Why bribing your child doesn't work.

Snapshots of life at home.
March 26 2010 7:17 AM

If You're Good, I'll Buy You a Toy

The difference between bribing your child and rewarding your child.

Illustration by Rob Donnelly. Click image to expand.

The very idea of rewarding children for good behavior rubs a lot of adults the wrong way. Even when they accept that positive reinforcement changes behavior a lot more effectively than punishment does, they still object to using rewards in general and specifically for behaviors one ought to expect as a matter of course from a child. They just can't bring themselves to praise a 7-year-old and give her points on a point chart or a little prize because she didn't throw a tantrum in the supermarket.

Rewarding the desired behavior is just one element of positive reinforcement, which a deep body of reputable research over several decades has established as the most effective way to change behavior. But many parents (and teachers and babysitters) reasonably object that giving a child a reward amounts to bribing him. He's just doing it for the reward, they say. They want him to behave for other, better reasons: because he should, because it's his responsibility, because they say so.

The families who come to Alan's Yale Parenting Center often resist using rewards, which leads to some typical exchanges with the trainers there.

Objection: I don't want to reward behavior the child should do anyway because it's part of his normal responsibilities.

That's a reasonable position. However, if the child is not doing the behavior now, or not often enough, or not without a struggle, it seems practical to ask: "What's the best way to jump-start it?" Reprimands and threats don't work very well, and they often have side effects that increase noncompliance and decrease the likelihood that the child will do what's requested. Typically, reprimands and threats function just like bribes. They might get the desired behavior for the moment, but the child won't continue the behavior beyond just responding to the threat. That is, the behavior will not be locked in as a habit or as an expression of a general characteristic we wish to develop, such as honesty, kindness, or generosity.

Objection: If we use rewards to get the behavior, the behavior will stop once we stop giving the rewards, which means we'd have to give the rewards forever to keep the behavior going.

That's a nightmarish scenario—the kind of perpetual bind that people taking certain medications, like those that lower blood cholesterol, can find themselves in. Will you have to give your child two stickers on her point chart for using the potty before she steps into the limo on the way to the senior prom? Fortunately, that's not how it works when you use rewards in an effective way to improve behavior.

You praise your child at the toddler stage for using the potty and for saying "thank you" and for not using the tablecloth to wipe her nose, but by the prom-going stage (and, ideally, by about 14 years before it), she has these habits down cold and no longer needs to be reinforced in them. The relatively brief period during which you praised these behaviors has been over for a very long time. Whether a behavior is maintained and becomes a habit has a lot to do with how the rewards are given. The key is not in the rewards themselves but in a consistent way of getting the child to carry out the behavior and praising or otherwise rewarding it when it occurs. (Don't forget: Attention and praise are rewards and are, in fact, your most reliable rewards.) Unsystematic one-shot incentives will lead to dependence on the reward, and the behavior is indeed likely to stop once you stop rewarding it.

From the perspective of changing behavior, that's the crucial difference between a bribe and a reward. A bribe, even in those cases when it's ongoing over a long period of time, is not part of a systematic effort to develop behavior so that it persists when the bribe is no longer offered. Even if you've paid the same guy in City Hall to fix tickets for you for years, if you stop paying him, he'll stop fixing your tickets. A reward, by contrast, when used properly in combination with antecedents (what you say and do to prompt the desired behavior) and shaping (rewarding partial success in order to build up the behavior) and other elements of a systematic effort to lastingly change behavior, is a temporary measure that can be phased out once the behavior is established. The behavior continues without the reward, just as a finished building continues to stand after the scaffolding comes down.

Objection: Rewards will ruin my child's inner motivation to do things for their own sake.

No, they won't, not if used properly. While an improvised, one-shot "do this for that" will not develop the behavior you want and may undermine intrinsic motivation, rewards used to systematically develop behavior can lock in habits that become independent of any rewards. An example many adults can relate to in their own behavior is exercise. At first, it may be a struggle to exercise regularly. When it's time to work out, you feel a strong urge to sleep in or quietly watch reruns of Cheers until the moment passes, and you may not get to the gym at all unless your significant other intervenes to offer a timely reward (you're an adult; use your imagination). But repetition of the behavior locks in the habit, and the behavior then becomes its own reward. You might even become a fanatic who gets mildly disturbed or seriously irked if illness, work, or some other annoying obstruction prevents you from getting to the gym on a given day. There's no magic in the rewards themselves. They are merely aids to get the behavior to occur repeatedly, so that the habit can develop.

Objection: Rewards are unnatural consequences. When my child grows up, she'll see that the world is not waiting around to hand out praise or video games for doing what she's supposed to do.

A fair point, but one might quibble with the idea that the home should be exactly like other situations the child will experience when she grows up. The home is a unique place where a child develops resilience, competences, attachments, and support in ways that have little to do with later experiences except that they prepare the child to weather and learn from life. Also, the proper way to give rewards has some other redeeming features, of which more in a moment.

Objection: Rewards don't work. I tried one of those point charts, and it didn't change a thing.

A point chart—if you even need one at all—can work very well. And yet, almost all families that come to the Yale Parenting Center have tried point charts and the rewards that go with them. Those schemes haven't worked, obviously, or they wouldn't be coming to the center to address their conduct problems. But that doesn't justify concluding that reward programs don't work, period. The equivalent would be driving in neutral with the emergency brake on and saying that cars don't work. There is research about how to make such programs work, and that same research explains why so many of them fail:

1.Winging it. Some parents improvise makeshift rewards on the spot and fire them from the hip at the moving target of the child's conduct. "Billy, go get my purse from upstairs and you can have 5 minutes more of TV"... "Billy, we're going to the store; if you get your shoes and coat right away, you can have a snack when we get back"... "Billy, if you go keep your sister company for a few minutes, we can make your bedtime a few minutes later." These parents are winging it, throwing together rewards on the fly to encourage a shifting range of behaviors. This approach confirms the wisdom of every objection we've just considered above. It will not develop intrinsic motivation or habits, or at least habits parents want. The child may do each desired behavior, but only for the specific reward at that moment—that is, for the bribe. Once the reward stops, each of the behaviors is likely to stop as well.

2. The Hail Mary reward system. Some parents give rewards for a big outcome that occurs at the end of a long period. Promising a mega-prize for good grades at the end of a year is perhaps the most common example. Parents who want their child to study, do homework, and get good grades—and who have learned that pleas, reprimands, and threats don't often work—often turn to the Hail Mary. Typically, they promise their child a big reward—a trip to Disney World, use of the car—if she gets all A's at the end of the term. This is a misuse of rewards that's nearly guaranteed to fail.

The goal of a reward program should be to build specific behaviors, actions, and habits: bringing home your homework assignments regularly, doing them regularly, studying with a parent and then by yourself, and talking a little bit at the dinner table about one or two things you have studied. All of these are specific behaviors we want to develop and lock in and that can in fact contribute to good grades, not to mention learning. The focus on long-term outcomes alone—A's at the end of the year—puts the emphasis on the wrong thing and creates a crucial lag between the proper day-to-day behavior and the reward intended to reinforce it. Even if the child gets the grades, that outcome can occur for all sorts of reasons (including cheating) and does not always necessarily help develop the good habits we want to develop.

3. Complex reward systems. Some parents who come to the Yale Parenting Center have created complex point systems featuring multiple prizes that can be earned with different point totals, and bonus gimmicks galore. They also make physically beautiful point charts on which earned points are tracked as, for instance, spots on a leopard or ladybug, or as planets or stars in space. There's no reason to object to charts that are more creative and fun—just as long as you bear in mind that it's not necessary or more effective. But if you get too fancy with the reward system itself, you can undermine your own results.

The more complex the system, the harder it is to stick to and administer properly. Yes, if done correctly, it may well work. Yes, variations of systems like this have a technical name—a token economy—and have been used in basic training in the military, in educational settings from preschool through college, in assisted-living facilities to increase activities and social interaction among the elderly, and in workplaces to promote safe shop-floor habits. In all these settings, even fairly complex rewards systems have been proven to work. They can work in the home, too, but you probably don't need to use them. Parents can shape most of the behaviors they want by offering as rewards no more than attention, praise, and a few minor privileges.

A parent's attention is very rewarding to a child, and praise is even better. Parents are giving attention all the time, and they are giving mild forms of praise, verbal and nonverbal (a smile, a touch, an affectionate or impressed look). Attention and praise are our main rewards, and often they're sufficient to change behavior on their own, without resorting to tokens, privileges, or prizes.

But attention and praise can be used more precisely than they usually are. It's natural to make the mistake of devoting most of our attention to misbehavior, rather than to good behavior, and praise is often sporadic, not very specific (the generic "good job" doesn't convey precisely what the child did right), not particularly enthusiastic, and not systematically connected to some behavior that the parents would like to develop.

When you're consciously using praise to change behavior, the key is noticing and praising the behavior promptly, specifically, and precisely so that it occurs again and can be praised again. We want repeated practice to lock in the habit, and praise helps that to happen.

To make praise effective:

1. Be specific about the behaviors you want. Explain to yourself, first, and then to your child what the specific behavior is you want to develop. Vague commands like "be nice" or "show respect" are too general. Rather, say, "When you're playing with your sister, keep your voice down and don't take her toys" or "When you talk to Grandma, keep your body calm and don't make that obscene gesture Grandpa taught you." When you see the desired behavior, praise it specifically and enthusiastically (the younger the child, the more enthusiastic you should be): "You stayed in your seat all through dinner and you used your inside voice. That's great!"

2. Identify a small number of behaviors. Start with no more than two or three behaviors you want to develop in the child. You will be able to replace them later with new behaviors once these first few are developed. Remember, the reward does not produce the results—rather, you want to encourage repeated practice of the behaviors, or of samples or approximations of these behaviors. You want to focus on getting a couple of behaviors locked in as a habit, then move on to the next ones.

3. Model the behaviors you want. Show the child exactly what the behavior would look like. Even if she "knows," it will help if you demonstrate. Then have the child do it and praise her for whatever parts she copied correctly from you. If you see other people in everyday life—while you are at the store with your child or out to dinner or hanging out—point out the desirable behavior you just saw. Other people can be used as models if you punctuate what you see with your comments and approval.

4. The key is repetition, so practice. We reward behavior in order to encourage repetition, which is called reinforced practice. If your child is already occasionally doing the behavior you want, systematic praise can lock it in as a habit. If the child does not do the behavior yet in the course of normal life, practice it in pretend, gamelike circumstances. You do it, have her do it, and praise.

A young child who may not go to bed right away at bedtime can get better at this if you practice a bit. This can be pretend practice during the middle of the day. Set it up as a game. He goes to his room and gets into bed, pretending it's bedtime. You praise all of that. Practicing this a few times in a given week will increase the likelihood that he goes to bed better at bedtime—and be sure to praise that, too.

Practicing the behavior in pretend situations is still practice and still builds the behavior. Remember, commercial airline pilots practice all of the time in "pretend" situations (called simulators), and that practice carries over to real situations. Boxers practice and train in the gym to develop habits they will call upon on fight night. It's the same with behaviors you want to develop in the home.

5. Shape the desired behavior by rewarding gradual approximations of it. If the behavior does not yet occur the way you want it—not a full hour of bassoon practice, not a full hour of homework each day—praise lesser durations and partial successes and build them up over time.

The use of praise to develop behavior in the way we've been describing it is systematic, temporary, and a purposeful addition to—but not a substitute for—the ordinary warm attention you normally give to your children just because you love them. Why not test it out for a brief period? Let's say three days. For three days, try to use praise as a reward in the ways outlined above. You should see a difference in your child's behavior and in the emotional climate of your household.

Praise is one in the larger set of positive experiences that builds the relationship between a parent and child. Many parents and others in authority rely instead on threats, hitting, and punishment more generally, especially when trying to teach respect. More punishment leads to children escaping and avoiding the punisher; more effective praise makes for closeness and better relationships. Because praise for positive behaviors decreases the need for punishment, it helps to make a family closer and warmer. In seeking to strike the balance between them, the rule of thumb when trying to change behavior is that praise for good behavior should be much more frequent than punishment for the behavior you want to eliminate.

None of this looks anything like bribery. Rewards tend to become bribery when they're unsystematic, unconnected to clear parental expectations, and used to try to get your kids to do things at odds with the model you provide with your own behavior.

Alan E. Kazdin, who was president of the American Psychological Association in 2008, is John M. Musser professor of psychology and child psychiatry at Yale University and director of Yale's Parenting Center and Child Conduct Clinic.

Carlo Rotella is director of American studies at Boston College.