Punishment, Reinforcement, and Discipline


I am curious about the line between discipline (which I think of as consistently reinforcing the right behavior and preventing the negative) and punishment. Do children this young ever need actual punishment and if so, what would be appropriate? I know this comes close to the spanking issue which may have a bunch of new research (as in newer than Dr. Spock; I’m pretty behind). —Tom

Thanks for bringing this up—it’s a complicated and sometimes confusing topic. Let’s go over some terminology, and then talk about what’s appropriate for young children. The words to think about are discipline, punishment, reward, and reinforcement. People often use those words interchangeably, and they are related—but they’re different in important ways.

Discipline means guidance toward appropriate behavior. Pretty broad, right? It can be (and is) approached in a wide variety of ways by different people with different values and different cultures. Basically it’s anything you do to teach children. But there’s a sense of control in it as well, which I think is important—how do you, as a caregiver, appropriately control children’s behavior, and how do you help them to become in control of themselves? I like to think of discipline as an ongoing process of handing over control to the child: the more they learn, the more they can be in charge of making good decisions about their behavior. The word discipline in our culture has come to have a negative connotation of punishment—in fact, the National Association for the Education of Young Children has stopped using the word “discipline” entirely, because of its connotations, and uses the word “guidance” instead. But discipline doesn’t necessarily mean anything negative at all. In fact, learning and guidance (i.e. discipline) are crucial and positive parts of every child’s life.

The related term that gets most inexactly bandied about is “reinforcement.” It’s a term from psychology and behaviorism with a very precise definition, but for our purposes we can say that reinforcement is something that happens because of a behavior that makes the behavior more likely to keep happening. Positive reinforcement is something good that happens that makes you want to keep doing whatever caused it. For instance, “Every time I hug mommy she smiles at me! I’ll keep hugging her!” or “Every time I poke my brother mom pays attention to me! I’ll keep poking him!” Negative reinforcement is something bad that stops happening when you do a behavior. You know when you start the car and there’s an annoying buzzing sound until you buckle your seat belt? That’s negative reinforcement that makes you more likely to buckle your seat belt, because every time you buckle it you feel better that the buzzing has stopped.

Reinforcement is how training works, and it’s complicated. For instance, let’s say you’re in line at the grocery store and your child whines and whines until you give in and buy him a candy bar. You’ve received strong negative reinforcement for the behavior of buying a candy bar—the annoying whining stopped! But your child received strong positive reinforcement for whining—he got candy. That interaction makes it more likely that you’ll both do the same thing next time.

Reinforcement is how the world teaches you how to behave, naturally. One time you didn’t know a pan on the stove was still hot and you got burned; now every time you touch anything on the stove, even if you know for sure it’s cold, you always double check. One time you found a five dollar bill on the ground outside the coffee shop; now you always finding yourself looking on the ground there, even though you know it’ll probably never happen again. Reinforcement often works unconsciously—in fact, though we often try to actively use reinforcement in our personal lives, humans are a little perverse and sometimes rebel if they realize they’re being trained.

Punishment and reward are similar to reinforcement, with a crucial difference: they happen after the behavior in question is done, and changing your behavior won’t change the punishment or reward. They therefore, intrinsically, don’t do as good a job changing behavior. Reinforcement is a good teacher because changing your behavior immediately causes a change in the reinforcement; reinforcement makes you actually do the desired behavior, in the moment. Punishment and reward are less effective teachers because changing your behavior has no immediate effect. Punishment is especially ineffective in a discipline setting because its strongest effect is to damage relationships. When children are punished they may think, “Boy, I’ll never do that again,” but they are almost sure to think, “I am mad” or “I am sad” or “I am scared” or “Nobody understands me” or “I’ll get back at them!” That’s usually not what we’re hoping to teach.

So, to your actual question: Do children ever need punishment? Well, there are a lot of things children need in order to learn to behave appropriately. They need clear boundaries for behavior. They need to know when they’ve stepped over the line and done something unacceptable. They need to know when you’re upset with them. They need opportunities to get it right next time. But punishment isn’t necessary to achieve any of those things.

I believe that physical punishment is simply wrong: it’s just not right to harm someone who you have power over and who depends on you and trusts you. But more importantly, I believe that punishment, physical or not, just doesn’t work very well. When I consult with parents about discipline, if their methods include punishment, I always ask, “Does it work? Does she stop doing it when you punish her?” Almost always the answer is, “No, not really.” It’s more effective to reinforce desired behaviors than to punish undesired ones (or to reward desired ones). If there’s an undesirable behavior that a child is doing repeatedly, dollars to donuts they’re getting reinforced for it; sometimes you’ve got to be a detective to figure out what the reinforcement is, but when you do you can often remove the reinforcement, and the behavior often stops, no punishment required.

Reinforcement is way harder than punishment. It takes much more attention, time, and energy. But it has the primary benefit of actually working a lot of the time, and it has the added benefit of making both the caregiver and the child feel better about themselves.

* * *

For more on successfully using reinforcement methods, I strongly recommend you read Karen Pryor’s Don’t Shoot the Dog!, a book that I have recommended to almost everyone I know. Whether you’re raising children or training your dog or just trying to get along with people at the office, this book will help you understand how to get the behaviors you want out of other living things.

And if you’re having trouble with discipline, I can help!

6 thoughts on “Punishment, Reinforcement, and Discipline

  1. Kerry December 6, 2012 at 4:10 pm Reply

    “Punishment and reward are similar to reinforcement, with a crucial difference: they happen after the behavior in question is done, and changing your behavior won’t change the punishment or reward. They therefore, intrinsically, don’t do as good a job changing behavior.”

    Great article, Jarrod. I’m confused by this sentence, however, and would love to hear more about it. Yes, changing the behavior won’t change the outcome in that instance, but the presence of a punishment or a reward immediately following the behavior will greatly change the odds for the increase or decrease of *future* behaviors. I only ask for clarification, because I have always thought of the reward as the vehicle for reinforcement.

    Overall, fantastic article. I love reading your stuff!

    • jarrodgreen December 6, 2012 at 5:01 pm Reply

      Oh, I see what you mean. I think of rewards as being the analogue to punishments – things that happen because of a behavior, but after the behavior is over. As opposed to reinforcements, which happen during or immediately on conclusion of the behavior. I suppose another way to think about it is that punishments and rewards are reinforcements with a time-delay. I think you’re right that rewards are reinforcing (and reinforcements feel rewarding), but I find it useful to distinguish between what happens in conjunction with the behavior and what happens later.

      That’s just how I think of it though. And I know the psych/behaviorism definition of reinforcement is quite specific, and may not entirely correspond to how I think of it.

      • Kerry Ryan December 7, 2012 at 12:30 am

        Ah, I see. Yes, the psych definition of punishment is anything that causes a behavior to decrease. Time is not a part of the equation.

        In terms of the reward system, Wikipedia offers a good description: “Certain neural structures, called the reward system, are critically involved mediating the effects of reinforcement. A reward is an appetitive stimulus given to a human or some other animal to alter its behavior. Rewards typically serve as reinforcers. A reinforcer is something that, when presented after a behavior, causes the probability of that behavior’s occurrence to increase. Note that just because something is labelled a reward does not necessitate it as a reinforcer. A reward can only be said to be a reinforcer if its delivery has increased the probability of a behavior.”

  2. Tom Birch December 8, 2012 at 5:39 pm Reply


    Thanks for this. It’s been really helpful to start looking at this subject from a different angle and with different language. Speaking of which, I have a few questions about language that aren’t blog-worthy if you have some time.

    First, what is the actual, precise psych/behavior definition of reinforcement?

    Second, I’m trying to find a word/concept label for something and I am hoping you might know. There is an Amish practice called shunning where people are allowed to stay in the community, but no one speaks to or interacts with them. I think the Amish are only different in that they identify this behavior and do it purposely. Is there an official name for this and do you know anyone who has studied it to any extent?

    Last, you & Amelia seem to know so much good stuff about human behavior. I have been consistently grouchy with most of the material (mostly intro level textbooks and – god help us – self-help) I have encountered. If you have some really good authors or titles in the behavior field to share, I’d love some good reading.

    Hope you’re doing well. Hugs to all of you!


  3. […] For my related thoughts on discipline and reinforcement and such, look here and here and here. […]

  4. […] debate is about how the punishments and rewards should be determined. But I would argue (hold on, I HAVE argued) that punishment and rewards aren’t the most effective way to get where you’re going. […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: