Schedules of Reinforcement: When to use positive and negative reinforcement

I’m back at my keyboard after a few weeks’ hiatus during which I spent all my free time landscaping.  I completely understand how the stereotypical scene of the stay at home mom working out in the garden developed.  You can use your hands in a productive manner, making progress rather than just cleaning up and fixing things.  There are small, well-defined projects that can be accomplished and checked off your list rather than the ongoing and endless refinement of child rearing.  And perhaps the most appealing feature of gardening is that your plants cannot talk, whine, yell, etc. 🙂

Back to the topic at hand, we’ve already had an introduction to operant conditioning (see Understanding Reinforcement vs. Punishment from 2/8/18 and Using Operant Conditioning to Train Your Children to Have Good Manners from 3/1/18) so now it’s time to delve deeper and discuss schedules of reinforcement.  Here we’ll continue to focus on parenting techniques derived from operant conditioning (e.g., positive and negative reinforcement) and develop an understanding of how when we use these techniques makes a huge difference in our child’s response.

There are two categories for reinforcement schedules.  First, a continuous schedule of reinforcement means that every single behavior is reinforced. For example, every time your child eats their veggies at dinner they get dessert.

Second, a partial schedule of reinforcement means that the behavior is only reinforced some of the time.  In this example, sometimes your child eats their veggies and they get dessert but sometimes no dessert is offered.

Which reinforcement schedule is better?  Researching his pigeons and rats, Skinner discovered what he dubbed the partial-reinforcement effect wherein behaviors that are only partially reinforced (i.e., not every time the behavior occurs) are longer lasting, less prone to extinction as he called it.  That’s a relief from a parental perspective because you don’t have to reinforce every single desired behavior to get your children to behave well.  According to this theory, if your child gets dessert every night and you stop that reward, they’ll quickly stop eating their veggies whereas if you only sometimes give your child dessert and then stop, they’ll continue eating their veggies for a while longer.

Things get a little more complicated when you look at the different ways that partial reinforcement schedules can be defined. There are 4 subcategories for partial reinforcement schedules: ratio, interval, fixed, and variable.  If the schedule is developed based on the frequency of a behavior, it is a called a ratio schedule.  If a schedule is developed based on theamount of time that has elapsed, it is called an interval schedule. Bear with me; these concepts are going to be so critical later when I introduce sleep training techniques like the Ferber Method and Cry It Out.  It’s about to get a little tricky because it’s the interaction of these subcategories that has utility.  We’re going to need a chart to wrap our heads around these and the different combinations of these subcategories.

  Ratio Interval
Fixed Fixed # of behaviors

Ex: Factory worker

Variable # of behaviors

Ex: Slot machine

Variable Fixed amount of time

Ex: Friday spelling tests

Variable amount of time

Ex: Pop quizzes

In a fixed-ratio schedule, your child’s behavior is reinforced after a fixed number of times, like a factory worker who gets paid $10 for every 100 toothpaste caps he puts on a tub of toothpaste.  For children, this might mean for every 5 times they eat their veggies, they get a treat.

In a variable-ratio schedule, the number of times a child has to exhibit the behavior to get the reinforcer varies randomly, as with slot machines.  So, if your child eats their veggies today, they get a treat, but they won’t get another treat until they eat their veggies 6 times, and after that it will be 3 veggie eatings to earn a treat.

There are still two subcategories to cover.  In a fixed-interval schedule, the behavior must be exhibited after a specified amount of time has elapsed before the reinforcer is given.  For example, a student who has a spelling test every Friday engages in the behavior of studying and is rewarded with a good test grade, but only on Fridays.  Studying during the week might help them on Friday but they only get the reward on Friday.  Back to the dessert example, if your child eats their veggies on Friday, they get dessert and it’s not dependent on whether they ate their veggies the rest of the week.

In a variable-interval schedule, there is a varying amount of time that must pass between rewarded behaviors, as in pop quizzes.  For a young child, this might mean dessert is offered tonight, then not for week, then not for two days, then offered the next day.  In this dessert example, the difference between variable-ratio and variable-interval schedules is subtle – the difference is simply how the reward timing is defined, by the accumulated number of behaviors oran amount of time that must pass before the one critical behavior that earns the dessert.

Which partial reinforcement schedule is the best?  It depends what behavior you are trying to change and what you know about your child’s emotional stability and understanding of delayed gratification.  There are some well-researched phenomena to help guide your reward distribution.  Ratio schedules (fixed or variable) are most likely to increase the frequency of a behavior – the more the child cleans up, the more likely they are to get the treat.  Compared to variable-ratio schedules, in fixed-ratio schedules, you tend to see more of a lull in the desired behavior immediately after the reinforcer is given because the child knows how many times they have to do the desired behavior before earning the next treat.  In fixed-interval schedules (like the spelling test), you tend to see long periods without the desired behavior (studying) then a surge of behavior prior to the end of the interval (the test).  Variable-interval schedules tend to result in consistent patterns of behavior where you study regularly just incase there’s a pop quiz tomorrow.  From a parental perspective, if you want to see change fast, implement a ratio schedule.  If you want to train your child to be consistent in their behavior, variable schedules, whether ratio or interval, are better than fixed schedules – keep them on their toes!  Variable schedules are also harder to extinguish, meaning that your child will keep up the good behavior for a longer time than with fixed schedules even if you remove the reinforcer.  If you’re a psychology nut like me, this is fascinating stuff, though a little tricky to wrap your head around at first.  And the parenting applications are numerous; more on that another day.

Using Operant Conditioning to Train Your Children to Have Good Manners

Circling back to our discussion on reinforcement and punishment (see Understanding Reinforcement vs. Punishment from 2/8/18), now that you have an understanding of the basics, it’s time to learn more about operant conditioning. As a reminder, below is the chart showing the 2 forms of reinforcement, positive and negative, as well as the 2 forms of punishment, positive and negative. Remember that positive means adding something while negative means removing something, as opposed to positive meaning good and negative meaning bad. If I were B.F. Skinner, I would have chosen less confusing terms.

Positive Negative
Reinforcement Adding something good

Ex: Praise

Removing something bad

Ex: Stropping a nagging song

Punishment Adding something bad

Ex: Cleaning house

Removing something good

Ex: Taking away a toy

Skinner’s original works used pigeons and rats to demonstrate his learning theory. Though your munchkins are infinitely brighter than pigeons and I would never compare them to rats, the same concepts apply, sometimes when I’m thinking about how to motivate my children’s behavior, I like to simplify things by thinking of them as little birds pecking away at a lever in order to earn their food pellet. The term shaping is used in behavioral psychology to describe the process of teaching a subject to perform a certain behavior by reinforcing successive approximations of the desired response, just as you patiently guide your children toward desired behaviors. I would like for all of my children to be polite and thankful. For example, I would like my 2 year-old to ask for milk by saying, “Mom, may I please have a glass of milk” but I can’t expect him to magically start speaking this way when he first develops communication skills. First, he cries to tell me he wants milk. Then he grabs or points. Then he says “milk”, then “milk please”, and so on until one day he learns to say the complete, polite request. Each of those steps was a successive approximation of the desired response and was rewarded along the way, but once he was ready to move on to the next step I had to stop reinforcing the previously rewarded behavior (i.e., when he was able to speak, I waited to give him the milk until he said “milk” rather than when he simply pointed to the refrigerator). Fortunately children have a well developed understanding of spoken language long before they can verbally express their own thoughts, so I could prompt him for the behavior by saying, “Would you like a milk? Say, “Milk please Mom.” Now that he can say complete sentences, I don’t hand him the milk until he makes a polite request. I have shaped that behavior over time. You have likely been through this process with your children for any number of desired behaviors. Remember that you have the power to set the ultimate goal, so if you want your children to be polite, don’t stop at “Mom, can I have a glass of milk?” (or one of my pet peeves, “Mom, I want milk.”). Choose not to reinforce that behavior until “please” is added at the end of the sentence.

The same concept of shaping applies to training your children to say “thank you”. In addition to asking politely for the glass of milk, I would like my children to say “thank you” when I hand it to them. This is a behavior that is learned through your child’s interactions with you and other people in their lives with whom they spend a significant amount of time, so get on the same page with other caregivers. Often young children have the best manners when they are new talkers because it is so rewarding for them to receive positive reinforcement from you when they say in their adorable little voices, “thank you” as you hand them the glass (or sippy cup) of milk with a big smile. Your praise acts as a reinforcer beyond the reinforcer of receiving the milk itself. This process of using reinforcement to pair the act of you handing the child the milk with a loving smile with the desired response of them saying “thank you” is called acquisition. This polite behavior may fade over time as the novelty of your child’s response wears off and you react less and therefore unintentionally withdraw part of the reinforcer (they are still getting the milk but your loving smile is not prominent). This is called extinction because you have extinguished, or removed, the connection between being given the milk and them saying “thank you” by removing the reinforcer.

Shaping is then needed to re-aquire the behavior. Now the reward must not be provided until the desired behavior is exhibited (i.e., you don’t give them the milk until they say “thank you”). The most commonly heard cue from parents trying to teach their children to say “thank you” is “What do you say?” after the child already has the milk in their hands. Taking a lesson from behavioral psychology, you’ll find that the desired response is more quickly learned and more consistently exhibited if you without the reward until after the child says “thank you.” To do this you might say “What do you say?” or “What’s the magic word?” while still holding the milk in your hand, but I prefer more subtle techniques like offering the child the milk but not releasing it until they say “thank you”. Don’t think of this as a tug of war but rather, picture your child’s surprised reaction as they go to the grab the milk and discover that it has not been released (this lesson is best taught with sippy cups or water bottles for spilling concerns). Without saying a word, you offer an expectant look (eyebrows raised with a knowing smile), and it’s as though you can actually see gears turning in their heads as they think through “What on earth is mom doing? What is she waiting for? Oh, I need to say thank you.” This can be such a fun learning moment; I have found that my children and I usually leave these interactions with a big smile on our faces, rather than the child feeling like they’re being nagged, yet again, to have good manners. This brings us full circle to the idea of maximizing reinforcement in your parenting. This is an example of negative reinforcement because I’m stopping something bad (hanging onto the milk) in order to increase the likelihood of a behavior (saying “thank you”). Reinforcement can be so powerful in your parenting and is so much more enjoyable as a parenting technique compared to punishment. See if you can get better manners out of your kids this week by being an operant conditioning pro!

Understanding Reinforcement vs. Punishment

I simply cannot blog another week without covering reinforcement and punishment. These are terms you have probably heard before and you may have a thorough understanding of them or an inkling about what they mean, but they are critical to parenting so let’s spend some time refreshing your memory. The quick psychology history lesson is that these terms are part of B.F. Skinner’s operant conditioning theory that developed in reaction to John B. Watson’s classical conditioning theory (think Pavlov’s dogs) and spawned the behaviorism movement in psychology. We’ll talk more about Pavlov and Watson another time. We now know that both forms of behaviorism are valid in different applications, and from the later cognitive psychology movement we know that thoughts and attitudes clearly cannot be ignored as influences on behavior. For today, let’s focus just on how parents can use reinforcement and punishment to improve their parent-child interactions. To understand these concepts, it is simplest to look at them in a grid:

  Positive Negative
Reinforcement Adding something good

Ex: Praise

Removing something bad

Ex: Stopping a nagging song

Punishment Adding something bad

Ex: Cleaning house

Removing something good

Ex: Taking away a toy

Reinforcement comes in two types, positive and negative. Positive reinforcement means adding a pleasant stimulus (adding something good) to increase the likelihood of a behavior. If your child says, “May I have the milk please,” you hand the milk and say, “Here you go Johnny, and I love how you used your good manners.” Johnny is happy to receive your praise and is more likely to say please in the future. Other examples of positive reinforcement are clapping when a child puts their shoes on by themselves, scheduling a play date when they show good behavior at home, or getting a lollipop at the end of the grocery trip if they sat nicely in the cart.

Negative reinforcement means removing an aversive stimulus (taking away something bad) to increase the likelihood of a behavior. A couple of my kids have gone through phases where they would get in the car and take forever to put on their seatbelts. After trying a few different approaches, I found the one that worked best for my munchkins is singing the seatbelt song, “The first thing you do when you get in the car is strap in, strap in” and simply saying that over and over again (with increasing volume if necessary) until they strap in. If you have a good singing voice, this technique may not work for you, but when I say those first few words of the song, they usually get to buckling up quite quickly now. More examples of negative reinforcement are a child putting away their shoes so mom stops nagging them, letting a child walk on the sidewalk without holding your hand when they agree to stay right next to you, or switching the radio station after a child adds “please” to their request.

Punishment also comes in two types, positive and negative. The term positive can be very confusing in this context but remember positive just means to add something (good or bad). Positive punishment means adding an aversive stimulus (adding something bad) but this time the stimulus is meant to decrease the likelihood of a behavior. For example, if my boys get too rough with each other, they are put to work cleaning the house. First they must “check-in” with the victim. In our house this means a sincere apology in a nice tone of voice and an inquiry as to how the guilty party may help to make it up to the victim (take a turn with a coveted toy, clean up the toys they were playing with, etc.). But if mom has to get involved because the roughness continues, they are off to wipe down the walls, doors, or table chairs. A little physical labor helps to get out that extra energy and deter further roughness. Other examples of positive punishment are speaking to your child in a firm tone, giving a disapproving look when you see your child behaving poorly, or having to say 10 nice things for every unkind thing they say to someone.

Finally, negative punishment means removing a pleasant stimulus (taking away something good) to help decrease the likelihood of a behavior. When one of my boys was 7 years old and regressed into biting siblings when he was angry, we tried a few approaches and ultimately had to step things up a notch by making the rule that if he bit someone, that person could choose any one of his Lego sets to play with for an entire week. This happened once and there were tears, complaints that the punishment was too harsh, and sadness when the Lego set was dismantled by a younger sibling, but we felt that this was a very serious behavior that needed to be stopped. It tugged on my heartstrings to see him so upset, but guess how many times he has bitten a sibling since then…zero! More examples of negative punishment are leaving a play date early for poor behavior, skipping dessert if you don’t eat your dinner, or removing toys if you don’t clean up after playing.

So there you go, four powerful tools in your parenting tool belt to help motivate your child’s behavior to be more in line with your expectations. Amazing parents use reinforcement throughout the day, every day. Punishment should be used infrequently after first attempting other options. Parenting can be very frustrating at times, and even the best parents sometimes loose their cool and yell at their children, but your goal should be to avoid that form of positive punishment. Physical punishment, such as spanking, should never be used. Now that you have a better understanding of these terms, take a look at how you interacted with your children today and see if you can fit examples from your own life into these four categories. If you’re heavy on the punishment side, make it your goal to focus on reinforcement tomorrow!