This is a technique widely used by behavioral root psychology to learn skills.
One of the behavior modification techniques that Burrhus F. Skinner systematized in the process of developing his operant learning paradigm, which is based on the association of certain responses with the obtaining of reinforcers or punishments. In this article we will analyze what chaining is, how it is applied and what types exist.
What does chaining consist of?
Chaining is a behavioral technique that is part of the operant learning paradigm. It is used to develop new chains of behavior (that is, complex sequences composed of a pre-established series of simple responses) from other behaviors that are already in the subject’s repertoire.
Some typical examples of learning that can be developed through this technique are those that allow elaborate behaviors such as playing a musical instrument or driving a vehicle. Chaining is widely used to teach basic skills to children with special needs, such as putting on clothes or grooming themselves.
The responses that make up the behavioral chain act as conditioned reinforcers of the previous behavior and as discriminative stimuli of the next. There are two logical exceptions to this: the first response, which does not reinforce any other, and the last, which does not act as a cue for further behavior.
A similar operant technique is shaping, also known as the “successive approximation method.” The central difference is that in shaping a behavior is progressively perfected using differential reinforcement, while chaining consists of combining a series of responses in a given order.
How to apply this operant technique
To use chaining properly it is very important to take into account a series of recommendations. In the first place, it is convenient, as far as possible, to select behaviors that the apprentice already masters in order to maximize the effectiveness of the procedure; furthermore, these should be as simple as possible or be divided into simpler segments.
The behaviors, or links in the chain, that are necessary and that the subject does not dominate must be developed during the process. Some operant techniques that can be helpful in this regard are shaping, modeling, physical guidance, and verbal instruction.
It is preferable that the reinforcement be of a social nature, such as congratulations and smiles, rather than material, since the behaviors acquired thanks to this class of reinforcers are maintained to a greater extent. It is also important to use fading to eliminate supportive behaviors that may have been learned collaterally from the chain.
The 5 types of chaining
Chaining can be applied in a number of different ways. In the first place, it is necessary to distinguish between the chaining by means of total and partial tasks ; in one the learner must execute all the steps of the behavioral sequence in each trial, while in the other the chain is divided into several segments and these are taught separately before combining them.
In turn, chaining by partial task can be subdivided into four different types : forward chaining, backward chaining, pure partial and progressive partial. Let’s see what each of them consists of.
1. Through total task
The basic type of chaining corresponds roughly to the procedures we have described so far. In these cases, the learner carries out all the behaviors that make up the chain in order in each of the training exercises. Reinforcement is provided once the behavioral sequence has been completed.
In this type of partial chaining, behaviors are taught one by one, starting the chain again each time a new segment is added. Thus, in the first place the first link is executed and a reinforcement is obtained ; then the first behavior in the chain is carried out again and then the second, and so on.
The partial backward chaining is carried out in the opposite way to the previous type, although the scheme is very similar: the person executes the last response in the chain and gets a reinforcement; later it has to do the same with the penultimate and then with the last one again to obtain the reinforcement, etc.
4. Pure partial
In pure partial chaining, behaviors are taught one by one, generally following a linear order (such as forward chaining). In this case, it is not necessary to execute any other response in the chain except the one that is being worked on to achieve the reinforcer, although finally an integration must occur.
5. Progressive partial
This type of chaining is similar to the previous one, although the reinforcement is more gradual, as the name suggests. It would consist, for example, in rewarding the execution of the first behavior in the chain, then the second, then the combination of both, then the third…, until the complete chain is reinforced.