Reinforcement: Dos and Don'ts

No doubt you use your fair share of reinforcement in your everyday life. For sure you use it with your kiddos, with your spouse, and with yourself. Reinforcement  is everywhere and whether or not you are trying to reinforce anyone's behavior, you definitely do throughout your daily life. So since this is such a big part of your existence, maybe we should break down what it is, and the dos and don'ts of reinforcing behaviors.

What is reinforcement?

Seems like a simple enough question that should have a simple enough answer. It doesn't. It's probably not what you think, at least not all of it. Essentially, reinforcement is the process of increasing the future frequency of behavior. So, it may be providing a Skittle for using the toilet. But not necessarily.

In the above example, the act of providing the skittle is just a behavior itself: your behavior of handing a Skittle to your child. It only acts as reinforcement if your child's toileting behavior increases in the future; that is, he/she continues using the toilet, particularly in exchange for Skittles. You can then argue that Skittles are reinforcing for your child for that behavior. 

Dos and Don'ts of Reinforcement

Do clearly define the behavior you want to increase. and be specific with your praise if you are using it as a reinforcer.

Do provide reinforcement contingent upon desired behavior. It will be very hard to build an association between the target behavior and the reinforcers.

Do provide reinforcement immediately, especially in the beginning. Research supports that positive reinforcement is most effective when it occurs immediately after the behavior, because there is a strong connection between the behavior and the reinforcement. This is especially true for learners who struggle with cause and effect. When reinforcement is delayed, there is also a risk that different behaviors will be accidentally reinforced (your provide reinforcement much after the initial behavior was demonstrated, therefore by the time they receive the reinforcement, the connection between the reinforcement and behavior made be made with a behavior they demonstrated after the initial behavior to increase). You do, however need to plan for delaying reinforcement overtime, as in real life, our behavior is not always reinforced immediately (think paychecks!)

Do pair secondary reinforcers with primary ones. This is how you get a token board to work -by having the learner associate the tokens with earning reinforcement. 

Do take the time to find out what's actually considered reinforcement for your learners (or your spouse, or you colleagues, or yourself). Just because someone "likes" something, does not mean they will engage in a specific behavior (brush teeth, complete homework, wash dishes) for that thing. Likewise, we assume all kids respond to praise, but this is not always the case. See my Preference Assessment post for more information on this.

Do continue to assess the effectiveness of your reinforcers. Oftentimes, what works for a period of time, no longer continues to be reinforcing. Think about it - when I was a child, my behavior was very reinforced by earning stickers or trips to buy new stickers. When I turned 14? Not so much. Reinforcers change, sometimes rapidly. Always be on the lookout for when reinforcement becomes ineffective and make changes.

Do plan to fade your reinforcement out by thinning the schedule. Making a plan to fade is key in maintaining behavior change. You won't be able to always reinforce every occurrence of a behavior, and even if you could, it would do the learner no good! Real life isn't like that. When you plan to thin the reinforcement schedule, do so systematically. Don't go from reinforcing every occurrence of behavior to reinforcing at the end of the day (particularly when the behavior occurs many times a day). In this example, you are expecting the learner to go from earning reinforcement every single time he does something, to after he does that same thing 50+ times. The learner will not be able to determine if reinforcement is still available, thus will likely decrease in his responding/no longer perform that behavior that once received reinforcement. In order to eventually go to a thin schedule of reinforcement, such as receiving reinforcement at the end of the day for engaging in the behavior you wish to increase, you need to gradually get there. You could start out reinforcing for every occurrence, then over time begin to reinforce on a variable ratio schedule, such as every 3 (varied, so sometimes it will be every 2 responses, sometimes, every response, sometimes every 3 responses, etc.). Further thin the schedule by switching to a time-based schedule

Don't allow free access to specified reinforcers for particularly tricky behavior. If a child always refuses to wash his face before bed, it may not be in your best interest to provide "reinforcement" in the form of a video game that he gets to play whenever he wants. He will just play that game later. If you have a special reinforcing item or activity that he earns once he completes his night time routine whoever, he will work for that reinforcer. (If he doesn't, then it's not potent enough).

Don't assume that once you find something reinforcing, that it will work with all behaviors in all contexts. Earning a Skittle after using the toilet may be reinforcing, but it may not hold the same weight when the expectation is completing an entire night time routine. In this case, the Skittles is not a potent enough reinforcer for the m ore complex or time-consuming behavior. The behavior doesn't even have to be labor intensive for the Skittles to not be reinforcing. It could just be a highly unpreferred behavior and one Skittle for a reward just isn't going to cut it. You need to be assessing what works in different situations and adjust your reinforcement accordingly.

Don't let reinforcers get stale. Reinforcers should be rotated. In order to prevent satiation of any reinforcer (item or activity), teachers, parents, and staff should be working on continually adding reinforcers to the list of potential reinforcement so that there is always something available that is potent for your learners.

Don't get discouraged! If you are having trouble finding reinforcers that work, or figuring out how to reinforce the behavior you want, while not reinforcing the stuff you don't, reach out to a professional. Find the BCBA in your school, or the district psychologist and ask for a consultation to help. You can further ask for support in running a classroom-wide reinforcement protocol and focus o n reinforcing the group for desired behavior. Whatever you do, keep trying, and it will pay off  :)

ABA at Play: reinforcement, schedules of reinforcement, scientific research, manipulating MOs, fading, generalization, evidence-based practices, token economies, behavioral contingencies, socially significant behavior, individualized supports, preference assessments

How do you use reinforcement with your learners? Any strategies you've found to work?