Differential Reinforcement

If I want my partner to cook dinner more often, take out the trash regularly without asking, or make the first pot of coffee in the morning, what is the best way to accomplish this? I could flat out ask them, and for some people that may work. But what if your partner needs a little more (okay, more like a lot more) encouragement? Differential reinforcement to the rescue!

What Is Differential Reinforcement?

Differential Reinforcement (DR) is when I reinforce the behaviour I want to see, and do not provide reinforcement for the behaviours I do not want to see. Reinforcement is the act of providing a reinforcer (this is different for everyone and varies from situation to situation – for some it is a preferred snack, for others perhaps it is an activity, some still it can be verbal praise) to an individual after a behaviour, to increase the future probability of that behaviour happening again under similar circumstances.

Why the future probability? Because I cannot affect something that has already happened. The client has already emitted a behaviour, I cannot after-the-fact influence that behaviour.

Differential Reinforcement (DR) also means that I do not provide reinforcement under those conditions if the desired behaviour does not occur.

How To Use Differential Reinforcement:

When using a strategy like Differential Reinforcement, you are likely to use additional ABA methods as well. One technique I like to use with differential reinforcement is shaping (check back for our future blog post about Shaping).

If, for example, I was working on animal sounds with a vocal client who was scrolling through responses (giving multiple animal sounds in succession), I would only provide their reinforcer (a high five and verbal praise such as “you got it!”) if their response immediately followed the SD (within 3 seconds of being asked), and not giving other responses either before or after their response.

Example:

Therapist, SD: “what does a cow say?”
Client response: “quack, neigh, moo”
Do not reinforce
Error Correction and Transfer Trial (watch the YouTube video for another example)

Therapist, SD: “what does a cow say?”
Client response within 3 seconds: “moo”
Provide reinforcement: high five and “you got it! A cow does say ‘moo’!”

The Different Types Of Differential Reinforcement:

In true ABA fashion, we love our initialisms. The world of reinforcement procedures is no exception, here we already introduced you to DR (differential reinforcement), these are a few more in the DR family: DRA, DRD, DRH, DRI, DRL, DRO, and DRP.

DRA: Differential Reinforcement of Alternative Behaviour

Providing reinforcement for a predetermined alternative behaviour while withholding reinforcement for the unwanted response.

Using DRA requires that you identify the unwanted behaviour, as well as identifying appropriate alternative behaviours you would like to see in its place. This procedure is a great way to teach adaptive skills to clients.

Example:

In the classroom, you want to extinguish shouting out answers and differentially reinforce hand raising for a specific student who calls out answers for teacher attention.
Reinforcement in the form of teacher attention and praise is provided only when the student raises their hand and is called upon to answer the question.
Reinforcement is withheld when the student calls out the answer without being chosen (regardless if their hand is raised or not).

DRD: Differential Reinforcement of Diminishing Rates

Providing reinforcement dependent upon the number of responses meeting or being below a predetermined limit during a specified time frame.

DRD procedures are to be used when the behaviour is appropriate to engage in, but the goal is to see a specific number of responses at a time. This can be used to gradually reduce the number of cookies eaten per day.

Example:

You have a client who can finish a box of cookies in one sitting and want to gradually reduce the number of cookies eaten.
A single box of cookies has 30 cookies inside. Your initial goal may be for the client to eat 25 cookies or less in one day. If they eat 25 cookies or fewer in a day, they would receive reinforcement (an identified valuable reinforcer was doing a painting craft).
When this criterion is mastered, the next goal would be a diminished (lowered) rate: 20 cookies or less. This procedure repeats until it meets a predetermined diminished rate.

DRH: Differential Reinforcement of High Rates of Behaviour

Providing reinforcement dependent on the responses meeting or being above a predetermined limit during a specified time frame.

DRH procedures specify that rapid responding is reinforced, at gradually increasing intervals, until a predetermined mastery criterion is met. Imagine you are playing a competitive game of 5 Second Rule with family over the holidays, where the goal is to name 3 items in an unknown category in 5 seconds or less.

Example:

Before the big holiday gathering, you and your partner practice naming 3 items on common lists, in under 5 seconds.
At first your partner is only able to think of one item in 5 seconds – but you have 2 weeks before the party! After dinner you practice together daily, and once your partner has increased their responding and is able to list 2 items per category in 5 seconds or less they receive the best reinforcer of all – a sweet kiss right smack on their cheek!
The next goal for them to get that wonderful smooch is to get 3 items per category in 5 seconds or less.

DRI: Differential Reinforcement of Incompatible Behaviours

Not all DRA’s are DRI’s, but DRI’s are a type of DRA. This time the alternative behaviour is one that is incompatible with the unwanted behaviour – they cannot occur at the same time.

With our DRA example above, a student can sit at their desk with their hand up and still call out. If the example did not specify that they needed to remain quiet until called upon, and received reinforcement for calling out with their hand up, what did we really accomplish? Another DRI example includes saying “good game” to opposing teams after a game instead of spewing profanities. A person simply cannot say two things at the exact same time.

DRL: Differential Reinforcement of Low Rates of Behaviour

Providing reinforcement dependent on the response occurring after a specified time frame with no occurrences.

DRL procedures are perfect for when the response is appropriate to have in the learner’s repertoire, so you do not want to extinguish it. Imagine having a learner who repeats “hello [name]” every time they see their teacher. This is a behaviour we want to have in their repertoire; however, it is not appropriate if they are saying it every 5 minutes.

Example:

The student would only receive reinforcement if they went 5 minutes without repeating “hello”, then we gradually increase the time interval until it reaches an appropriate predetermined criterion (e.g., after 2 hours of not seeing the teacher).

DRO: Differential Reinforcement of Other Behaviours

Providing reinforcement for the non-occurrence of the unwanted behaviour.

DRO is similar to DRA in that reinforcement is withheld for the unwanted response, but since you cannot provide a reinforcer for a lack of response, another behaviour that is contextually appropriate is reinforced. This does not need to be a specific alternative response.

Check out our YouTube video for an example of how to use a DRO procedure.

DRP: Differential Reinforcement of Paced Responding

Providing reinforcement for behaviour occurring within a minimum and maximum time limit.

DRP is a procedure that is used when the timing of responses is imperative to success. Music teachers, dance choreographers, and others are familiar with the concept of DRP (even if they are not familiar with the label). This is when the timing of responses matter: whether it be chords played on an instrument, movements made by dancers, or number of basketball dribbles per step.

Watch our YouTube video on Differential Reinforcement for more info!

Leave a Reply

Your email address will not be published. Required fields are marked *

Captcha loading...