomes noticeable, an external one, be it a checkmark, a point total, or a small treat, fills the gap and keeps the behavior alive long enough for its natural benefits to appear.

Psychologist Edward Deci demonstrated this dynamic in a 1971 laboratory study. Participants who found a puzzle task inherently interesting were split into two groups: one was paid for solving puzzles during a middle session, the other was not. During the paid session, engagement with the puzzles rose. But once the payment stopped, the previously paid group spent noticeably less free time on the puzzles than the group that had never been paid at all. The reward had temporarily boosted behavior, then left a smaller appetite for the activity than existed before the money was introduced.

That finding was extended two years later by Mark Lepper, David Greene, and Richard Nisbett, who studied preschoolers who already enjoyed drawing with felt-tip markers. Children who were told in advance that they would receive a certificate for drawing later spent less time drawing on their own, compared with children who received no reward or an unexpected one. The expectation of a reward, tied directly to the activity, was enough to change how the children related to something they had previously done purely for enjoyment.

These two studies became the foundation for what researchers now call the overjustification effect: when a reward is added on top of an already-engaging task, people can start attributing their behavior to the reward rather than to their own interest. Once the external justification disappears, so does a chunk of the motivation to continue, because the internal reason was partially crowded out rather than reinforced.

None of this means rewards are the wrong tool for habit formation. It means the size, timing, and framing of a reward determine whether it supports a habit or eventually competes with it.

What the Research Says About the Shift Toward Intrinsic Motivation

Edward Deci and Richard Ryan’s self-determination theory offers the clearest framework for understanding why some habits survive after the rewards stop and others don’t. The theory proposes that people are intrinsically motivated to the extent that a behavior satisfies three basic psychological needs: autonomy (feeling the behavior is chosen, not imposed), competence (feeling capable and improving over time), and relatedness (feeling connected to other people through the behavior). A reward that supports these needs tends to strengthen motivation; a reward that undercuts them tends to weaken it.

This distinction shows up clearly in a 1999 meta-analysis by Deci, Richard Koestner, and Ryan, which pooled data from 128 experiments on rewards and intrinsic motivation. The analysis found that tangible rewards tied directly to task completion or performance reliably reduced intrinsic motivation for interesting tasks, while unexpected rewards or verbal praise that conveyed competence had a much smaller negative effect, and sometimes a positive one. The key variable was not whether a reward existed, but whether it was perceived as controlling (“do this to get the prize”) or informational (“you’re getting better at this”).

A 2001 meta-analysis by Judy Cameron pushed back on parts of this conclusion, arguing that rewards for outperforming a specific standard, or rewards for tasks people found boring to begin with, could actually increase engagement rather than reduce it. The disagreement between these two lines of research points to a practical takeaway rather than a contradiction: reward effects depend heavily on context, and a habit that starts out low-interest (like flossing) responds differently to reward removal than a habit that starts out high-interest (like a hobby someone already loves).

Habit-formation research adds a timing dimension to this picture. In a widely cited 2010 study published in the European Journal of Social Psychology, Phillippa Lally and colleagues tracked 96 participants who chose a new eating, drinking, or activity behavior to repeat daily for 12 weeks, measuring how automatic the behavior felt each day. Automaticity increased following an asymptotic curve, and the average time for a behavior to reach its plateau of automaticity was 66 days, though individual habits in the study ranged from 18 to 254 days depending on the person and the behavior. That range matters for reward fading: a habit that hasn’t yet reached its automaticity plateau is a poor candidate for removing support, because the internal, self-sustaining routine the reward was supposed to help build may not exist yet.

A 2018 study in BMC Psychology by Benjamin Gardner and colleagues followed 118 participants adopting flossing or a daily vitamin C habit over 16 weeks. It found that pleasure and intrinsic motivation predicted stronger habit growth per repetition, independent of how often the behavior was simply repeated, and that performing the behavior in a stable context (same time, same place, same trigger) also helped convert repetition into automatic habit. In other words, the emotional and situational quality of each repetition mattered as much as the count of repetitions, which is a strong argument for building enjoyment and consistent cues into a habit before assuming volume of repetition alone will make rewards unnecessary.

A different strand of research shows what happens when reward or punishment structures are handled carelessly. Economists Uri Gneezy and Aldo Rustichini studied ten day-care centers that introduced a small fine for parents who picked up their children late. Instead of reducing late pickups, the fine increased them, and late pickups remained higher even after the fine was removed. The introduction of a price appeared to replace an implicit social and moral obligation with a market transaction, and once that transaction existed, removing it didn’t restore the original norm. The lesson for reward fading is the mirror image: introducing and removing extrinsic incentives can permanently reshape how people frame a behavior, so the fading process needs to preserve the identity-based or values-based reasons for the habit, not just the reward-based ones.

The Reward-Fading Protocol: A Step-by-Step Guide

The following protocol assumes a habit that has already been running for at least a few weeks, using an external reward such as points, streak bonuses, small purchases, or app-based badges. It is built around gradually reducing reward frequency and intensity while increasing attention to internal cues, using the automaticity window from the Lally research as a rough pacing guide.

A few notes on applying this protocol well:

1. Start the clock only after the habit is stable. If a behavior is still irregular, week 1 of the fading protocol should wait until the behavior itself is happening on most days, not just some days.
2. Move the reward’s target, don’t just cut its size. Rather than reducing a $5 reward to $1, shift from rewarding every action to rewarding a streak. This keeps the reward feeling meaningful while reducing how often it’s expected.
3. Pair each fade step with a reflection prompt. A one-line note (“what felt automatic today?”) gives the brain something to attribute the behavior to besides the missing reward.
4. Watch for a specific slip signal. If completion rate drops more than about 20 percent after thinning a reward, pause the fade for a week and hold the current frequency steady before continuing.
5. Keep the cue and routine constant while the reward changes. The Lally research found that consistent context helped convert repetition into automaticity; changing the trigger and the reward schedule at the same time makes it hard to tell which change caused a slip.

Signs a Habit Has Become Self-Sustaining

Reward fading should be guided by evidence, not a fixed calendar. The five-phase timeline above is a starting frame, but the actual pace should track how the habit is behaving.

If most of these signs are present, the fading protocol can move faster than the default timeline. If few of them are present by the time a fade phase is scheduled to begin, hold the current reward level for another two to three weeks before thinning it further.

It’s also worth distinguishing a habit that has become self-sustaining from one that has simply become low-effort. A habit can feel easy to perform and still be running mostly on the reward attached to it; the test is what happens on a day the reward is unavailable. A self-sustaining habit survives an unrewarded day with only a small dip in the desire to do it. A reward-dependent habit tends to produce a noticeable internal negotiation (“is it even worth doing without the points?”) on that same day.

How to Choose Which Reward to Fade First

Most habits accumulate more than one reward over time: a streak badge in an app, a small treat, praise from a friend, and a general sense of accomplishment can all be attached to the same behavior at once. Fading works best when it targets rewards in a deliberate order rather than removing everything simultaneously.

Start with the reward that is most tangible and most closely tied to a single instance of the behavior, such as a point earned for one gym session or a treat purchased after one meal-prepped lunch. These instance-level rewards are the ones most strongly associated with the overjustification effect, since they are easy for the brain to treat as the reason for the action rather than a side effect of it. Milestone rewards, such as a celebration after a full month of consistency, can stay in place longer because they reinforce a pattern rather than a single repetition.

Leave social or relational rewards, such as a workout partner noticing your progress or a family member commenting on a healthier dinner, until last, or don’t fade them at all. These rewards tend to satisfy the relatedness need described in self-determination theory, and removing them doesn’t reduce dependency on an external prize so much as it removes a genuine source of connection that happens to support the habit.

A simple ordering to follow: fade instance-level tangible rewards first, then fade or space out milestone-level tangible rewards, and treat social recognition and personal reflection as the layer that stays in place indefinitely, since neither functions like the kind of controlling reward that produces the overjustification effect.

Common Mistakes When Fading Rewards

Three mistakes account for most failed attempts at reward fading.

Cutting the reward all at once. An abrupt stop removes the scaffolding before the habit has had time to build its own support structure. Because the overjustification effect can leave motivation lower than baseline once a reward is withdrawn, a sudden cutoff risks a sharper drop than existed before the reward was ever introduced.

Fading before automaticity develops. Since the median time to reach a habit’s automaticity plateau in the Lally data was 66 days, with a wide individual range, fading a reward at week three is often premature. The habit hasn’t yet built the cue-response link that’s supposed to replace the reward.

Replacing one extrinsic reward with another instead of building intrinsic reasons. Swapping a points app for a cash jar doesn’t fade the reward, it just changes its shape. Real fading requires shifting attention toward competence (visible progress), autonomy (the habit is a choice, not an assignment), and relatedness (the habit connects to people or values that matter).

Frequently Asked Questions

How long should I keep rewarding a new habit before I start fading?

Wait until the behavior is happening on most days without much internal resistance, generally at least four to six weeks, and ideally closer to the 66-day average automaticity window identified in habit-formation research. Fading too early is a more common mistake than fading too late.

What if my completion rate drops when I reduce the reward?

A drop of more than roughly 20 percent after thinning a reward is a signal to pause, not abandon, the fade. Hold the current reward frequency for another one to two weeks, reinforce the cue and routine, and try the next fade step again once completion stabilizes.

Should I fade all rewards, including praise from other people?

Not necessarily. Research on cognitive evaluation theory suggests that praise or feedback conveying competence, rather than praise tied to receiving a prize, tends to support rather than undermine intrinsic motivation. Unexpected, informational feedback can often stay in place even after tangible rewards are gone.

Is it bad to keep a small reward forever for a habit I find boring?

Not automatically. Meta-analytic research has found that low-interest tasks are less susceptible to the undermining effect, and a small, sustained reward can be a reasonable long-term strategy for habits like flossing or medication adherence that may never become intrinsically enjoyable.

Can I use the same fading timeline for every habit?

Treat the five-phase timeline as a default, then adjust based on the signs of self-sustaining motivation described above. A habit tied to a hobby you already enjoy may fade faster; a habit that’s inherently unpleasant may need a longer reward tail or a permanent light-touch reward.

What should replace the reward once it’s fully faded?

A brief, honest check-in with yourself about what changed because of the habit works better than a substitute prize. Naming a specific, felt benefit reinforces the internal reasons for the behavior instead of quietly reintroducing an external one.

Conclusion

Rewards are not the enemy of a lasting habit. Used carelessly, tied to every single repetition indefinitely, and removed all at once, they can leave a habit weaker than if no reward had been used at all. Used deliberately, on a schedule that respects how long automaticity actually takes to build, and paired with attention to competence, autonomy, and connection, rewards can hand a habit off to its own internal reasons for existing. The fading protocol above is a starting structure. The real signal to follow is what the habit does on the days the reward isn’t there.

References

• Deci, E. L. (1971). Effects of externally mediated rewards on intrinsic motivation. Journal of Personality and Social Psychology, 18(1), 105-115.
• Lepper, M. R., Greene, D., & Nisbett, R. E. (1973). Undermining children’s intrinsic interest with extrinsic reward: A test of the “overjustification” hypothesis. Journal of Personality and Social Psychology, 28(1), 129-137.
• 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.
• Cameron, J., Banko, K. M., & Pierce, W. D. (2001). Pervasive negative effects of rewards on intrinsic motivation: The myth continues. The Behavior Analyst, 24(1), 1-44.
• Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55(1), 68-78.
• Lally, P., van Jaarsveld, C. H. M., Potts, H. W. W., & Wardle, J. (2010). How are habits formed: Modelling habit formation in the real world. European Journal of Social Psychology, 40(6), 998-1009.
• Judah, G., Gardner, B., Kenward, M. G., DeStavola, B., & Aunger, R. (2018). Exploratory study of the impact of perceived reward on habit formation. BMC Psychology, 6, 62.
• Gneezy, U., & Rustichini, A. (2000). A fine is a price. The Journal of Legal Studies, 29(1), 1-17.