Applied Behavior Analysis and Teaching Children with Autism

1
Applied Behavior Analysis and Teaching Children
with Autism
2
Characteristics of Applied Behavior Analysis

1. Massed vs. mixed trials
2. Errorless prompting
3. Differential reinforcement
4. Behavioral momentum

3
Characteristics of Applied Behavior Analysis

1. Emphasis on causes of behavior in applied
   settings (functions)
2. Emphasis on observable, measurable behavior
3. Science requires repeated measurement of behavior
4. Change environment to change behavior

4
Characteristics of Applied Behavior Analysis cont.

1. Methods and rationales can be defined precisely
2. Socially significant effects size and importance
   of behavior sought
3. High value placed on accountability for everyone
   involved in the behavior change effort
4. Treatment strategies can be refined and adjusted
   based on objective data of progress
5. Address separate and specific behavioral deficits
   excesses (language, social skills, behavior
   problems)

5
Terminology

• SD Discriminative Stimulus A stimulus (i.e.,
  instruction), in the presence of which, a
  particular response is likely to be reinforced
• SR Reinforcement The process by which some
  consequent stimulusor removal ofincreases the
  probability of a behaviors occurrence

6
ABCs of Behavior

• Antecedents - What happens before
• Behavior/Response
• Consequence - What happens
• after/during

ABC Model
SD
SR
Antecedent Instruction
Behavior Response
Consequence Reinforcer
7
Teaching Methodology

• Common behavioral approaches
• Discrete Trial Training (e.g., Lovaas, 1981)
• Highly structured, very specific
• Natural Language Paradigm (e.g., Koegel, ODell,
  Koegel, 1987) Natural Environment Training
  (Sundberg Partington, 1998)
• Incidental, specific to childs reinforcers
• Task Analysis (Axelrod, 1983)

8
Discrete Trial Training

• DTT uses Applied Behavior Analytic (ABA)
  technology. They are not the same thing.
• Discrete trial training (DTT) is only one method
  for teaching skills to children with autism and
  other developmental disabilities.

9
Discrete Trial Training

• A discrete trial method is not to be considered
  the only way to teach new behavior in fact many
  behaviors do not lend themselves to discrete
  trial training, and must be taught using
  alternate methods (e.g., task analysis training).
• Teaching should occur outside the DTT format as
  well.

10
Key Features

• Several key features make a discrete trial
  teaching method effective
• A trial comprises a single teaching unit
  (Lovaas, 1981)
• Concise and clearly defined - Provides clear
  expectations for the teacher and the student.

Lovaas, O.I. (1981). Teaching developmentally
disabled children The ME book. Austin Pro-Ed,
Inc.
11
Key Features

• The method allows for repeated presentations of
  trials, which appears to be critical for many
  children with developmental disabilities.
• Since a trial is clearly and objectively defined,
  a students performance is easily measured.

Lovaas, O.I. (1981). Teaching developmentally
disabled children The ME book. Austin Pro-Ed,
Inc.
12
Discrete Trial Training

• Discrete trial training is typically provided in
  11 direct instruction, but is also useful in
  providing incidental instruction and teaching in
  the context of various activities.

13
Discrete Trial Training

• Major parts to a discrete trial
• The Trainers Presentation/Instruction
• Clear, concise and phrased as a statement
• Given only once
• Not too many words
• Consistency in wording initially, vary later

14
Discrete Trial Training

• Major parts to a discrete trial
• The Childs Response
• Correct
• Incorrect
• None
• Allow 3-5 seconds
• When incorrect or none, intervene (prompting)

15
Discrete Trial Training

• Major parts to a discrete trial
• The Consequence
• Correct - Immediately present enthusiastic praise
  with other identified reinforcers.
• Incorrect/None - Prompt/guidance (may need to
  re-start the teaching sequence reinforce?)

• Maurice, Green Luce (Eds.). Behavioral
  Intervention for Young Children with Autism A
  Manual for Parents and Professionals.

16
Behavior Modification vs. ABA

• Does ABA create robotic behavior?
• Does ABA use bribery in the form of goodies to
  get kids to perform?
• Does ABA tout dramatic treatment effects, but
  only in isolated environments?
• Is ABA concerned with behavior in the natural
  environment?
• Does ABA rely on punishment as a primary
  treatment method?

17
Prompting
SR
SD
Instruction
Prompt
Response
Reinforcer

• The prompt gets the desired response so that it
  may be reinforced.
• The objective is always to minimize, fade, and
  eliminate prompts.

18
Prompting Techniques

• Types of Prompts

Verbal/Gestural Positional Model Touch Physical Other

• Prompt Fading
• Gradually removing a prompt to facilitate
  successful and independent performance

19
Prompting

• Use correction trials for prompted responses
• Errorless prompting
• Antecedent intervention (not how to respond to
  mistakes, how to prevent mistakes)
• New skills most-to-least (errorless prompting)
• Older skills least-to-most
• Prompt fading Transferring stimulus control

20
Prompt Fading

• Fade prompts on two levels
• Latencytime between SD and prompt
• Topographyintrusiveness of the prompt (type of
  prompt)
• Physical (partial to full)
• Gestures (demonstrations to small motions)
• Sounds (demonstrations to one phoneme)
• Positional (changing the placement of items)
• Other (pictures, lines, mats)

21
Prompt Fading

• Desired Response Prompt
• Receptive Instruction Full physical
• i.e., Touch your head. Partial physical
• Touch
• Model all of response
• Model part of response
• Gesture
• Most-to-Least Least-to-Most

22
Prompt Fading

• Desired Response Prompt
• Labeling Full model
• i.e., What is this?
• Two phonemes
• One phoneme
• Mouth movement
• Most-to-Least Least-to-Most

23
Prompt Fading

• Desired Response Prompt
• Eye Contact Withhold activity
• Hold R near
• Say students name
• Say Look at me.
• Physical prompt
• Most-to-Least Least-to-Most

24
Reinforcement

• R should be varied in type, intensity, and
  duration
• Differential R Better R for better
  responses
• Boredom is often satiation w/ R
• Track and record different effective reinforcers
• Always pair social praise w/ other reinforcers

25
Examples of Reinforcers
Reinforcing Dimension of Stimulus Examples
Social Smiles, being with another person, praise, playing games with others
Gustatory Taste or consumption related Food, drinks, candy
Auditory Music, singing, sounds
Visual Colors, bright lights, pictures, things that spin
Adapted from Fovel, J.T. (2002). The ABA
Companion
26
Examples of Reinforcers
Reinforcing Dimension of Stimulus Examples
Tactile Hugs, tickles, massage, roughhousing, vibration
Proprioceptive Throwing a ball, stretching, exercising
Olfactory Smells flowers, perfume, food smells
Vestibular Rocking, swinging, bicycle, jumping
Adapted from Fovel, J.T. (2002). The ABA
Companion
27
A Language Training Environment (Sunberg
Partington. 1995. Teaching Language to Autistic
and Developmentally Disabled Children.)

• Language training is viewed as the key feature by
  teachers, parents support personnel, etc.
• Language training is incorporated into all other
  activities (e.g.,self-care, play, non-verbal
  behavior).
• There must be a large number of daily trials
  under a variety of stimulus and motivational
  conditions.
• All relevant types of language training trials
  should be conducted (including requesting,
  labeling, conversation, etc.), not just receptive
  language.
• Data should be collected on performance.

28
Analysis of Verbal Behavior

• Actively addresses one of the key skill deficit
  areas for students with autism
  communication/language
• Based upon B. F. Skinners Verbal Behavior (1957)
• Verbal behavior is defined by its FUNCTION rather
  than form.
• Same word can have different meanings based on
  conditions under which it is acquired

29
Functional Language Development

• Words can have different functions at different
  times.
• Requesting
• Labeling
• Echoing
• Conversations
• Early language instruction should always start
  with requesting.

30
Functions of Language

• Want pencil pencil receive
  pencil MAND
• See pencil pencil praise
  TACT
• Write with pencil praise
  INTRAVERBAL
• Hear pencil pencil praise
  ECHOIC

31
Verbal Operants

1. Mand Asking for reinforcers (request)
2. Tact Naming or identifying objects, actions,
   etc. (label)
3. Echoic Repeating what is heard
4. Intraverbal Answering questions in response to
   verbal stimuli (conversation)

32
Verbal Operants

1. Textual Reading written words
2. Imitation Copying someones motor movements
3. Receptive Following instructions

Stop!
33
Natural Environment Training

• Stimulus items are chosen by the student
  (increased attention, motivation)
• Natural (rather than contrived) reinforcers
• Ideal conditions to teach requesting
• Interaction and instruction take place within the
  context of using those items
• Reduces the need for elaborate generalization
• Reduces the potential for inappropriate behavior

34
Natural Environment Teaching
Student ________________
Activity ________________
35
Reinforcement

• One of the most important components of discrete
  trial.
• Differential Reinforcement
• Your reinforcement efforts should be based on the
  students level of independent responding and
  attention. That is, if the student performs the
  task independently, the level of reinforcement
  should be higher than if prompts are required.
• there is a positive consequence if they can do
  the task without assistance.
• Helps to avoid prompt dependency.

36
Reinforcement

• Needs to be varied in type, intensity, and
  duration.
• Much of what people describe as boredom can
  actually be looked at as satiation with the
  environment and the reinforcers.
• Develop a section in the students notebook to
  log the reinforcers you have tried and found to
  be effective.
• Remember, just because a student doesnt find an
  item or activity rewarding the first time you
  introduce it, this doesnt mean that he/she never
  will. Keep experimenting and be creative.

37
Reinforcement

• Limit the use of primary reinforcers to
  independently correct trials.
• Social reinforcers should always be paired with
  primary rewards.
• Thereby the student learns to enjoy a variety of
  secondary reinforcers.

38
Reinforcement vs. Bribery

• Do not use rewards as bribery.
• Do not tell the student in advance about the
  reinforcer he will receive (dangling a carrot).
• Do not remind the child of the reinforcer he
  would be getting if disruptive behaviors were not
  occurring.
• Do not offer additional reinforcers when behavior
  escalates in attempts to calm him down.

39
Prompting

• Prompts should occur before the behavior, not
  after the behavior as a consequence for incorrect
  responding
• Fade prompts as soon as possible
• while ensuring correct and errorless responding
• Errorless Prompting
• With new skills, use an adequate prompt
  immediately. With maintenance skills, allow a
  few-second delay.

40
Prompt Fading

• First fade physical dimension of prompt. If you
  do have to use a physical/full verbal prompt,
  immediately represent and try for an independent
  response or a less intrusive prompt.

41
Introducing New Targets

• 1. Transfer Procedure
• Instructor Swim little
• STUDENT fish
• Instructor Whats this?
• STUDENT fish
• 2. Errorless Prompting (prompt with a 0 second
  delay)
• Instructor Whats this? Fish
• STUDENT fish

42
Pre-teaching/Pre-trial Prompts

• Some children will respond well to pre-teaching
  or pre-trial prompts
• Prompt is provided before the SD
• Instructor This is a fish? What is it?
• STUDENT Fish

43
Non-Responding

• If the child does not respond within 2-3 seconds
• give him the correct answer
• wait for him to imitate you
• ask the question again to get an unprompted
  response if possible.  
• Instructor What do we sleep in?
• STUDENT ltNo response for 2-3 seconds after
  SDgt
• Instructor Bed.
• STUDENT Bed.
• Instructor What do we sleep in?
• STUDENT Bed.

44
Incorrect Responding

• If the child gives an incorrect answer
• repeat the question and immediately say the
  answer (prompt with a 0 second delay)
• wait for the child to imitate you
• ask the question again to get an unprompted
  response
• Instructor Whats this?
• STUDENT woof, woof
• Instructor Whats this? Dog.
• STUDENT Dog.
• Instructor Whats this?
• STUDENT Dog.

45
Prompt Fading

• It is vital to fade prompts so that the student
  does not become dependent on prompting and so the
  response comes under control of the stimulus and
  the target verbal SD.
• This is accomplished by asking the question again
  in attempts to get an unprompted response.
• Instructor Whats this? Cat.
• STUDENT Cat.
• Instructor Whats this?
• STUDENT Cat.

46
Massed Trials vs. Mixed Trials

• It is not always possible to get an unprompted
  response right away
• Be careful and avoid frustrating the child if
  this is the case.
• Children vary in their ability to tolerate
  multiple trials (massed trials) but as a general
  rule, if you are still unable to get an
  unprompted response after the third attempt,
  accept the prompted response and move on.

47
Behavioral Momentum and Mixed Trials

• Separate prompted from unprompted responses with
  easy tasks (those you know the child will
  respond to correctly)
• Then go back to the missed item.
• Increase the number of easy tasks gradually
  while still going back for an unprompted
  response.  

48
Example

• Instructor Swim little
• STUDENT fish
• Instructor Whats this?
• STUDENT fish
• Instructor Touch the fish!
• STUDENT lttouches fishgt
• Instructor Whats this? (holding up fish)
  
• STUDENT fish
• Instructor Great job!

49
Errorless Learning and Behavioral Momentum

• We want to prompt new skills and thereby avoid
  incorrect responses
• we do not want the child practicing the wrong
  response.
• Immediately repeat the drill when the student
  responds incorrectly in order to keep the child
  from inadvertently learning to chain incorrect
  and correct responses.
• Mixing difficult targets with easy (mastered)
  responses increases the amount of reinforcement.
  
• Make sure you are using differential
  reinforcement when using behavioral momentum
  (make sure to use a stronger reinforcer for new
  targets than for mastered skills.

50
New Targets/Acquisition Skills

• 1.  Start teaching session with several
  maintenance items first (to build success and
  behavioral momentum).
• 2.  Fully prompt the acquisition item.
• 3.  Immediately provide a second learning
  opportunity (correction trial) for independent
  responding.
• 4.  Provide several maintenance items.
• 5.  Ask for the acquisition response again. (we
  are trying to reduce the level of
  prompts).
• 6.  If response is independent (no prompt
  required), reinforce immediately! If student
  does not respond independently, give a full
  prompt.

51

• 7.  If a full prompt is required, use
  correction trial
• procedures.
• 8.   Provide several maintenance items.
• 9.   Ask acquisition item with faded prompt.
• 10.  Ask for an independent response (if the
  response
• is independent or with a reduced prompt,
• reinforce immediately. If the same level
  of
• prompting is required, but the student is
  still
• activity attending, reinforce effort).

52
Maintenance Skills

• 1.      If student misses maintenance items, give
  full prompt.
• 2.      Ask for an independent response.
• 3.      After getting independent response, you
  do not need to
• stop immediately to reinforce the
  maintenance item.
• Provide social reinforcement (thats
  right) and proceed
• with the teaching session.
• In maintenance trials present SD only once, then
  wait three seconds for the student to respond
  independently.
• If the student does not respond (or responds
  incorrectly), use least-to-most prompting. Then,
  present one correction trial using the same
  strategies.

53
Teaching Guidelines

•  Provide SD (Discriminative Stimulus) only once,
  then prompt. Avoid making repeated demands.
• Intersperse skills (mix receptive, motor
  imitation, expressive, etc). Do not do all
  trials of a program at once.
• New Skills present the SD and immediately follow
  with a prompt (errorless learning). Follow the
  prompted response with social reinforcement and a
  second learning opportunity for the skill
  (correction trial)
• fade prompt to a one second delay (give him a
  chance to respond independently)

54
Teaching Guidelines

1. Once the student begins responding within the
   one-second delay, fade to two, then a
   three-second delay. If no response occurs within
   three seconds, always prompt the response.
2. Fade the type of prompt in addition to the timing
   of the prompt.
3. Each teaching session should begin with
   requesting (manding) to establish the therapist
   as a conditioned reinforcer and to determine the
   childs interest and motivation (which ensure
   consistent and optimal responding).

55
Teaching Guidelines

1. Once you ask for the reinforcer back, present the
   next task immediately (within one second). Do
   not make the student wait without access to a
   reinforcer.
2. Occasionally, ask for the reinforcer back but
   provide the student with the opportunity to
   immediately access it again by requesting
   (manding).
3. Reinforcers should be delivered after
   approximately 3-5 responses in the beginning.
   These numbers can be increased over time.

56
Teaching Guidelines

• Keep your pace up. You want short time between
  your instructions. This helps to create fluent
  responding (accurate and quick) and behavioral
  momentum.
• Remember to add in maintenance items to ensure
  good, fast responding (behavioral momentum).
• When working on maintenance items, make sure you
  use time delay (2-3 seconds before prompting).
  The goal is to have independent responding.
• Use Errorless Learning
• (it is our job to make learning fun)

57
Behavioral Terms

• Shaping- A process through which we gradually
  modify the childs existing behavior into what we
  want it to be. This is typically done by
  adjusting the requirements before reinforcement
  is given. For example, if a child is just
  learning to say words, he may just be asked to
  touch and item before receiving it. Later, we
  may require the beginning sound, a syllable and
  eventually the word.