. . . . . . . .

The new behaviour you want to build may be a series or chain of behaviours.  A behaviour chain is a series of related behaviours, each of which provides the cue for the next and the last that produces a reinforcer.

Almost everything we do can be considered part of a behaviour chain.  For example, when you are reciting the alphabet, you start with “A”, then “B”, then “C” and so on until the task is completed at “Z”.

Each step serves as a cue for the next step; a chain is really a series of signals and behaviours.  The completion of one behaviour in a chain produces the signal for the next action.  Saying “G” is the signal to say “H” next.

Practically any complex behaviour we do in the way of operant behaviour is part of a chain or a multitude of chains: eating, getting dressed, using the computer, counting, brushing your teeth, riding a bike, walking to school and so on.  Behaviour chains are very important to all of us; as is the procedure for building chains, which is called chaining.

Chaining is the reinforcement of successive elements of a behaviour chain.  If you are teaching your dog to retrieve, you are attempting to build a chain, if you are teaching your dog a freestyle routine, you are attempting to build a chain, and if you are teaching your dog to walk to heel, you are also attempting to build a chain.

There are two chaining procedures, forward and back chaining:

Forward chaining

Forward chaining is a chaining procedure that begins with the first element in the chain and progresses to the last element (A to Z).  In forward chaining, you start with the first task in the chain (A).  Once the dog can perform that element reliably, you have him perform the first and second elements (A & B) and reinforce this effort. Do not teach “A”, then teach “B” separately; “A” and “B” are taught together.  When these are mastered, you can move to “A”, “B” and “C”.  Notice they are not taught in isolation; hence the term ‘chain’.

Back chaining

This is often a very effective way of developing complex sequences of behaviour.  In forward chaining, you are teaching A to Z; in back chaining, you are teaching Z to A.  Back chaining is a chaining procedure that begins with the last element in the chain and proceeds to the first element.

As an example of back chaining, here’s a summary of how I teach the retrieve:

• Step 4: Hold, with dummy presented from my hand, once reliable (at least 80% ten second hold) add cue “fetch”

• Step 3: Release to my hand, once reliable add cue “give” or “dead”

• Step 2: Retrieve from the floor, this is shaped in small steps by presenting the dummy lower and lower, until the dog will retrieve from floor; fetch, hold, and release to my hand

• Step 1: Retrieve at distance, this is shaped in small steps by throwing the dummy further and further, until the dog will retrieve from distance; fetch, hold, and release to my hand.

This is a brief condensed example, throughout the game I also shape position such as retrieve to heel, sit and release etc.

Guidelines for chaining

• Define the target behaviour:  To teach a dog to perform the links of a chain, you need to know exactly what those links are.  Sometimes the links are very obvious, other times links are not so obvious.  It may be helpful to run through the target behaviour visually and make notes of all the steps involved.
  

• Reinforce successive elements of the chain: The elements in the chain must be reinforced in sequence.  Reinforce them as they happen.  In the example given, once your dog is reliable at step 4 and you begin to teach step 3, you will be reinforcing steps 4 and 3.  You will either be reinforcing at the end of the chain or at the end of as much of the chain as the dog has learned.  What your dog learns in a chain is not just a number of tasks; he learns to perform those tasks in the right order.  You can start at the beginning of the chain and work your way to the end or vice versa.

• Monitor Results: As with any training, you must keep track of the effects of your efforts.  Is a particular element reliable?  Should it be taught and reinforced a few more times?  Is it time to move on to the next element?  These are judgments that must be made during the chaining process, and they can be made accurately only if you carefully monitor the results you are getting.

Note: In a behaviour chain it helps to understand that a cue is also a conditioned reinforcer. In teaching a behaviour chain, by carefully timing your cues, you can mark a behaviour and reinforce it and cue the next behaviour simultaneously.

The video below shows a very good example of clicker training the retrieve using back chaining:

Related post - Shaping: Guidelines For Success