Why Children With Autism Often Fail to Acquire a Functional Intraverbal Repertoire and What to do About it Mark L. Sundberg, Ph.D., BCBA-D ()

Why Children With Autism Often Fail to Acquire a Functional Intraverbal

Repertoire and What to do About it

Mark L. Sundberg, Ph.D., BCBA-D

(www.marksundberg.com)

Intraverbal Behavior• The most significant aspects of human behavior involve

intraverbal behavior• language acquisition• education • social behavior and relationships• knowledge, thinking, and intellectual behavior• literature, poetry, stories, novels • understanding• employment • beliefs, views, & opinions • religion• international relations• facts of history....

Intraverbal Behavior• A young child’s early intraverbal behavior may be relatively

simple such filling in the words of songs or fun activities (e.g., “Ready, set...” “Peek-a-…”)

• These response may be multiply controlled (e.g., MOs)• But intraverbal behavior soon becomes quite complex and a

significant part of a developing child’s daily verbal behavior

What is Intraverbal Behavior?• Intraverbal behavior is a type of verbal behavior where the form of the

response (what is said, signed, written, etc.) is under the functional control of an antecedent verbal discriminative stimulus (SD) and some type of generalized conditioned reinforcement (Skinner, 1957)

• The verbal stimulus and the verbal response do not match each other (i.e., no “point-to-point correspondence”)

• Verbal stimulus control Verbal responses• “What’s your favorite movie?” “Cars”• “What’s your name?” “Neil”• “What kind of work do you do?” “I’m a behavior analyst”• “What causes autism?” “There are probably a

variety...”

How is the Intraverbal Different from the Mand, Tact, & Echoic?

• Antecedent Behavior Consequence• Motivation (MO) Mand Specific reinforcement• Nonverbal SD Tact Generalized reinforcement• Verbal SD Echoic Generalized reinforcement

(w/ a match)• Verbal SD Intraverbal Generalized reinforcement

(w/o a match)

These are all called “expressive language” in traditional treatments

Common Intraverbal Problems Experienced by Children with Autism

• Absent or weak intraverbal behavior, despite strong mands, tacts, and listener (receptive) skills

• Rote intraverbal responses• Difficulty answering questions--especially complex questions• Excessive rote scripting• No conversational skills • Echolalia with intraverbal questions• Poor peer intraverbal interaction• Irrelevant intraverbal behavior• Self as a listener with overt intraverbal behavior

Why the Intraverbal Repertoire may be Absent, Weak, or Impaired

• There are many potential causes of intraverbal problems. Here are a few...

• The child has not received formal intraverbal training • The child is given training, but it’s too early to focus on intraverbals• The specific target responses are not in the child’s repertoire as tacts,

listener discriminations (LDs), or listener responding by function, feature, and class (LRFFCs) (e.g., “What vehicle has wings?”)

• Single verbal stimuli and single verbal responses have been over conditioned

• The intraverbal curriculum is out of developmental sequence• The child does not have sufficient training on verbal conditional

discriminations

Special Problems Related to Teaching Intraverbal Behavior

• There are many complexities involved in the intraverbal. Here are a few...

• Verbal stimulus control vs. nonverbal stimulus control• Transitory and constantly changing SDs vs. the general static

nature of tact, mand, matching-to-sample• A separate curriculum is required• Tact, mand, and listener (receptive) prerequisites• Many words have no clear referent, but change the meaning of

words in a sentence (if, but, can’t, or, usually, its, for, of)• Verbal stimulus and response classes must be established • Verbal conditional discrimination training is usually necessary

Conditional Discriminations

• Conditional discrimination: “When the nature or extent of operant control by a stimulus condition depends on some other stimulus condition” (Michael, 1993, p. 14)

• One discriminative stimulus (SD) or motivational operation (MO) alters the evocative effect of a second stimulus, and establishes the second stimulus as an SD or MO, and they collectively evoke a response

Verbal Conditional Discriminations (VCD)

• What constitutes a verbal conditional discrimination and an intraverbal response?

• Two components of a verbal stimulus where one verbal stimulus alters the evocative effect of the second verbal stimulus, and collectively they evoke a differential intraverbal response

• Skinner (1957, p. 76) calls this a “compound verbal stimulus,” but does not use the term “verbal conditional discrimination”

• Antecedent Response• Verbal SD

1 + Verbal SD2 Intraverbal Response

Verbal Conditional Discriminations (VCD)

• Examples...• Antecedent (Verbal SD

1 + Verbal SD2) Intraverbal Response

• Big animal Lion • Little animal Mouse• Big vehicle Boeing 747• Little vehicle A toy bike

• VCD=VSD1 alters the evocative effect of VSD

2 or vice versa• A correct response is dependent on VCD between words, if

individual words are the source of control errors will occur

Teaching Intraverbal Behavior to Children with Autism

• Many children with autism have a difficult time acquiring intraverbal behavior because beyond simple intraverbals (“A kitty says...”), most intraverbal responses are part of a VCDs.

• For example...• “What’s your cat’s name”• “What’s my cat’s name?”• “What did the cat chase?”• “What did the dog chase?”

Current Study• Is there a general sequence of increasingly complex verbal stimuli

and VCDs that can be used for assessment and intervention?• When are typically developing children successful at these tasks?• A revised version (v. 5-2) of the 80-item intraverbal subtest of the

VB-MAPP was designed with increasingly complex intraverbal tasks. This version was modified as function of previous field-test data

• This is the 3rd large-scale administration of the VB-MAPP IV subtest

Current Study

• Approximately 8,500 intraverbal responses were collected• 40 typically developing children and 71 children with autism

served as participants (bringing the project total to 91 typically developing children and 262 children with autism)

• Ages ranged from 23 months old to 15 years old• Parents and professionals administered the assessment based on a

set of written instructions

Acknowledgments for the People who Conducted the Field Testing

• Kristen Albert• Judah Axe• Vincent Carbone• Lori Chamberlain• Anne Cummings• Carla Epps• William Galbraith• Rebecca Godfrey• Lisa Hale• Ally Labrie• Heather Law

• Mike Miklos• Paige Raetz• Rikki Roden• David Roth• Rachael Sautter• Carl Sundberg• Cindy Sundberg• Brenda Terzich• Joel Vodovic• Kaisa Weathers

Intraverbal Assessment: Level 7: Multiple SDs with Prepositions, Adverbs, & Negation

Verbal SD Score Response

What do you eat with?

What animal moves slow?

Tell me something that is not a food

What do you write on?

Where do you talk quietly?

What is something you can't wear?

What do you sit at?

What is between the blankets and the bed?

What animal goes fast?

What's something that is not a musical instrument?

4030201000

10

20

30

40

50

60

70

80

Typically Developing Children Intraverbal Scores

Participants

Ag

e in

Mo

nth

s an

d I

ntr

aver

bal

Sco

re

Age in Months

Intraverbal Score

3-years-old

Results and Error Analysis for the Typically Developing Children

• 1½-year-olds (20-30 mands and tacts)• IV Assessment scores: Less than 5• Generally no IV behavior

• 2-year-olds (100-200 mands and tacts)• IV Assessment scores: 10-30 range• Some intraverbal behavior, but no VCDs• Can do song fill-ins and fun IVs, some associations, animal sounds,

common fill-ins; limited WH answers (e.g., name, or one word answers); lots of echoic responses


• 2 ½ -year-olds (200-400 mands and tacts)• IV Assessment scores: 20-40 range• Some simple intraverbal behavior, getting some easy WH questions,

minimal VCDs• Frequent echoic responding, or “What?” “I don’t know” “Things”

When some intraverbal control was demonstrated, often simple IV relation, minimal verbal conditional discriminations, the last, or prominent word was usually the source of stimulus control

• “What do you smell with?” ... “Poopies” • “What grows on your head?” ... “Shoulders”• “What animal moves real slow? ... “Drink water slow?”• “What helps a flower grow?” … “Up”


• 3-year-olds (500-1000 mands and tacts)• IV Assessment scores: 40-60 range• Well established basic intraverbal repertoire, 100s of IV relations• But verbal conditional discrimination errors were prevalent• “What grows on your head? ... “Plants” • Many “WH” questions cause problems • “Why do people wear glasses?” ... “Because they do”• “When do we set the table?” ... “So we don’t make a mess of food”• “Why do you use a Band-Aid?” ... “A rainbow”• “Where do you eat? ... “Food”• Rote responses were evident• “What day is today?” ... “Rainy” (it was sunny)


• 3-year-olds (cont.)• Problems with prepositions, adjectives, adverbs in VCDs • “What’s under a house? ... “roof”• “What something that is sharp?” ... “Giraffe”• Trouble with negation, time, personal information (except first name) • “What’s something you can’t wear?” ... “Shirt”• “Tell me something that is not a food” ... “We don’t throw food.”• “What day is today?’ ... “Sunny”• “What is your last name? ... “Noah” “Jon” “Sofia” “Neil”• Three causes of errors: VCDs, complexity of the different parts of

speech, and meanings of individual words (e.g., “clothing” evoked echoic responses for many 3 year olds)


• 3½-year-olds (500-1200 mands and tacts)• IV Assessment scores: 50-70 range• Verbal conditional discrimination errors were still common • “What grows on your head?” ... “Hat”• “Name some clothing” ... “For the body”• Negation still a major major problem • Still having problems with, prepositions, adjectives, adverbs in VCDs• Still problems with time concepts • Still emitted echoic responses when no intraverbal occurred


• 4-year-olds (800-1800 mands and tacts)• IV Assessment scores: 50-75 range• Verbal conditional discrimination errors were still common• “What do you smell with?”... “A skunk”• But VCDs are clearly getting stronger• “What’s above a house?”...“An airplane, and stuff that’s on the

roof”• Negation, time concepts, prepositions, and adjectives in a VCDs

continued to be a problem for many children• Specific words and concepts like “different,” “between,” “take”

“how,” & “why” caused problems


• 5-year-olds (1000-2500 mands and tacts)• IV Assessment scores: 55-76 range• They get it! They are much better at VCDs. • “What’s in a balloon?” ... “Helium” “Air”• However, they still have problems with negation, time concepts, and

prepositions• Many children missed “What day is today?” “What day is before

Tuesday” “What’s your last name.” “How is a car different from a bike?” “What number is between 6 and 8?”

Results and Error Analysis for the Children with Autism

• 71 children served as participants• Ages ranged from 35 months old to 15 years old• Consultants, classroom staff, and in-home

providers administered the assessment

7060504030201000

10

20

30

40

50

60

70

80

90

100

110

120

130

140

150

160

170

180

190 Children with Autism Intraverbal Scores

Participants

Ag

e in

Mo

nth

s an

d I

ntr

aver

bal

Sco

re

Age in Months

Intraverbal Score

4030201000

10

20

30

40

50

60

70

80

Typically Developing Children Intraverbal Scores

Participants

Ag

e in

Mo

nth

s an

d I

ntr

aver

bal

Sco

re

Age in Months

Intraverbal Score

3-years-old

Error Analysis for theChildren with Autism

• The children with autism made the same types of errors as typical children who scored at their level

• Verbal conditional discriminations were hard for all children especially those involving “WH” questions and the different parts of speech

• Rote responding was more obvious, and more firmly established• Echoic responses were more frequent• Negative behavior was higher with increasing complexity of the

verbal stimulus

Errors Made by Typically Developing Children and Children with Autism

Scoring at the Same Level

• IV Assessment Score of 0-19• Typically developing children and children with autism who scored

at this level emitted similar errors • Gave an echoic response to the question• Gave no answer• Gave a standard single answer like “yeah,” or “yes”• Pointed at something or pointed in some location (listener behavior)• Negative behavior was higher for children with autism (avoidance

and escape)

Errors Made by Typical Children and Children with Autism Scoring

at the Same Level

• IV Assessment Score of 20-29

• Samples Typical Autism• What can fly? “All gone shirt” “Water”• What can you sing? “Yes” No Response• What’s outside? “Outside” “Outside”


at the Same Level


• Samples Typical Autism• What are some colors? “1, 2, 3”

“Coloring”• Why do you use a Band-Aid? “On my finger” “Happens”• Where do you take a bath? “Mommy and daddy” “With toys”


at the Same Level


• Samples Typical Autism• What grows outside? “Sand” “Playground”• What shape are wheels? “Triangle” “Cars”• What do you wear on your head? “A ear” “Boo boo”


at the Same Level


• Samples Typical Autism• What color are wheels? “Circle” “Red”• What do you eat with? “Cheese” “Pizza”• Name some clothing. “Clothing” “Clothing”


at the Same Level


• Samples Typical Autism• What’s in a balloon? “It pops” “String”• What makes you sad? “Cry” “Cry”• What grows on your head? “Hats” “A plant”


at the Same Level


• Samples Typical Autism• What day comes before Tuesday? “Wednesday” “Wednesday”• What’s your last name? Gave full name Gave full

name• What number is between 6 and 8? “9” “9”

Implication for Intraverbal Intervention Programs

• Use the acquisition of intraverbal behavior of typically developing children as a guide for an intraverbal curriculum

• Assess a target child’s existing intraverbal repertoire using a tool like the one presented here

• Look for balance across the scores of the VB-MAPP (Is the child ready for intraverbal training?)

• Assess the corresponding mand, tact, and listener repertoires and be assured they are present prior to intraverbal training (e.g., “big” and “little” as tacts and LDs)

• Carefully sequence the intraverbal tasks (e.g., the VB-MAPP intraverbal task analysis)


• Typically developing 3-year-olds emit 1000s of intraverbal responses a day, thus there needs to be many trials for language delayed children.

• Don’t move from the nonverbal context too quickly• Don’t be in a rush to move up the curriculum• There is a need for massive verbal stimulus and response

generalization• Use LRFFC as a stepping stone to IV• Always analyze errors and the sources of control. Back down the

curriculum sequence if necessary• Mixed and rotated VB trials on one topic mimic general

conversation and provide a bridge to the natural use of the skill


• Respect the complexity of verbal conditional discriminations (VCDs)• Establish a strong repertoire of simple intraverbal relations prior to

moving to VCDs and WH questions (100s of IVs from Group 1 and 2)• Be aware that a (rote) correct answer to a question may occur when

the VCDs element is removed (e.g., all “What color” questions, or all “What shape” questions, or all “Where” questions). Mixed and rotated VB trials can solve that problem

• Be assured that the individual words evoke generalized intraverbal responses prior to combining them in a VCDs task (e.g., “grows” “head” “garden” individually evoke a variety of intraverbal responses)

Conclusions• The intraverbal behavior of typically developing children can

serve as an important guide for an intraverbal curriculum for children with autism or other developmental disabilities

• The primary behavioral relation relevant to intraverbal behavior almost always involves a verbal conditional discrimination

• Children with autism made the same intraverbal errors as typical children who scored at their level

• A behavioral analysis of these verbal errors can help us better understand the errors made by children with autism and avoid those errors by better sequencing the curriculum tasks, and by providing the necessary instruction, especially on VCDs

Conclusions• There is very little behavioral research on VCDs and their relation

to intraverbal behavior (for a review see Axe, 2008)• Existing conditional discrimination research can serve as a guide

for VCD intraverbal research (e.g., Saunders & Spradlin, 1989)

Funny Intraverbal Responses:“Kids say the darnedest things”

• Why do people wear glasses? ... Because they are old and tired

• What’s something you can’t wear?... I can’t wear my Halloween shirt• What do you write on? ... Not on the wall• What can you kick? … We only kick balls• Where do you put dirty clothes? ... In the dishwasher• Who drives a car? ... Daddy slept in the car• You sit on a... Time-out chair• Twinkle, twinkle, little ... Starbucks

Thank You!

For an electronic version of this presentation visit:marksundberg.com/ABAI

For a copy of VB-MAPP intraverbal subtest visit:Avbpress.com

Click on “Updates and downloads”

Why Children With Autism Often Fail to Acquire a Functional Intraverbal Repertoire and What to do About it Mark L. Sundberg, Ph.D., BCBA-D ()

Documents

weak intraverbal behavior

teaching intraverbal

intraverbal different

intraverbal curriculum

formal intraverbal

type of verbal behavior

behavior analyst

common intraverbal problems