Chapter 13: Getting a new sequence of behaviours to occur with behavioural chaining

Behavioural Chaining Example

• getting and chewing gum

  • SD1 (pack of gum in pocket) → R1 (reach into pocket)

  • SD2 (hand in pocket) → R2 (pull out pack of gum)

  • SD3 (pack of gum in hand) → R3 (pull out 1 stick of gum)

  • SD4 (1 stick of gum in hand) → R4 (unwrap stick of gum)

  • SD5 (unwrapped stick of gum in hand) → R5 (put gum in mouth)

  • SD6 (gum in mouth) → R6 (chew gum in mouth) → terminal reinforcer (flavour of gum in mouth)

Is studying an example of chaining?

• no

  • Studying is not continuous — it involves taking breaks that do not reset your progress back to 0

Behavioural Chaining

• sequence of discriminative stimuli and responses in which each response produces the discriminative stimulus for the next response with the last response in the chain producing a terminal reinforcer

  • Each discriminative stimulus in the chain also serves as a conditioned reinforcer for the previous response and its ability to reinforce is related to its association with the terminal reinforcer

Behavioural Chaining — Linking

• Discriminative stimulus 1 → response 1 → discriminative stimulus 2 → response 2 …

  • each discriminative stimulus to response is a link in the chain

    • If a response fail to produce a discriminative stimulus, the chain fails at that point (weak link)

      • If someone is having trouble completing a chain you need to find the weak link and strengthen it

3 Methods for Teaching Sequences of Behaviour

• total task presentation

• Backward chaining

• Forward chaining

*not much difference between these in effectiveness

Total Task Presentation

• learner attempts all the behaviours in the sequence in each trial and continues until all the steps in the chain are mastered

  • Often used in sports to teach a certain technique — swinging at baseball

    • Provide feedback on each step

• Commonly used in natural environment

• Method of choice for simpler tasks with a small number of steps

Total Task Presentation — Advantages

• teacher spends less time in partial assembly or disassembly of a sequence in preparation for training

  • Other methods require breaking things down into sequences

• Can produce results quicker

• Maximizes the learner’s independence in early in training, particularly if steps are already familiar to them (don’t need to train these)

Backward Chaining

• systematically construct the chain in reverse order starting with the discriminative stimulus and the response that produce the terminal reinforcer, then in the next step add in the discriminative stimulus and response that produce the discriminative stimulus in the terminal link, and so on

• Used to teach dressing, grooming, work, verbal behaviour to special needs individuals

  • Putting on a coat — put it on fully for them without the zipper → … → put it on themselves

• Likely more effective for more complex tasks

Backward Chaining — Advantages

• always strengthening the discriminative stimuli as conditioned reinforcers by associating them with the terminal reinforcer

Forward Chaining

• teach initial link first using the terminal reinforcer, then train the initial and second links followed by the terminal reinforcer

  • Commonly used in natural environment

  • Likely more effective for more complex tasks

Total Task Presentation vs. Forward Chaining vs. Backward Chaining

Chaining vs. Shaping and fading

• similarities

  • All involve progressing through a series of steps to produce a new behaviour, new stimulus control over a behaviour or a new sequence of stimulus-response steps

• Differences

  • Shaping: steps consist of reinforcing closer and closer approximations to the final desired behaviour

  • Fading: steps consist of final desired behaviour coming under control of closer and closer approximations to the final desired stimulus

  • Chaining: steps consist of reinforcing more and more specific stimulus-response links that comprise the chain

    • Exception in total task presentation: all the links are taught from the beginning so not really adding to a sequence of behaviours

Factors influencing effectiveness of behavioural chaining

1. Task analysis

2. Encourage use of prompts by learner

3. Conduct a preliminary modeling trial

4. Begin training the behavioural chain

5. Use ample social and other reinforcers

6. Decrease extra assistance at each step as quickly as possible

Task Analysis

• break down task into smaller steps or component responses

• Order of sequence must be maintained

• Components need to be sufficiently simple to be learned without difficulty (Each step should be fairly simple)

• Select components so that completion of a response produces a clear change in stimuli that signals completion of the component (conditioned reinforcer) and occasions the next response (discriminative stimulus)

  • Clear indication of when to stop behaviour of holding glass under running water to fill and switch to next behaviour of removing the glass before it overflows (make a fill line)

Encourage use of prompts by learner

• learner uses prompts to guide their learning of a chain of behaviour

• Prompts can be visual, written, or verbal (self-instructions)

Conduct a preliminary modeling train

• model the entire sequence to the learner while verbally describing what you are doing and the discriminative stimuli that signal behaviours in the chain

Begin training the behavioural chain

• give learner an initial request to begin work and to complete the steps of the task

• Proceed with error correction if learner fails to begin responding or performs incorrectly

Error Correction

• provide necessary instruction to prompt response or use physical guidance to help learner perform step correctly

Use ample social and other reinforcers

• Use social reinforcers (praise) upon completion of each link when learner first begins to learn the chain

• Use reinforcer other than the natural reinforcer at the completion of the chain (food, toy…)

• Gradually eliminate these additional reinforcers so that the chain is maintained by the natural (terminal) reinforcer

  • Ex; putting on winter clothes

    • Praise every step

    • Candy when done

    • Natural reinforcer:opportunity to go play outside with friends

Decrease extra assistance at each step as quickly as possible

• fade additional instruction and error correction used when first training the chain as quickly as possible

Pitfalls

1. Unaware Misapplication Pitfall

2. Partial Knowledge Misapplication Pitfall

Unaware Misapplication Pitfall: Adventitious Chaining

• contains response elements that are not necessary for reinforcement

  • Response component that is not functional for producing reinforcement is the “superstitious component” (where superstitious behaviours arise)

    • Should avoid superstitious component or extinct them if they happen

  • Reinforcer will strengthen the functional response as well as the unnecessary response that precedes it

Unaware Misapplication Pitfall Example

• overeating due to snacking during commercials

  • Watch TV → Commercial starts which signals movement to kitchen to get food → return with food when TV program resumes

    • May be happening without awareness

    • Strong reinforcer to get TV program rather than commercials and get food

    • Break chain at weakest point to stop chain from reoccurring (don’t need to break each part)

Partial Knowledge Misapplication Pitfall

• children tend to make errors when an error leads to an imitation trial (instructor asks question then models the answer)

• Errors are reinforced by instructor providing answer and then reinforcing the student for imitation

  • Solution: higher reinforcement for correct responses on question trials; lower for imitation

Partial Knowledge Misapplication Pitfall Example

• teacher asks question and child gets it wrong

  • Teacher asks question again and models answer

    • Child learns to make errors to get reinforcement with an easier question

Guidelines for Effective Application of Chaining

1. Do task analysis

2. Consider strategies (different ways to get there)

3. Decide on chaining method (Total task presentation, backward chaining, forward chaining)

4. Use fading to decrease prompting

5. If using backward or forward chaining, ensure the learner performs all steps learned so far well before moving on to the next one

   • fade prompting to low level or non before adding new step

6. Start with reinforcement of each step, gradually decrease this reinforcement

7. Ensure the terminal reinforcement is effective