In Applied Behavior Analysis, interventions fall into a variety of categories. Antecedent interventions can be used to prevent the “trigger” that often leads to a behavior or response (see Antecedent Interventions: Complete Guide). Skill development interventions help teach skills and replacement behaviors. Here we discuss consequence interventions.
A consequence is anything immediately following a behavior in which we are interested. Often, the consequence makes the behavior more or less likely to happen in the future. Consequences occur frequently without intention or planning. A consequence intervention can be used to intentionally reinforce desired behaviors.
Consider some of your own behavior:
- What is the consequence of pouring coffee into a coffee cup? You have access to coffee.
- What is the consequence of going to work every day? You earn a paycheck.
Nearly everything you do has a consequence. Both of the above are examples of positive reinforcement (assuming coffee and money are both motivating for you), but there are other types of consequences as well.
The key to understanding consequence interventions is an awareness of the way different categories of consequences affect an individual’s behavior.
ContentsUnderstanding the Categories of Consequences The Difference Between Reinforcement and Punishment Is the Distinction Between Positive and Negative Reinforcement and Punishment Necessary? What’s better, punishment or reinforcement? Choosing Effective Consequence Interventions Effectively Reinforcing Target Behaviors Ethical Considerations of Reinforcement and Punishment The Distinction Between Positive Reinforcement and Bribery
As with any treatment, the strategies discussed here should only be used with the assent of the learner. Read our post Understanding Assent and Assent Withdrawal in ABA for more information on how and why to obtain assent.
Understanding the Categories of Consequences
ABA defines consequences by 2 different variables:
- Something added or taken away
- Behavior occurs more or less often in the future
Combining these variables in different ways provides us with 4 basic categories of consequences:
- Positive reinforcement
- Negative reinforcement
- Positive punishment
- Negative punishment
These terms create confusion, especially for people on the fringes of the field such as parents, speech pathologists, paraprofessionals or teachers. The chart below can help in clarifying each of the categories.
A stimulus (plural stimuli) is anything that evokes a response from an organism (in this case, a human being). For the sake of this discussion, the stimulus we refer to is often an activity, attention, or something tangible.
|Positive||The addition of a desired stimulus following a behavior that results in that behavior occurring more often (or with more intensity or for longer durations) in the future.||The addition of an undesired stimulus following a behavior that results in that behavior occurring less often (or with less intensity or for shorter durations) in the future.|
|Negative||The removal of an undesired stimulus following a behavior that results in that behavior occurring more often (or with more intensity or for longer durations) in the future.||The removal of a desired stimulus following a behavior that results in that behavior occurring less often (or with less intensity or for shorter durations) in the future.|
Understanding these categories of consequence will help you in providing effective interventions for your clients.Back to Top
The Difference Between Reinforcement and Punishment
Differentiating between reinforcement and punishment seems an easy thing to do, but it can be quite confusing. Determining whether a stimulus served as a reinforcer or a punisher requires looking at the frequency of the behavior…in the future.
A consequence is not considered a punishment unless the behavior actually happens less frequently (or with less intensity or over a shorter duration) in the future. The same is true for reinforcement. A stimulus is only a reinforcer if the behavior occurs more frequently in the future.
When we apply a consequence after a behavior, we hope for behavior change in the desired direction, but reality may find the opposite occurs. Sometimes what we expect to be a punisher ends up reinforcing the behavior.
Here’s an example:
Parents and teachers commonly use time out with the intent for it to serve as punishment following a maladaptive behavior; however, this intervention may reinforce escape maintained behaviors. In the following example, if Michael wants to avoid the math worksheet, at least for a little while, throwing his pencil worked for him:
|Presentation of math worksheet||Throws pencil||Put in time out|
The above example demonstrates negative reinforcement if the behavior (throwing pencils or other items) occurs more often in the future. The math worksheet was removed (at least for a little while) and the behavior increased. If the behavior occurs less frequently in the future, then time out served as a punishment.
Something can be considered to be positive reinforcement when a stimulus (i.e. toy, activity, attention, etc.) is added and the result on behavior is an increase in the frequency, duration, or intensity of that behavior in the future.
For example, you tell your client to sit at the table. He sits at the table and you immediately give him an M&M. The M&M is considered positive reinforcement, but only if the result is that your client sits more often when told to do so in the future. If the behavior does not occur more often in the future, then it would not be reinforcement. It’s possible that your client does not like M&Ms or they are not motivating for him.
As professionals, we often use positive reinforcement in the form of access to something tangible or social praise. In order for our efforts to be effective, we must pay attention to whether the target behavior is increasing, decreasing or remaining the same. If it’s not happening more frequently, what we are doing isn’t working and we need to make a change.
Negative reinforcement requires the removal of a stimulus that results in an increase in the behavior preceding it. The individual engages in a behavior that results in an object, activity, or sensory experience terminating or being removed. The individual finds the stimulus aversive thus generating a reinforcing effect upon removal.
Here’s an example: You are in a public restroom with your client and she asks to leave the restroom when the hand dryer comes on. You respond by taking her out of the restroom. If she asks to leave noisy situations more often in the future, then leaving the restroom (providing an escape from the aversive noise) is a reinforcer.
Negative reinforcement can be effective when teaching replacement behaviors for escape maintained behaviors such as leaving the bathroom. Teaching your client to ask for a break during work tasks or providing access to noise canceling headphones are common methods of negative reinforcement.
Positive punishment occurs when a stimulus (i.e. sitting in a “time out” chair, spanking, etc.) is added and the result is a decrease in the behavior in the future. As a general rule, ABA avoids the use of punishment unless positive strategies alone have been determined to be insufficient to elicit necessary change.
It is entirely possible that when attempting to implement positive reinforcement you inadvertently implement a positive punishment procedure. If you make assumptions around what you think should be reinforcing for your client rather than conducting a preference or reinforcer assessment, you are at risk of this occurring.
Negative punishment occurs when a stimulus (i.e. access to TV or driving the car, etc.) is removed and the result is a decrease in the behavior in the future. This type of punishment is common in parenting advice books and schools. The danger is that, as with positive punishment, there is a risk of inadvertently reinforcing the behavior you are attempting to reduce.
For example, schools often use suspension (removing access to school and the learning environment) as a form of punishment. When parents restrict privileges following an undesired behavior (such as staying out past curfew), they are attempting to use negative punishment.Back to Top
Is the Distinction Between Positive and Negative Reinforcement and Punishment Necessary?
Professionals in the field of Applied Behavior Analysis ABA) strive to disseminate the technology to outsiders. Often, professionals must teach and train parents, teachers and professionals from other fields to utilize strategies for behavior change. The seemingly endless terminology and acronyms creates a significant barrier to accomplishing this task.
Much of the terminology within ABA provides clarification that sets the technology apart from that offered in other fields. At the same time, some terms create more confusion than clarification. Newcomers to the field spend considerable time studying the terminology rather than the technology. The terms positive and negative reinforcement and punishment appear to be among the most confusing for outsiders. This raises the question, is the distinction really necessary?
The terms positive and negative reinforcement and positive and negative punishment create confusion when training new staff or collaterals. Many confuse positive punishment with negative reinforcement and frequently misuse the terminology. This can lead to ineffective implementation of of behavior change interventions.
Terminology that is unnecessarily complex or confusing alienates practitioners. Rather than being seen as the source for answers, the terminology makes professionals unapproachable.
Check out this example from the Big Bang Theory to find out how the terms negative reinforcement and positive punishment might be confused.
Take a look at these examples of positive and negative reinforcement and punishment:
Positive Reinforcement: Following your client stacking blocks, you offer praise and a short video, resulting in your client stacking blocks more frequently in the future.
Negative Reinforcement: Following your client asking you to turn off the music, you turn off the music, resulting in your client asking you to turn off the music more frequently in the future.
Positive Punishment: Following your client’s tantrum that included throwing toys, you require him to restore the environment, resulting in a decrease in the frequency of these tantrums in the future.
Negative Punishment: Following your client’s aggressive behavior, you restrict his access to recess, resulting in a decrease in the frequency of this behavior.
Does the Positive Exist Without the Negative?
I’m not trying to go all existential, but this is important in considering the necessity of the distinction between positive and negative reinforcement and punishment. As you consider these examples, consider how the distinction becomes questionable during application:
- Could you provide access to the praise and video if you hadn’t first restricted this access?
- Would you be able to you turn off the music at your client’s request if you didn’t first expose your client to the music?
- Could you require your client to clean up if you hadn’t first given him access to items to throw?
- Would you be able to restrict access to recess if you didn’t grant access first?
Why Continue with the Distinction Between Positive and Negative Reinforcement and Punishment?
Given the frequent confusion in the terminology and the reality that one doesn’t exist without the other, why do experts in the field continue with the distinction? As a science, it is our duty to question the rationale behind our technology.
According to Cooper, Heron and Heward (2007), we are bound by “philosophic doubt” to be skeptical. This means that as professionals in a scientific field we question what is currently accepted as fact, look for evidence that our assumptions are incorrect, continually seek to expand understanding, and assume the potential of new information to override existing beliefs.
Although professionals in ABA continue to use these terms, we should question if the need to distinguish between them truly exists. Would referring to a condition as either reinforcement or punishment provide sufficient context? Continue to question and challenge, using philosophic doubt to ensure the efficacy of our field.
It’s critical for the advancement of the field of Applied Behavior Analysis that we continue to evaluate why we do what we do. Professionals in the field must be intentional in their interventions, but must also carefully consider their use of terminology as well.
Jack Michael (1975) published Positive and Negative Reinforcement, a Distinction that is no Longer Necessary; or a Better Way to Talk About Bad Things in Behavior analysis: Research and application. Michael asked this question nearly 45 years ago yet there has been no change in the usage of the terminology.Back to Top
What’s better, punishment or reinforcement?
There isn’t one right answer that’s universally true. It’s up to you as the professional to understand the risks and benefits of the interventions you choose to implement.
Both reinforcement and punishment procedures present the opportunity for unintended effects on behavior. Professionals must plan for these possibilities prior to initiating interventions. You can mitigate some of these effects more easily than others and must carefully weigh the potential for harm against the potential benefits.
|Behavioral contrast-behavior receiving reinforcement in one setting decreases in other settings where reinforcement is unavailable||May result in a higher rate of aggression|
|May lead to an increase in undesired or a decrease in other desired behaviors that are part of the same response class||Punishment procedure may model inappropriate behavior to the individual|
|When access to a reinforcer is restricted, maladaptive behavior may escalate||Punishment effects may provide negative reinforcement for the interventionist increasing the possibility of abuse|
|Learner may rely on extrinsic reinforcement if the schedule of reinforcement is not thinned to naturally occurring contingencies||Usually only effective in the presence of the punisher|
|Effects of immediate reinforcement contingencies may reduce other more desired behaviors that result in delayed reinforcement||Risk of interventionist becoming a conditioned punisher|
|Reinforcement may lead to behavior that negatively impacts an individual’s health, safety, relationships, etc. (i.e. addiction)||Punishment does not establish an alternative behavior so another undesired behavior may take its place|
Example of Unintended Consequences of Reinforcement
Watch the video below for a common example of the use of reinforcement and identify any possible negative effects the parent might encounter as a result of using this intervention.
Although as a professional, you might recommend that Roy give Brad a reason to sit at the table. Do you see the potential for the following unintended consequences?
- Brad stops eating lunch at school where reinforcement for sitting at the table is unavailable.
- Brad begins to overeat, resulting in considerable rapid weight gain.
- When Roy begins to restrict the tablet, making it only available at mealtime, Brad begins to hit Roy, demanding that he have it when he wants it.
- Brad will only sit at the table if he has access to the tablet throughout the entire meal. As soon as the tablet is removed, he’s up and running again.
- Even though Brad’s teacher uses a reinforcement system for Brad’s task completion at school, his willingness to sit to do schoolwork goes down since that reinforcer is significantly more delayed than the reinforcer he earns for sitting to eat at home.
- Brad begins to find food so inherently reinforcing, especially foods high in carbohydrates, that he begins to demand more and more of these foods.
Knowing that the above unintended consequences are possible, would this change your recommendation to use reinforcement to encourage Brad to sit at the table to eat? Since it’s unlikely that there’s a simpler solution for Roy to implement at home, it’s probably still the best intervention for him to try. Identifying potential unintended effects allows you to plan for them.
The Problem with Punishment
One of the big drawbacks to punishment is the risk of inadvertently reinforcing the behavior you are trying to reduce. This can happen when the parent doesn’t understand the function (reason behind) the behavior.
For example, 4 year old Jill throws a toy at Mom when she is asked to clean up her toys. Mom puts Jill in the “time out” chair for 4 minutes (1 minute for every year of age). Jill repeatedly gets up from the chair causing Mom to bring her back to finish her “time out.” In the end, Mom sits in the chair and holds Jill in her lap until her “time out” is finished.
In this example, Jill’s mom didn’t do the work to determine the function of the behavior. She used the punishment she believed would be effective. But, if Jill threw the toy because she wanted to get Mom’s attention, then Jill’s mom actually reinforced this behavior. Every time Jill’s mom directed her back to the chair, and then when she held her on her lap, Jill got the attention she was seeking. It’s likely that Jill will continue to throw toys when asked to clean up because when she does this she knows she can get attention from Mom.
A better approach is to use reinforcement.
Reinforcement is the opposite of punishment. It makes a behavior more likely to happen in the future. Okay, you might be a little confused here. If you want to stop your learner from doing something (like throwing toys), then you have to use punishment, right? You don’t want the behavior to happen more often!
It’s reasonable to make this assumption. But when you’re trying to change a behavior you can look at it from two perspectives:
- You want to stop Jill from throwing her toys.
- You want Jill to pick up her toys when asked.
Focus on the behavior you want to see more of by using positive or negative reinforcement. Then provide no reinforcement for the behavior you’re trying to stop. While many parents and professionals might tell you to ignore a behavior – which withholds reinforcement in the form of attention – this isn’t always the best solution. Watch our video What to do instead of ignoring a behavior: The Matching Law to learn what to do when a behavior can’t be ignored.
Positive reinforcement happens when something is added, and the result is an increase in the behavior. This can be an object such as a toy, or an activity like watching a favorite TV show. It can also be in the form of social praise such as, “great job!” although often children with autism don’t find social praise as reinforcing as other children might.
It’s important to note that this type of reinforcement can be either intentional or inadvertent.
Let’s look at an example of unintentional positive reinforcement. Three-year-old Ben is out shopping with his dad at the grocery store. They made it through almost the whole shopping trip without any serious problems, but as they stand in line waiting to check out Ben grabs a candy bar from the shelf. Dad takes it away and tells him he’s not getting candy today. Then Dad watches in horror as Ben starts to wind up for a tantrum. He is getting agitated and has started to cry. Dad has seen this many times before and he knows the whole store will stop to watch when he really gets going. To avoid this, Dad shows Ben the candy and tells him he can have the candy after they pay for it.
In this case, the tantrum is more likely to happen in the future because Ben received the candy.
On the other hand, the intentional application positive reinforcement can have significant benefits. If instead Ben’s dad had anticipated a tantrum at the checkout he could have implemented positive reinforcement. Before Ben became agitated Dad tells Ben that he can have a piece of candy if he waits patiently while they pay for the groceries. If Ben succeeds and doesn’t have a tantrum, he gets the candy. If he has a tantrum then Dad puts the candy back and they try again another day.
Let’s look at some more examples of positive reinforcement.
Positive Reinforcement Examples
Jack is 6-years-old and his parents have decided it’s time he learn to sit with the family at dinnertime. Normally he grabs a bite of food passing by the table, but they feel he’s old enough to stay in a chair while they all eat together. To do this, his parents have decided to use positive reinforcement. They will let him watch 3 minutes of his favorite YouTube video after sitting for 3 minutes at the table.
Five-year-old Clara is afraid of dogs. Their neighbor has just brought home a golden retriever puppy and Clara runs crying whenever the puppy comes near her. Her parents want to teach her that the puppy is not going to hurt her, so they have decided to use positive reinforcement. They begin by having Clara watch the puppy out the window. When she does this, she gets an M&M. Over time they will gradually bring the puppy closer to Clara, but they might need to choose a stronger reinforcer to help Clara overcome her fear.
Choosing a Reinforcer
Keep in mind that if the reinforcer you choose isn’t something your learner finds reinforcing you might actually be using positive punishment. For example, Mom is trying to teach 5-year-old Julie to put on her shoes. To begin, Mom has decided that every time Julie picks up her shoes Mom will clap excitedly and tell Julie, “good job!” Mom doesn’t understand why Julie now picks up her shoes less often than she did in the beginning.
The problem in the scenario above is that not only did Julie not find the loud clapping and exclamation of “good job!” reinforcing, she the sudden noise scared her, causing her to be afraid to pick up her shoes.
Negative reinforcement makes a behavior more likely to happen in the future, but it does this by removing something. Like positive reinforcement, this can be applied unintentionally, causing undesired behaviors to be reinforced.
For example, 7-year-old Beth is afraid of the loud hand dryers found in public bathrooms. When she hears one she begins rocking, humming and puts her hands over her ears. If the sound continues she begins to scream as well. It has gotten so bad that her parents have stopped taking her into public restrooms all together.
This inadvertently reinforces Beth’s agitated behavior.
Let’s say that instead Beth’s parents decide to intentionally use negative reinforcement. They bring noise canceling headphones that Beth can use to drown out the sound of the hand dryer. When the hand dryer starts they show Beth how to put on the headphones to remove the sound that she finds so distasteful.
Negative reinforcement works extremely well for replacing an undesired behavior (Beth’s agitation) with a more desirable behavior (using headphones to drown out the sound of the dryer). Let’s look at some more examples.
Automatic or Socially-Mediated Negative Reinforcement
Some behaviors result in direct reinforcement. For example, if you cover your ears when you hear a loud noise, that behavior is reinforced automatically as it results in an alleviation of the noise. The noise isn’t as loud when you cover your ears, so you are more likely to cover your ears in the future when you hear a loud noise, assuming that you don’t like loud noises.
Other behaviors require the mediation of another person. If I’m sitting in the backseat of a car and the music on the radio bothers me, I need to ask someone in the front seat to turn the music off. I can’t reach it on my own and need the help of someone else to make the music stop. If the person in the front seat turns the music off, then my behavior is reinforced through socially-mediated negative reinforcement.
Both of the above examples involve adaptive behaviors (covering ears and making a request); however, maladaptive behaviors are often reinforced in the same way, although unintentionally.
Consider these examples:
If a child screams when he hears a loud noise and you respond by giving him headphones or removing him from the loud noise, you not only alleviate his discomfort, you negatively reinforce the screaming.
If a child kicks the back of the seat because the music on the car radio bothers her and you respond by turning off the music, you not only ease her frustration, you negatively reinforce the kicking.
Applying Negative Reinforcement
Negative reinforcement provides an effective method for reinforcing a variety of adaptive behaviors, including both automatically and socially mediated forms of reinforcement. When a child demonstrates an aversion to a particular item, task, activity or sensory experience, consider using negative reinforcement to teach an adaptive way for that child to escape or avoid that stimulus.
We all engage in a variety of behaviors to avoid things we don’t like. You use mostly socially appropriate behaviors to accomplish this. The children you work with may creatively engage in less adaptive behaviors that work for them. Guide the child in choosing a more appropriate way to get what he wants.
Negative Reinforcement Examples
Nine-year-old Liam often gets frustrated when doing his homework, especially math. When he gets stuck on a problem he will rip his homework page to shreds. Because of this, the problems he does solve can’t be graded. His parents have decided to use negative reinforcement to teach Liam to ask for help. When Liam asks for help one of his parents will sit with him and explain the math problem, making it easier for him to understand.
Andrea is 7-years-old and getting her to clean her room feels like a constant battle. When Mom asks her to clean up her room she ends up taking more toys out instead, making the room an even bigger mess. Mom realizes that cleaning the whole room might be too big a task for Andrea to do all at once, so she has decided to use negative reinforcement to make the project easier. Instead of telling her, “clean up your room,” like she normally does, Mom tells her, “pick up all the books you can find and put them on the shelf.” By doing this she has made the task easier for Andrea, and she is able to complete it without argument.Back to Top
Choosing Effective Consequence Interventions
As I mentioned in the beginning, almost everything we do has a consequence. Consequences can be unintentional or planned. For the purpose of the rest of this article we will focus on planned consequences that are intended to change the behavior of a child.
Often when we apply artificial consequences, we make assumptions that what we do will have a reinforcing or punishing effect on behavior. We may become frustrated when we offer a child access to something we believe should be reinforcing, yet it doesn’t produce the outcome we were hoping for.
The same is true when we apply consequences we intend to be punishing. Think about schools applying a consequence of “suspensions” for inappropriate behavior. For some children, this can have a punishing effect, but for others, it may be highly reinforcing, especially if the function of the behavior that caused the suspension is to escape from school or classwork.
Differential reinforcement is a powerful consequent strategy that can be applied in a variety of ways. When you use differential reinforcement, you typically reinforce one behavior while putting another behavior on extinction. Our post Differential Reinforcement: a Complete Guide goes into more detail than what you’ll find here.
Differential reinforcement of other behavior (DRO)
Differential reinforcement of other behavior (DRO) provides reinforcement in the absence of target behavior. This time-based system offers reinforcement on a fixed or variable interval schedule when the specified behavior does not occur.
For example, you want to eliminate a child’s spitting. In baseline, the behavior occurred about every 2 1/2 minutes. You decide to use DRO on a FI2 schedule to reinforce the absence of spitting. You set a timer for 2 minutes and for every 2 minutes during which spitting does not occur, the child earns a reinforcer. If the child spits during the 2 minute interval, the timer is reset to 0 and the interval begins again. The child earns reinforcement only after 2 consecutive minutes of no spitting.
Differential reinforcement of alternative behavior (DRA)
Differential reinforcement of alternative behavior (DRA) offers reinforcement for appropriate alternative behavior to the target behavior. For this intervention, choose an alternative behavior you wish to reinforce. This alternative behavior should serve the same function (access, escape, automatic) as the challenging behavior.
This intervention provides reinforcement on a fixed or variable ratio schedule.
You could use DRA to reduce a variety of challenging behavior while teaching an appropriate alternative behavior. For example, if your client jumps up and down repeatedly, making loud noises upon impact which is disruptive to others, reinforce jumping on a small trampoline instead. It’s not enough to just direct him to engage in the alternative behavior, you must also provide reinforcement for it!
Differential reinforcement of incompatible behavior (DRI)
Differential reinforcement of incompatible behavior (DRI) provides reinforcement for a behavior that is incompatible with the target behavior. Similar to DRA, for DRI, choose what incompatible behavior you wish to reinforce. This behavior again should be functionally equivalent to the challenging behavior. DRI also provides reinforcement on a fixed or variable ratio schedule.
When choosing an incompatible behavior, consider one that produces the same result for the child while also ensuring that the child can’t perform both the target behavior and the incompatible behavior at the same time.
Let’s look at an example. You’re working with a child who constantly taps his pencil on the desk when completing worksheets. You decide to use DRI to reduce this behavior. To do so, you decide to reinforce writing since he cannot write and tap his pencil at the same time.
Differential reinforcement of lower rates of behavior (DRL)
Differential reinforcement of lower rates of behavior (DRL) reduces a behavior without eliminating that behavior. This intervention is ideal for behaviors that are simply occurring too frequently. For this intervention, you choose your interval (time period) and the maximum number of occurrences of the behavior that will earn reinforcement.
Here’s an example: your client raises her hand to respond every time the teacher asks a question. She becomes upset when she’s not called on, but she’s not allowing the other children in the class to respond to the teacher’s questions. You want to reduce her rate of hand raising without eliminating the behavior. You decide that a more reasonable rate of responding would be for her to raise her hand no more than 5 times during a 10 minute period. If she raises her hand 5 or fewer times during that 10 minute period she earns reinforcement. If she responds more than 5 times then reinforcement is withheld for that period and the next interval will begin.
When using DRL, you can shape lower rates of responding by gradually changing the reinforcement criteria. If in the example above, your client currently raises her hand 25 times in a 10 minute period, you might begin your criteria at 20 times in a 10 minute period and gradually change the criteria as she becomes successful.
Differential reinforcement of higher rates of behavior (DRH)
Differential reinforcement of higher rates of behavior (DRH) reinforces a higher rate of responding. Use this intervention when your client is demonstrating a skill but not at a high enough rate.
For example, your client completes simple math problems at a rate of 1 per minute. While you’re pleased that he completes his work, with 100 problems on the paper, he will not complete the worksheet within the designated math time. DRH provides reinforcement for completing work at a faster rate. You decide to shape faster responding by providing reinforcement if he completes at least 2 problems in a minute. Over time, you gradually increase the criteria required to achieve reinforcement.
Other Differential Reinforcement (DR) Strategies
In some instances, you may differentially reinforce a more desired behavior with a more potent reinforcer and a slightly less desirable behavior with a slightly less desired reinforcer as part of a plan to shape behavior. For example, when trying to teach a child to echo the word “mama,” you might choose to give the child one car when he says “ma” and 10 cars when he says “mama,” assuming that more cars are more reinforcing for the child.
This strategy works well for toilet training as well. You can differentially reinforce sitting on the toilet with a lower preference item and producing (urinating or deficating) on the toilet with a high preference item.
There’s a lot to know about differential reinforcement, so we’ve created a complete guide that’s available in a printable PDF.
Reinforcement is provided contingent on the occurrence of a specified behavior. For many children who are new to ABA, an initially dense schedule of reinforcement helps the child understand the relationship between doing what is asked and achieving reinforcement. Initially, reinforcement may be provided on a FR1 schedule and thinned over time.
Token economy visually presents progress toward earning reinforcement. Through a pairing procedure, tokens (stickers, velcro images, coins, etc.) become conditioned reinforcers which are then traded in for a backup reinforcer. This system works well for teaching delayed reinforcement and self-management strategies.
You can use a token board, calendar, dry erase board, sticker chart, paper, or any other method of displaying progress toward the child’s goal. In the image below, there is a spot for a picture of what the child is working for and room for 4 tokens. To make this more interesting and potentially more motivating, use images of the child’s favorite characters as tokens and make the board visually appealing.
To learn more about this strategy read our post Token Economy: Examples and Applications in ABA
Extinction reduces challenging behaviors by withholding reinforcement. Although this can be an effective strategy, it’s not always a necessary component of a consequent package. Putting a behavior on extinction presents many potential disadvantages and should be implemented with care.
Often with the application of extinction procedures, the behavior escalates prior to reducing to low levels. For some children, this is an acceptable risk. If the child already demonstrates dangerous behavior, using this strategy may not be worth the potential risk involved. For comparatively innocuous behaviors, such as whining or verbal protesting, extinction may be a viable intervention.
Using differential reinforcement or other strategies without including extinction may produce similar results without the potential for an extinction burst.
Restitutional overcorrection requires the individual to restore or repair a damaged space or situation as a consequence for undesired behavior. For example, a child who throws toys when angry may be required to clean up not only the toys thrown, but all other objects out of place in the room.
For many children, this may have a punishing effect, although this won’t be the case for every individual. Keep in mind that this consequence is only a punishment if it results in a decrease in the behavior in the future. It’s possible that the attention required to ensure compliance with restoring the environment could have reinforcing effects. Always be conscious of the potential that an intended punishment may serve as a reinforcer when implemented.
Positive Practice Overcorrection
Positive practice overcorrection, similar to restitutional overcorrection, requires the individual to engage in sustained effort following an undesired behavior with the intent of having a punishing effect. This intervention requires the individual engage in a desired alternative of the challenging behavior.
For example, a child who frequently slams the door when walking into the house may be required to “practice” closing the door quietly 10 times each time he slams the door. For many children, this repeated effort has a punishing effect. Just as with restitional overcorrection, be cautious of the potential for unintentionally reinforcing the behavior you attempt to reduce.
In ABA, time out means time when reinforcement is unavailable. While it may look similar to the traditional format for time out in common parenting practice, this is not necessarily the case.
Time out is most effective when a dense schedule of reinforcement is available in the absence of the challenging behavior. When you provide a high rate of reinforcement at all times, and that reinforcement stops, the child quickly understands the consequences of his behavior.
As with all other forms of punishment, time out has the potential to have undesired outcomes. Children’s behavior may escalate when they are denied access to a desired reinforcer. For some children, this can be potentially dangerous.
Using Consequences to Change Behavior
Carefully planning consequences leads to positive behavior change. Use reinforcement alone whenever possible and use punishment with extreme caution and supervision. Carefully consider the risks and benefits of using any punishment procedure.
When implementing consequent procedures, ensure you consider the child as an individual. What reinforces behavior for one individual may punish behavior for another.
As with any treatment, the strategies discussed here should only be used with the assent of the learner. Read our post Understanding Assent and Assent Withdrawal in ABA for more information on how and why to obtain assent.Back to Top
Effectively Reinforcing Target Behaviors
In order for reinforcement to be effective, there are several factors to keep in mind:
- Ensure the reinforcer you choose is of high value to the child
- Deliver reinforcement immediately after the behavior you’re interested in (within 3 seconds) until you begin to systematically teach delayed reinforcement
- Reinforce on a consistent schedule
Remember a reinforcer is only a reinforcer if the result is an increase in the behavior you’re interested in. Food and videos don’t motivate all children. A little creativity promotes success when choosing potential reinforcers.
Many children satiate quickly, especially with a dense schedule of reinforcement. Conduct frequent preference assessments to ensure that your client remains motivated.
Choose valuable reinforcers
For many adults and typically developing children, social praise can serve as a source of reinforcement. If your boss came up to you and told you what an amazing job you were doing, and did so in a sincere way, you would probably be more likely to continue to work hard. Many children enjoy hearing that they have done something well. However, for children with autism, social praise will likely have a lot less meaning or value, at least initially.
Anything your learner enjoys can be used as a reinforcer. This includes toys, games, tickles, hugs, music, or other activities she prefers. In addition, small snacks or treats (edibles) can be effective reinforcers. Use caution if your learner has problems with eating or weight; however, even in these cases, it may be possible to use edibles for brief periods.
For reinforcement to be effective, the reinforcer must have value to your learner. If your learner loves the iPad, but can use the iPad whenever he wants, then it will have no value as a reinforcer. It’s essential to limit access to anything you intend to use as a reinforcer so that it continues to have value for your learner.
If you had a money tree in the back yard, even if you loved your job would you still be motivated to go to work? You don’t need the paycheck if you can go into the backyard and pick money off the tree. The same is true for your learner. If he can pick up the iPad whenever he wants it, what is his motivation to do anything you ask of him?
Looking for Reinforcers
When selecting reinforcers consider how your learner responds. Remember it is of utmost importance that the reinforcer you use be something your learner considers reinforcing.
- How enthusiastic are you when delivering reinforcement to your learner? Some children respond well to loud praise and a quick tickle when you give reinforcement. Other children respond better to calm, quiet praise and a soft rub on the back. What works best for your learner?
- Change reinforcers often. Children satiate on reinforcers that are used often and they begin to lose their effectiveness. Rotate between some of your learner’s favorites then be on the lookout for things your learner shows interest in. Anything that your learner is excited about can be used as a reinforcer.
- Pairing reinforcement with social praise will, over time, teach your learner to value social praise. Always combine any tangible reinforcement with social praise: a verbal comment, tickles, a pat on the back, a smile, etc.
For more on selecting reinforcers, read our post Choosing Reinforcers: Reinforcer Assessments or Preference Assessments.
Should You Use Edibles?
This question has no easy answer and will depend entirely on your situation and your learner’s needs. Edibles offer a solution when your learner has little to no interest in other possibilities as reinforcers.
Use edibles if:
- You have a plan for increasing other possible reinforcers for your learner
- You plan to use them short-term to teach a particularly difficult skill
- Your learner easily satiates on all other forms of reinforcement
- Your learner continues to eat a variety of nutritious foods
However, use these edibles with caution.
Avoid using edibles if:
- Your learner has a restricted diet consisting of fewer than 10 foods
- You have concerns about your learner’s nutritional needs
- Your learner engages in dangerous behavior to gain access to food
- You’re not sure how to use edibles to create new reinforcers for your learner
- You’re not sure which edibles to use
When using a food reinforcer, it is critical that the portions you provide are tiny. For example, use 1 sip of juice, 1 mini M&M, ¼ of a Starburst, 1 baby Goldfish, 1 slice of a strawberry. Children can become satiated on food reinforcers, making them ineffective. Keeping the portions tiny can slow this process down. Small portions also reduces nutritional concerns for your learner.
Using Reinforcement Effectively in ABA
Using reinforcement effectively is the single most important thing you can learn do to teach new skills and change behavior. While the three points listed above are of primary importance, there are other things to think about:
- Some children will stop positive behaviors when they are pointed out. If your learner does this, use shaping to help your learner adapt to and learn to accept reinforcement for those positive behaviors. Continue to provide reinforcement for those behaviors in a quiet, casual way until your learner becomes comfortable with you recognizing those behaviors.
- The amount of reinforcement should match how hard your learner perceives the task to be. There will be things that your learner finds especially difficult or unpleasant. Set aside your learner’s most preferred items for these tasks.
- If it seems as though reinforcement is not effective, take a look at what you are using for reinforcers. It may be time for a change.
For more on using reinforcement with children with autism watch ABA Autism Training – Chapter 2 – Reinforcement:
Schedules of Reinforcement
Schedules of reinforcement determine how often and how predictably reinforcement occurs. Each of the 4 different types of schedules provides different advantages and disadvantages. Understanding these will help you choose the most appropriate schedule for your client.
Schedules of reinforcement fall on a continuum with extinction at one end and continuous reinforcement (CRF) at the other. Extinction provides no reinforcement for a behavior while CRF provides reinforcement for every occurrence of a behavior. Along the continuum are the intermittent schedules of reinforcement.
There are 2 categories of schedules to describe the delivery of reinforcement.
- Ratio schedules=reinforcement after a specified number of occurrences of behavior
- Interval schedules=reinforcement after a specified time period
Each of these types of schedules can be implemented on a fixed or variable schedule, leading to 4 different types of schedules of reinforcement:
- Fixed ratio
- Variable ratio
- Fixed interval
- Variable interval
The 4 schedules of reinforcement offer a method of prescribing the frequency and predictability for staff to provide reinforcement for a particular behavior. Clearly identifying the appropriate schedule helps to ensure the best outcomes for your clients.
Reinforcement after a predetermined number of responses
Reinforcement after a predetermined average number of responses
Reinforcement after the first response after a predetermined amount of time
Reinforcement after the first response after a predetermined average amount of time
The prescription is written to include the number of responses or the amount of time with the abbreviation of the schedule type. For example a FR 1 schedule of reinforcement indicates a continuous reinforcement (CRF) condition. A FR 2 schedule provides reinforcement after 2 responses. A VI 5 min schedule provides reinforcement for the first response after an average of 5 minutes has passed.
Fixed ratio schedule
A fixed ratio (FR) schedule provides reinforcement after a fixed number of occurrences of the behavior. For example, if you’re teaching your client to tact cat and every time he sees a picture of a cat you give him a reinforcer, then you are using a FR1 schedule of reinforcement. The number indicates the specified number of responses required to achieve reinforcement. The smaller the number, the denser the schedule. If you have a FR2 schedule, you might thin the schedule by moving to a FR3 then a FR4 schedule over time.
Advantages of FR Schedules
- Often produce a high rate of responding
- Fast responding results in faster delivery of reinforcement
- May be able to influence the rate of reinforcement by adjusting the ratio requirement
- There is often very little hesitation when responding
- Easy to implement consistently
Disadvantages of FR Schedules
- Produces a post-reinforcement pause
Variable ratio schedule
A variable ratio (VR) schedule makes reinforcement available after an average number of occurrences of the behavior. A VR schedule provides the most resilient form of reinforcement as the child can’t predict exactly when he will earn reinforcement. Consider how motivated people are by a slot machine, even when they lose more money than they win. People continue to drop coins in the slot knowing that eventually they will win something. This is a variable ratio schedule of reinforcement at work.
Advantages of VR Schedules
- Produce steady rates of responding
- Tends to influence quick rates of responding
- There is typically no post-reinforcement pause
- Often larger ratios produce more rapid responding, to a degree
Disadvantages of VR Schedules
- More difficult to implement in a natural setting in a planned way
- Requires careful planning and some creative thinking to implement
Fixed interval schedule
A fixed interval (FI) schedule provides reinforcement after a preset amount of time. Interval schedules reinforce the first behavior that occurs after the specified time period. This reinforcement schedule creates a slow rate of responding at the beginning of the interval with higher rates of responding at the end when the individual begins to anticipate the pending reinforcement.
You might use a fixed interval schedule to reinforce social interactions in a busy after school program where it’s not possible to keep track of every occurrence of the behavior. This reinforcement schedule is comparatively easy to implement when it’s impossible or impractical to use a ratio schedule.
Advantages of FI Schedules
- Easy to implement consistently
- Allows for time when the client works independently as it does not require staff to watch the child’s responding during the interval
Disadvantages of FI Schedules
- Produces a post-reinforcement pause followed by slow responding and ultimately more rapid responding toward the end of the interval
- Children learn to predict the amount of time required to achieve reinforcement and they learn that responses immediately after reinforcement are never reinforced
Variable interval schedule
A variable interval (VI) schedule provides reinforcement after a variable amount of time around a specified average. Similar to a fixed interval schedule, the variable interval schedule offers reinforcement for the first response after a designated period of time; however, this time varies around a specified average amount of time.
For example: you’re working with a child in a school and you want to reinforce completion of math problems, but you’re responsible for many children in the class and you can’t keep track of each problem each child does. You decide to use a VI5 schedule and provide reinforcement to each child on average every 5 minutes. You might reinforce the completion of a problem after 3 minutes, 5 minutes, 7 minutes, etc. The intervals are inconsistent, but average 5 minutes.
Advantages of VI Schedules
- Produces a stable, steady rate of responding
- Does not produce a post-reinforcement pause
- Allows for time when the client to work independently as it does not require staff to watch the child’s responding during the interval
Disadvantages of VI Schedules
- Requires careful planning and implementation
- Produces low to moderate rates of responding
Choosing a Schedule of Reinforcement
Now that you know the advantages and disadvantages of each type of schedule of reinforcement, you’re equipped to make a decision about which is right for your client’s behavior. When choosing a schedule, consider:
- The skill of the staff implementing the intervention
- The desired rate of responding
- The need for consistency in responding
While a VR schedule of reinforcement offers the most consistent and highest rate of responding, if your staff can’t implement it correctly, you may be better off choosing a FR schedule and accepting the post-reinforcement pause. If you plan to teach a skill where fluency and a consistent rate of responding are a high priority, focus on training staff to appropriately implement a VR schedule.
If your client must work independently for a period of time, a FI or VI schedule might be most appropriate; however, you will need to accept a lower rate of responding.
Taking into account the advantages and disadvantages of each schedule of reinforcement helps guide your decision-making to ensure the best outcomes for your clients.Back to Top
Ethical Considerations of Reinforcement and Punishment
Positive reinforcement is often seen as the epitome of ethical practice in ABA. Similarly, punishment carries with it a stigma associated with control and abuse of power. In reality, these conditions occur naturally and neither are inherently good or bad. The distinction lies in the way these contingencies are applied by professionals.
Despite this fact, governing bodies insist on the implementation of reinforcement strategies before considering punishment procedures. In fact, the BACB’s Professional and Ethical Compliance Code for Behavior Analysts states:
4.08 Considerations Regarding Punishment Procedures
(a) Behavior analysts recommend reinforcement rather than punishment whenever possible.
(b) If punishment procedures are necessary, behavior analysts always include reinforcement procedures for alternative behavior in the behavior-change program.
(c) Before implementing punishment-based procedures, behavior analysts ensure that appropriate steps have been taken to implement reinforcement-based procedures unless the severity or dangerousness of the behavior necessitates immediate use of aversive procedures.
(d) Behavior analysts ensure that aversive procedures are accompanied by an increased level of training, supervision, and oversight. Behavior analysts must evaluate the effectiveness of aversive procedures in a timely manner and modify the behavior-change program if it is ineffective. Behavior analysts always include a plan to discontinue the use of aversive procedures when no longer needed.
Misconceptions about the Ethics of Reinforcement and Punishment
This emphasis on reinforcement over punishment is intended to ensure a focus on teaching new skills, encouraging growth, taking a strength-based approach and avoiding power struggles or an abuse of power. This drive toward reinforcement procedures is an important one and not to be taken lightly, but it’s equally important to understand that reinforcement in and of itself is not good and there are ways to misuse it. Consider applying reinforcement to a behavior that is harmful to the individual. A reinforcement procedure is not necessarily ethical.
The same holds true for punishment procedures. Punishment, remember, is the addition or removal of a stimulus that reduces the future occurrence of that behavior. A child who asks to go home from school at 9:00 every morning and is denied this request repeatedly is likely to eventually reduce the frequency of this request. Refusing the child’s request is punishment if it reduces that requesting behavior. From an objective perspective, honoring that request might be the unethical thing to do.
In this video, Dr. Chris Manente discusses some common misconceptions about the ethics of reinforcement and punishment.
Reinforcement and punishment are often more complex than they initially appear. Many factors impact their effectiveness and even their ethical application. Carefully consider all potential risks and benefits associated with the use of these contingencies and become more aware of their presence in the natural environment.Back to Top
The Distinction Between Positive Reinforcement and Bribery
Many parents I talk to say “isn’t reinforcement just another word for bribery?”
In fact, reinforcement is the opposite of bribery in one important way. Bribery is used to stop a behavior. Reinforcement makes a behavior more likely to appear again. Bribery also results in the unintentional reinforcement of undesired behaviors. Helping parents understand the distinction not only reduces their resistance to using reinforcement, but also prevents reinforcing the wrong behaviors.
The infographic below helps explain the distinctions between the two and can be a helpful resource for parents.
Positive reinforcement requires forethought and planning. When rewards are introduced during a behavior, the “reinforcer” becomes a bribe. Even if you know you made a mistake and should have presented a reinforcer before attempting to evoke a behavior, you can’t present a reinforcer once challenging behavior emerges. Regardless of your intentions, presenting a reinforcer after challenging behavior occurs is bribery. Bribery builds behavior chains that are difficult to break. Take a look at this example:
Jenny is a 5-year-old diagnosed with autism who attends a public kindergarten. She typically transitions well between different activities, but every once in a while she flops to the floor when told to transition. The behavior appears random in that it occurs for some transitions but not others and it’s not always the same transition. You decide to observe her paraprofessional to determine if you can spot the variables impacting this behavior.
During the observation, Jenny transitions smoothly through the first 4 major transitions of her day. Her paraprofessional praises her behavior and gives her access to some fidget toys when she gets where they are going. When her paraprofessional tells her it’s time to go to the rug for story time, Jenny flops to the floor. Her paraprofessional reaches into her bag and pulls out a special fidget toy that she has been saving for those really tough transitions. She shows it to Jenny and says, “first go to the rug, then fidget.” Jenny stands up and walks to the rug.
Her paraprofessional doesn’t understand why Jenny continues to flop. She makes sure that Jenny earns a fidget toy after every transition. Maybe Jenny isn’t sleeping well at night. Jenny may be using her behavior to tell her paraprofessional that she’s tired of the other fidget toys and needs something new. Help Jenny’s paraprofessional understand the importance of varying the reinforcer and offering a choice of reinforcers before starting the transition.
Parent and Caregiver Training
There are no courses required to become a parent. Most parents use many of the strategies their own parents used.; however, when raising a child with autism, these strategies are insufficient. Parents need guidance to help them understand their child’s unique needs.
Often parents and caregivers of children with autism believe they use positive reinforcement and many feel that it doesn’t work. They may have heard the term and know that they need to provide a reward for good behavior; however, few really understand positive reinforcement. They rely on you to guide them.
Parent and caregiver training should include didactic (classroom style learning) as well as in-vivo (hands-on) learning. Through the training, parents and caregivers should have opportunities to apply what they learn and receive feedback in real time. While the concepts aren’t difficult, they also aren’t intuitive. These aren’t their parents’ parenting techniques and they need time to practice.
Model effective use of interventions. Provide in-vivo examples of when to provide reinforcement and when to withhold it. Guide the parents and caregivers in planning for reinforcement and avoiding the bribery trap.
Using Examples of Positive Reinforcement
Adults often retain information better through storytelling to help make connections between what they know and what they learn. Consider different ways parents and caregivers come into contact with reinforcement themselves.
If they work, their paycheck serves as a reinforcer. There’s a limit to what they would be willing to do if their paycheck didn’t come when it was supposed to. If they suddenly found a money tree growing in their backyard, the paycheck would be much less reinforcing.
Tie this piece of information to positive reinforcement for their child. Their child needs to predict when he will earn reinforcement. Their child can’t have unlimited access to what they want to use as a reinforcer or the value of that reinforcer diminishes.
One of my favorite examples of positive reinforcement is from the Big Bang Theory. I often share this video clip with parents to help them understand reinforcement in a less serious way.
The Temporal Relationship of Positive Reinforcement
Many parents and caregivers tell you that they “give rewards” or “give reinforcement” for positive behaviors, but many struggle with understanding the temporal relationship that exists. In order for positive reinforcement to effectively impact behavior it must immediately follow that behavior (until the child learns to understand delayed reinforcement). This means that the reinforcer needs to be presented within 3 seconds of the behavior.
Parents and caregivers who are desperate for impactful behavior change may try for goals that are too broad or vague. Out of routine, they offer rewards such as:
- If you’re good at school you can have ice cream when you get home
- If you don’t scream in the grocery store, we can go to the playground when we are done (meaning after they take the groceries home and put them away and then deal with other things as they come up)
Using delayed reinforcement opens the door to reinforcing behaviors unintentionally. Consider the examples above:
- What happens if the child does well at school, but has a tantrum when walking in the door when he comes home?
- What should a parent do if the child starts screaming and demanding the playground while the parent tries to put the groceries away?
Parents are comfortable with these types of rewards because they probably received these rewards themselves as children. Guide parents toward offering smaller reinforcement more frequently for very specific behaviors. Using positive reinforcement in this way achieves results which parents find reinforcing. Help them experience success and they will continue to apply what they learn.
References and Related Reading
Behavior Analyst Certification Board. (2014). Professional and ethical compliance code for behavior analysts. Littleton, CO: Author.
Cooper, J. O., Heron, T. E., & Heward, W. L. (2007). Applied behavior analysis, p. 6-7.
Fisher, W., Piazza, C., Cataldo, M., Harrell, R., Jefferson, G., & Conner, R. (1993). Functional communication training with and without extinction and punishment. Journal of Applied Behavior Analysis, 26(1), 23-36.
Hanley, G. P., Piazza, C. C., Fisher, W. W., & Maglieri, K. A. (2005). On the effectiveness of and preference for punishment and extinction components of function‐based interventions. Journal of Applied Behavior Analysis, 38(1), 51-65.
Michael, J. (1975). Positive and negative reinforcement, a distinction that is no longer necessary; or a better way to talk about bad things. Behaviorism, 3(1), 33-44.
Perone M. (2003). Negative effects of positive reinforcement. The Behavior analyst, 26(1), 1–14. doi:10.1007/bf03392064
Timberlake, W., & Farmer-Dougan, V. A. (1991). Reinforcement in applied settings: figuring out ahead of time what will work. Psychological Bulletin, 110(3), 379.