Reinforcement and ABA
Reinforcement can mean lots of things to different people. In the science of Applied Behavior Analysis, it has a very specific and narrow definition. That it is narrowly defined by its function does not narrow the range of possibilities: it can be money, smiles, warm water or an infinite number of things.
Reinforcement is any stimulus (something that a sensory organ can experience) that will increase the likelihood of a behavior reappearing.
Can a high pitched noise be a reinforcer? Yes, if the organism finds it pleasurable. Can a punch in the face lead to reinforcement? Yes, if it eliminates some of the throbbing pain of a toothache. A practitioner of Applied Behavior Analysis will seek out the function of a behavior by questioning how a consequence of the behavior creates reinforcement for the client/patient/student.
Reinforcement on a Continuum Reinforcement happens along a continuum from primary reinforcement ( food, water, other physical reinforcers ) to social reinforcers, such as social attention, praise or recognition. Many children with disabilities do not respond to secondary or social reinforcers, since they don't actually function to provide reinforcement. A child who has spent money will find a quarter reinforcing whereas a child with severe autism or cognitive disabilities will not find a quarter reinforcing.
Typical children and most adults generally respond to secondary and social reinforcement. We work long hours for monetary amounts that are deposited electrically into bank accounts we access online or with a credit card. The goal of ABA is to move children along the continuum to secondary reinforcers, so that they, too, will work for a pay check and learn to make choices about how they use the result of their own labor. For many children with disabilities, that needs to be taught, and it is often learned by "pairing" primary reinforcers with social or secondary reinforcers.
Once the replacement or target behavior is defined in an operational way, the ABA practitioner needs to find "reinforcers" that will drive the student's/client's behavior. Children with significant disabilities may need to be reinforced with primary reinforcers, such as favorite foods, but unless this reinforcement is paired with social or secondary reinforcers, it can create an unhealthy and unsustainable reinforcement strategy. Many sensory reinforcers can be successful with children with significant disabilities, such as low functioning autism, when you can discover what sort of sensory toy appeals to children. I have used buzzing toys, spinning toys, and even water play successfully as reinforcers with students with significant language and developmental disabilities. Some of these children like to play with musical toys.
It is important to create a rich menu of reinforcers, and continuously add items into a child's reinforcement menu. Reinforcement, like all matters of taste, change. Also, students can sometimes become satiated by too much of a single reinforcer, whether it's Blue's Clues or Reese's Pieces.
Often, practitioners will begin with a Reinforcer Assessment which can be done a number of different ways. A successful practitioner will ask parents or caregivers for the childs preferred foods, television shows or characters, activities and toys. These are often a good place to start. Reinforcers can then be presented in a structured or unstructured way. Sometimes two or three items are placed in front of the child at a time, often pairing preferred items with new items. Sometimes you can present a child with a large number of reinforcers at one time, and eliminate items a child ignores.
Research has evaluated regular reinforcement (on a schedule, from each correct response to every three or four responses) as well as variable reinforcement (within a range, such as every 3 to 5 correct behaviors.) It has shown that variable reinforcement is the most powerful. When a children/clients discovers that they are reinforced for every third correct response, they rush to the third response. When they don't know exactly when they will be reinforced, they tend to have stronger responses, tend to generalize across environments and tend to retain the new behavior. The ratio is important: too high a ratio too early may not help the subject learn the target behavior, too low a ration may lead to reinforcement dependence. As a child/subject learns the target behavior, the practitioner can "thin" the reinforcement schedule, increasing the ratio, and spreading out the reinforcement over more correct responses.
Discrete Trial Teaching
Discrete Trial Training, or Teaching (more acceptable now) is the principal delivery method for instruction in ABA, though ABA is increasingly employing more naturalistic methods, such as modeling and role playing. Still, each trial is a three step process: Instruction, Response and Feedback. The reinforcement happens during the feedback part of the trial.
During feedback, you want to name the target behavior¸ and in initial trials, you want to begin with a one to one reinforcement schedule. You will reinforce every correct response (or approximation. See Shaping) in a "one to one" schedule, so your student understands that he/she gets the goodies every time they give you the behavior you want.
Success in Reinforcement
The most successful reinforcement is when a child/client begins to reinforce themselves. That is the "intrinsic" reinforcement that some of us receive for doing the things that we value or enjoy the most. But let's face it. None of us would go to work without the paycheck, though many of us accept a lower paycheck (as lowly teachers) because we do love what we do.
Success, for many students with disabilities, is to learn to find social interaction, praise and appropriate social interaction as reinforcers, so that they acquire age appropriate social skills and function. Our hope is that our students will gain the level of social and cognitive function that will give them full and meaningful lives. Appropriate reinforcement will help them achieve that.