Shaping

1
Shaping

• Chapter 19

2
Shaping

• Systematically and differentially reinforcing
  successive approximations to a terminal behavior
• Its a treatment and a natural process
• Typically used to teach behavior that is not yet
  occurring at all
• Example from Wolf, Risley, and Mees (1964)
  Wearing glasses

24
3
Differential Reinforcement

• The process
• Reinforcement is provided for responses that have
  a certain predetermined quality AND
• Reinforcement is withheld for responses that
  dont have that quality (extinction)
• The effects
• Responses similar to those that have been
  reinforced occur more frequently
• Responses similar to those that have NOT been
  reinforced occur less frequently (extinction)
• Extinction typically leads to new behavior

24
4
Successive Approximations

• Begin shaping by reinforcing responses that
• Are similar or a prerequisite to the terminal
  behavior
• And are already in the learners current
  repertoire
• When the initially reinforced responses become
  more frequent
• Shift the criterion for reinforcement to
  responses that are a closer approximation of the
  terminal behavior

24
5
Wolf et al. (1964)
Placing glasses in position
Putting glasses up to face
Putting glasses up to face
Picking up glasses
Picking up glasses
Picking up glasses
Touching glasses
Touching glasses
Touching glasses
Touching glasses
24
Beginning
End
6
Wolf et al. (1964)
RESPONSE Step 1 Touch glasses
Step 2 Pick up glasses Step 3 Put up to
face Step 4 Place in position
SR
Put on your glasses
No SR
RESPONSE Step 1 N/A
Step 2 Touch glasses Step 3 Pick up glasses
Step 4 Put up to face
No SR
(EXTINCTION)
7
Advantages of Behavioral Shaping

• Positive procedure
• Can be used to teach new behaviors especially
  those that cannot easily be leaned by physical
  cues or verbal prompts
• Can be combined with other behavior change
  procedures such as chaining

8
Disadvantages of Behavioral Shaping

• Time consuming
• Progress is not always linear
• The individual does not always proceed from from
  one behavior to the next in a continuous,
  uninterrupted flow
• Clinical judgment is involved how rapidly to
  increase requirements and how much each
  requirement should be increased.
• Often there are subtle changes in the behavior
  that are difficult to detect by an unskilled
  clinician
• Very hard to get IOA on intermediate behaviors

9
Increasing the Efficiency of Shaping

• Use a prompt such as
• Instruction
• Physical guidance
• Model
• Give the person a rule
• If you do x, you will get y

24
10
Before Deciding to Use Shaping

• Estimate how long it will take to reach the
  terminal behavior
• Can the person who will do the shaping be trained
  to do it well in the time available?
• Would it be more efficient to simply target the
  terminal step?

24
11
Shaping Guidelines

• Select the terminal behavior
• Consider behavior that will continue to be
  reinforced in the natural environment
• Define the behavior precisely
• Determine the criterion for success at each step
• Analyze the response class
• Response class Group of behaviors with the same
  function
• Helps you identify the approximations that might
  be emitted in the shaping sequence
• Identify the first behavior to reinforce
• It should already occur sometimes
• Should be a member of the targeted response class
• Eliminate distractions
• Proceed in gradual stages
• Limit the number of approximations at each level
• Continue to reinforce when the terminal behavior
  is achieved

24
12
Questions

• How are shaping and chaining alike? How are they
  different?
• How are shaping and prompting alike? When do you
  use each?
• What research design would you use when shaping a
  particular behavior over time?

13
Chaining

• Chapter 20

24
14
Behavior Chain

• A sequence of responses in which
• the result of each response (except for the last)
  is
• a conditioned reinforcer for that response
• an SD for the next response
• the reinforcer for the last response maintains
  the reinforcing effectiveness of the conditioned
  reinforcers in the chain

15
Putting on Pants

• Pick up pants at waistband facing forward
• Sit down
• Put feet through pant leg openings
• Pull pants over ankles
• Stand up
• Pull pants up to waist

16
SR Pants are up
Step 5 Stand and pull pants up to waist
Step 4 Pull pants over ankles
SR / SD Pants are over your ankles
Step 3 Put feet through leg openings
SR / SD Your feet are through the leg openings
SR / SD Youre sitting down holding pants
Step 2 Sit down
Step 1 Pick up pants at waistband facing
forward
SR / SD Pants are in your hand and
facing forward
17
Definitions

• Behavior chain
• The result of the performance of a sequential
  series of behaviors
• Chaining
• The way in which these behaviors are linked
  together
• E.g., ?

18
Task Analysis

• Before a chain is taught, each of its behaviors
  must be identified
• Task Analysis the process of breaking a complex
  skill or series of behaviors into smaller,
  teachable units
• Also, the list of the steps
• Observe, consult an expert, do it yourself!
• Should be individualized and include mastery
  criteria for correct over time and relevant
  behavioral dimensions (e.g., latency at each step)

19
Practice

• Work with someone else and write a task analysis
  for a skill
• Actually go through the motions of the skill if
  you can!
• After you write the steps, practice to make sure
  you didnt skip any!
• Modify as necessary

20
(No Transcript)
21
Assessing Mastery Level

• Determine which components of the task analysis a
  person can already perform independently 2
  ways
• Single-Opportunity Method
• No prompting or contrived reinforcement
• Present the SD and record or on your task
  analysis data sheet
• Stop the individual as soon as you score a -
• Do this over several days or sessions
• Multiple-Opportunity Method
• Same as single, but you allow the individual to
  continue if a - occurs

22
3 Types of Chaining

• Forward Chaining
• Teach the first step of the task analysis
• Reinforce this response
• Next, teach the second step
• Reinforce after the first AND second steps have
  occurred in succession
• Each succeeding step requires the cumulative
  performance of all previous steps in the correct
  order

23
Making the Bed Forward Chain

• Pull top sheet up as far as it will go.
• Tug at both sides to smooth out the wrinkles.
• Pull blanket up and smooth out the wrinkles.
• Fold blanket and sheet back together.
• Pull bed spread up and smooth out.
• Fold bed spread back.
• Put pillow on.
• Bring bed spread up over pillow.

24
3 Types of Chaining

• Total Task Chaining
• Variation of forward chaining
• Learner receives training on every step in the
  task analysis during every session
• Teacher prompts when necessary
• Continue teaching until learner is independent
  will all steps

25
Making the Bed Total Task

• Pull top sheet up as far as it will go.
• Tug at both sides to smooth out the wrinkles.
• Pull blanket up and smooth out the wrinkles.
• Fold blanket and sheet back together.
• Pull bed spread up and smooth out.
• Fold bed spread back.
• Put pillow on.
• Bring bed spread up over pillow.

26
3 Types of Chaining

• Backward Chaining
• Teach the last step of the task analysis
• Reinforce this response
• Next, teach the second to last step
• Reinforce after the second to last AND last steps
  have occurred in succession
• This sequence proceeds backward through the chain
  until all the steps in the task analysis have
  taught in reverse order and practice cumulatively
• Advantage learner comes in contact with the
  terminal reinforcer during every teaching session
• Examples pasting with Robby, teaching Bailey to
  blow a bubble

27
Making the Bed Backward Chain

• Pull top sheet up as far as it will go.
• Tug at both sides to smooth out the wrinkles.
• Pull blanket up and smooth out the wrinkles.
• Fold blanket and sheet back together.
• Pull bed spread up and smooth out.
• Fold bed spread back.
• Put pillow on.
• Bring bed spread up over pillow.