Although incentives can be good motivators, they can backfire if they aren’t used properly. Being offered a reward for doing something doesn’t necessarily make people more motivated to do that thing—it just makes them motivated to get a reward. The same goes for threats of punishment. When people are only motivated by incentives, they lose interest in work that has no incentive attached to it, and often do not keep up the behavior once the incentive is gone.

Rewards and punishments are very commonly used as tools for self-regulation, parenting, and workplace management—but the people employing them do not realize that relying on incentives can sap other intrinsic motivations to do things. By employing incentivization too often, we may be making ourselves less passionate, less creative, and less engaged with our work.

Incentives are based on classical psychological theory

Incentivization is rooted in behaviorism, a theory of learning that gained popularity in the mid-20th century. According to behaviorism, all animals (including humans) learn how to act in the world based on the consequences of their past behavior. Simply put, if you act a certain way and it leads to a positive outcome, that behavior will be reinforced, and is more likely to recur in the future. On the other hand, if behavior is punished or leads to a negative outcome, the animal becomes less likely to repeat it. This type of learning is known as operant conditioning. (This contrasts with classical conditioning, which you might be familiar with through the story of Pavlov and his dogs.)

One of the major players in the behaviorist movement, and the one most associated with operant conditioning, was B.F. Skinner. Skinner labeled himself a “radical behaviorist,” because, in his view, the way an individual thinks or feels about a situation is irrelevant; all that matters is whether they have had a given behavior reinforced in the past. Even if you don’t agree with such a hardline stance, it’s undeniable that Skinner was able to use the principles of behaviorism to achieve some pretty remarkable things: during World War II, the military even hired Skinner to train pigeons to guide missiles. (He also trained pigeons to play ping pong, though that project probably was not funded by the U.S. army.)

The power of operant conditioning showcases the influence that reward and punishment have on our behavior. Once we have learned that doing something will bring us a positive outcome, we’re motivated to do it; and when we learn that doing something will lead to negative consequences, we’re less likely to repeat that mistake again. Incentivization plays on this very basic principle of human behavior.

Dopamine drives us to “want” incentives

Behaviorism had its heyday nearly a century ago, but it continues to be influential in modern-day psychology. In fact, in the decades since B.F. Skinner and his whimsical ping pong-playing pigeons, neuroscience research has fleshed out the biological reasons behind the power of incentives.

Even if you don’t know much about neuroscience, chances are you’re familiar with dopamine. Dopamine is a neurotransmitter, or brain chemical, that is most often associated with reward. You’ve probably seen it mentioned in media coverage of any topic relating to pleasure or compulsion, whether it’s about the addictive power of sugar or why we can’t stop using our smartphones. A lot of popular discussion of dopamine labels the chemical as the thing that creates pleasure.

Dopamine does indeed play a big role in pleasurable experiences. However, things are a bit more nuanced than simply having your brain flooded with dopamine anytime you see a Cheeto. Research has shown that our brains have multiple reward systems, which serve different functions. As it turns out, the mechanisms that make us “like” something are different from the ones that make us “want” it.

The “liking” system is active when we are having an enjoyable experience, such as eating an ice cream cone. This pathway actually does not rely on dopamine, but instead of endogenous opiates (“endogenous” meaning “produced within the body”). In contrast, the “wanting” pathway—the system that motivates us to act if there’s an incentive on the line—is dominated by dopamine.

“Wanting” is also associated with a specific area of the brain, known as the mesolimbic pathway. This pathway begins with a part of the midbrain known as the ventral tegmental area (VTA) which is rich in dopamine-producing neurons. When we want something, these neurons release dopamine into an area called the nucleus accumbens. This process gives rise to incentive salience, or the expectation of pleasure.

The distinction between these two systems is important, because it means we can want something without necessarily liking it. This is especially relevant to drug addiction. People who have a substance use disorder compulsively seek out opportunities to use the substance they’re addicted to, but this is less because they enjoy the experience of using and more because they are driven to crave the substance by the neurochemical processes of the “wanting” pathway.

Even outside the context of drug addiction, many (if not all) of us have experienced the feeling of being compelled to do something, even after our enjoyment of that thing decreases—for example, going in for more dessert even after you’re already stuffed. Once we have learned to associate something with pleasure, dopamine often kicks in to motivate us to seek out more of it.

Incentives offer extrinsic motivation

Although dopamine does drive us to pursue rewards, the full picture of motivation is more complex than just that. Needless to say, there are many things that could potentially motivate us to do something, and many levels in which they can motivate. There are some tasks we are actively excited to do, that give us a sense of satisfaction—and then there are others that we only do because we have to, in order to achieve some other end.

Psychologists often categorize these motivators according to whether they are “intrinsic” or “extrinsic.” Things that are intrinsically motivating are inherently enjoyable; we do them because they bring pleasure or a sense of fulfilment. Meanwhile, extrinsically motivated tasks are things we do because they lead us to some separate outcome that we find appealing.

In cases where we lack any intrinsic motivation to do something, having an incentive to act as an extrinsic motivator can push us to follow through. In one study, for example, participants were offered financial rewards for going to the gym. Initially, they were offered $25 to go to the gym once in the following week; after that, some of the students were additionally offered $100 to exercise eight more times over the following four week. (An unlucky final group, the control group, received no incentives.) The researchers found that the group that was offered $100 for eight gym visits had significantly higher gym attendance both during and after the experiment, compared to the other groups.

The results of research like this suggests that incentives can work as a kind of “self-commitment device”: they can motivate us to persevere with a new habit for long enough for it to stick, so that even after the incentive disappears, we’ll keep up the new behavior.

Incentives might “crowd out” intrinsic motivators

Although adding a source of extrinsic motivation can sometimes help to kickstart a change in behavior, there can be drawbacks to doing so. Being extrinsically motivated may be better than having no motivation at all, but compared to intrinsic motivation, it has a lot of downsides, including lower enjoyment of an activity, lower effort, and a greater likelihood of blaming other for negative outcomes.

The real danger of relying too much on incentives lies in something known as the overjustification effect, or the crowding-out effect. This describes how in some cases, where people treat an activity just as a means to some other end (i.e. when they are extrinsically motivated), it can undermine intrinsic motivators for doing that same task. Paradoxically, this means that some incentives can actually backfire, and achieve the opposite of the intended result.

In a famous study that illustrated the overjustification effect, nursery school children were given art supplies to play with, including markers, crayons, construction paper, and so on. At some point, the children were encouraged to use the markers to draw a picture for an adult who was visiting the school. One group was also told that students who drew pictures for the guest would receive a “Good Player Award.” In the few days following the experiment, the researchers then measured the amount of time the children spent drawing with the markers during their free time. The results showed that children who had been given an incentive to play with markers—the Good Player Award—spent significantly less time using markers afterwards, compared to students who hadn’t been given an extrinsic motivator.

Some incentives only work in the short term

Another weakness of incentivization is that its effects often don’t last once the incentive is removed. For example, in one study that looked at the long-term effects of incentives to help people quit smoking, participants were offered $20 for each smoking-cessation class they attended, and another $100 if they quit smoking for 30 days after the program was over. In the short term, the incentives were helpful: significantly more people who were offered the cash rewards completed the class and quit smoking for those 30 days, compared to people who weren’t offered an incentive. However, after six months, there was barely a difference in quit rates between the incentives group and the others.

This doesn’t mean that incentives are a waste of time. In some cases, short-term success might be good enough to achieve a goal. For example, if the goal of a program is to help pregnant women quit smoking just until their pregnancy is over, short-term results are still much better than nothing. However, if the goal is to encourage a change in behavior that will last for a long time, incentives are probably not going to be enough.

The reason that incentives are only temporarily effective is that they don’t involve changing someone’s underlying attitudes about an issue, or spark a more meaningful, long-term commitment. Instead, they encourage people to be compliant as long as they have to be—and unfortunately, this doesn’t tend to result in the permanent change that people like business managers or educators are hoping for.

Incentives focus on quantity, not quality

Most incentives are based on improving aspects of performance that are easy to measure, such as speed or volume of output. But even if incentivization can improve the quantity of someone’s work, it might come at the cost of quality.

When people are incentivized to do something, they are more motivated to receive a reward, or avoid punishment, than they are to do a good job. This often means people avoid taking risks because they don’t want to risk the incentive. Studies have found that when people have been offered a reward to do something, they try to make the job as easy as possible, minimizing difficulty. This results in less creativity; less thinking outside of the box.

Incentives are commonly used in many different situations to try to encourage us to behave a certain way. We might even try to give ourselves incentives to help make a lifestyle change—things like promising ourselves an expensive meal out if we go to the gym a certain number of times, or a vacation if we succeed in saving our money. On the flipside, for those of us who are parents, or work in management or leadership positions, incentivization is often the go-to strategy to try to encourage certain kinds of behavior from others.

In general, our culture prizes reward and punishment as a method of getting the results we want. Given how ubiquitous this mindset is, it pays to know when incentives work and when they don’t, and why. With a solid understanding of incentivization, you’ll be less likely to waste your time throwing incentives at projects where they’re not likely to work, and better positioned to figure out an effective incentive plan in cases where it does.

The takeaway from everything about isn’t that incentives are ineffective; it’s that they have to be applied thoughtfully. Here are some general rules of thumb about when incentivization can be most effective.

Use incentives to build new habits

Even though incentives are generally more likely to work best with shorter time frames, this could still be translated into long-term change if the behavior being encouraged becomes an established habit. Although people may initially just be going through the motions because there’s an incentive to do so, if there are enough positive consequences to adopting a new behavior, they might end up being intrinsically motivated to keep it up on their own.¹¹

The effectiveness of incentives in getting people to exercise more is a good example of this. Going to the gym is a daunting task for many of us, but regular exercise comes with a lot of benefits. Some of these are benefits that will only show up in the long term, such as decreased risk of heart disease and diabetes—but others, such as improved mood and increased energy, can be felt pretty quickly. “Side effects” like these might be enough to get people intrinsically motivated to keep exercising. Incentivization, then, is just a way to kickstart the formation of this new habit.

Deliver incentives sooner rather than later

If the only benefits of exercise were a decrease in our chances of getting sick in the future, most of us would have a much harder time doing it regularly. It’s much easier to get motivated to do something if the positive consequences of doing it are immediate and tangible, not abstract, and far off in time. This logic applies to incentives, too: trying to incentivize behavior by promising a reward (or threatening punishment) in the distant future is much less likely to work than incentives delivered quickly.

Use incentives sparingly

Imagine a workplace where management constantly uses incentives, such as bonuses, to motivate their employees. Over time, workers lose interest in their jobs, and are only motivated when there’s an incentive in place. Supervisors see this and say, “You see? If you don’t offer them a reward, they don’t get anything done.”

Overusing incentives creates a self-fulfilling prophecy: incentives are put in place because it’s believed that they are necessary for motivation, but in turn, having an external incentive saps people of intrinsic motivation to do things. In the long run, this will hurt people’s performance and output, rather than help it.

The takeaway here is not to rely too much on carrots and sticks to change behavior, whether it’s your own or someone else’s. Only use incentives where they’re likely to work, and don’t do it too often. And for the times where incentivization might not be the best choice, there are many other strategies that you can use to tackle behavior change, such as behavioral nudging.

Operant conditioning, or the use of reward and punishment to reinforce certain patterns of behavior, was first studied by Edward Thorndike, a famous American psychologist. Thorndike studied behavior by putting cats into a “puzzle box,” which could be opened from the inside by pushing a lever, and timing how long it took them to get out. After they escaped, he would put them back in and time them again. On the first try, cats only got out of the box when they accidentally stumbled upon the lever, but as the experiments went on, they got faster and faster at escaping—because they learned how the lever worked. Based on this, Thorndike coined the “law of effect,” which states that any behavior that is followed by a positive outcome is likely to be repeated.

The law of effect became the basis for the work of B.F. Skinner, who developed the concepts of reinforcement. Similar to Thorndike’s puzzle box, Skinner created a chamber now known as the “Skinner Box,” which contained (among other things) a lever that rats could push to dispense food for themselves, as well as an electric grid that could be used to give the rats a shock.

In one experiment, elementary school students in Ohio were offered financial incentives as high as $100 if they could improve their grades in school. The results showed that this program worked to improve grades—but only in math, and not in other subjects such as reading or social science.

Why would incentives work only for math? One possible explanation is that incentives are good for concrete tasks, such as practicing doing equations, but not for more conceptual topics.

Generally speaking, people are trusting of one another; we are intrinsically motivated to be kind and caring towards one another. Incentivization can potentially damage this trust.

Research has shown that adding an incentive for behaving altruistically can trigger the overjustification effect—that is, detract from our intrinsic motivation to be altruistic. If we believe somebody has an extrinsic reason to be kind to us, we become distrusting of them: “They’re only being nice because there’s something in it for them.” This means that trying to incentivize good social behavior might backfire, and undermine the bonds that naturally exist between us.

What it is

Incentivization is the use of rewards and punishments to spur people to act a certain way.

Why it works and why it doesn’t

On the one hand, incentivization provides extrinsic motivation to do something, and plays to the “wanting” pathway in our brain that causes us to seek out reward. On the other hand, incentives can diminish our intrinsic motivation to act, and can hurt the quality of our work if they are used too often. Incentives are also more effective for short-term behavior change than sustained changes, unless the new behavior becomes a habit that is fulfilling enough to sustain itself.

Example 1 – Incentives and grades

Financial incentives to improve grades in school were successful in one experiment—but only for math. This suggests that incentives work best when the task at hand is concrete, not conceptual.

Example 2 – Incentives and trust

People are intrinsically motivated to trust and be kind to one another—and incentives can damage this by triggering the overjustification effect. If there is an external incentive to be kind, we attribute people’s good behavior to that incentive, and become distrusting of their motives.

How to avoid misusing it

To avoid the negative consequences of incentivization, don’t use external incentives too often, and only apply them to situations where they are likely to work (for example, to kickstart the formation of a new habit). Also, make rewards immediate and tangible.