Professional Documents
Culture Documents
5000
4500
4000
BA Members 3500
3000
2500
2000
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
Years
Introduction
• Successful interventions for children with autism are probably
the main cause for the increase in ABA’s numbers
• This success with behavioral interventions comes as no
surprise to many in the field of behavior analysis
• Over 55 years of empirical and clinical research and
applications with the developmentally disabled
• Most of the basic teaching procedures for children with autism
were in place during the 1970s, including mand and intraverbal
training. None of this is “new.”
• Let me hear your voice (1993) by Catherine Maurice brought
widespread attention to behavior analysis
Introduction
• However, with this success there is some bad news
• Everybody is now an expert in behavior analysis
• Widespread dissemination of behavioral techniques, often by
unqualified people
• Simplifying the concepts and procedures beyond recognition
• Parallels to B-mod and education in the 60s
• “What happened to the promise of behavior modification?”
• Similar concerns for the current popularity of the use of
behavior analysis for the treatment of children with autism
Introduction
• What constitutes a “behavioral approach” to treatment of children with autism?
• Consumers must be confused because there are so many models out there, often quite
different from each other, but all claiming to be a “behavioral approach”
• DTT
• Lovaas model
• CARD model
• ABA
• Pivotal response training
• VB approach
• CABAS
• Competent learner model
• Natural language paradigm
• Milieu language training
• Incidental teaching
Introduction
• In addition, there are many • Sensory integration
other approaches and • Weighted jackets
treatments such as... • Deep pressure
• Floor time • Special diets
• RDI • Vitamins
• Son-Rise • Medications
• Holding therapy • Swimming with dolphins
• TEACCH • Decompression chambers
• Secretion • Chelation
• Auditory training • Faciliated communication
Introduction
• Thanks to the efforts of Dr. Gina Green and others, many of these
“pseudoscientific” approaches have been identified and consumers have been
warned about their ineffectiveness and even potential danger to children.
• However, recently Dr. Green has added the “Verbal Behavior Approach” to
her list of pseudoscientific approaches and suggested that it has similarities to
facilitated communication (ABA, 2005).
• In her recent presentation at the NY-ABA titled “Verbal Behavior;” An
evidence-based technology for autism intervention?” Dr. Green (2005)
concluded “the ‘VB’ approach to autism intervention does not appear to meet
accepted criteria for evidence-based practice or transferable behavioral
technology”
• Others have expresses concern about the dissemination of verbal behavior
procedures. Carr and Firth (2005) stated “little research exists to support such
widespread dissemination (of the VB approach).”
Introduction
• The purpose of the current presentation is to address the concerns
raised by Dr. Green, Carr and Firth, and others about the
“unwarranted dissemination of the verbal behavior approach” to
language training for children with autism.
• The goal of clarify what constitutes a “verbal behavior approach,”
while demonstrating its empirical foundation, on-going research
agenda, its value to children with autism, and hopefully, to get Dr.
Green to remove verbal behavior from her list of pseudosciences.
What Constitutes a Verbal Behavior
Approach to Autism Treatment?
• First, I share Dr. Green’s concern for the need for additional
verbal behavior research. I believe this point is uncontroversial.
(Sundberg, 1991: “301 Research topics from Skinner’s book
Verbal Behavior”).
• As of 2003, over 80 of these topics have been addressed
(Sundberg, 2003)
• Research was the primary purpose for starting the journal The
Analysis of Verbal Behavior, now in its 22nd Volume, and
published by ABA: International
• Second, I share Dr. Green’s concern about the improper
dissemination of behavioral concepts and procedures
What Constitutes a Verbal Behavior
Approach to Autism Treatment?
• Hung (1980)
• Simic & Bucher (1980)
• Lamarre & Holland (1985)
• Pierce, Epling, & Boer (1986)
• Hall & Sundberg (1987)
• Carroll & Hesse, (1987)
• Stafford, Sundberg, & Braam (1988)
• Yamamoto & Mochizuki (1988)
• McPherson & Osborne (1988)
• De Freitas Ribeiro (1989)
A Sample of Research on the EO, and the
Mand and Tact
• Sigafoos, Doss, & Reichle (1989).
• Sundberg, San Juan, Dawdy, & Arguelles (1990)
• Sigafoos, Reichle, Doss, Hall, & Pettitt (1990)
• Baer & Detrich (1990)
• Braam & Sundberg (1991)
• Sprague & Horner (1992)
• Williams, & Greer (1993)
• Twyman (1996)
• Drasgow, Halle, & Ostrosky (1998)
• Drash, High, & Tudor (1999)
• Brown, Wacker, Derby, Peck, Richman, & Sasso (2000)
A Sample of Research on the EO, and the
Mand and Tact
• Knutson & Harding (2000)
• Barnes-Holmes & Barnes-Holmes (2000)
• Goh, Iwata, & DeLeon (2000)
• Sundberg, Loeb, Hale, & Eigenheer (2002)
• Arntzen & Almas (2002)
• Ewing, Magee, & Ellis (2002)
• Winborn, Wacker, Richman, Asmus, & Geier (2002)
• Chambers & Rehfeldt (2003)
• Ross & Greer (2003)
• Nuzzolo-Gomez & Greer (2004)
• Taylor, Hoch, Potter, Rodriguez, Spinnato, & Kalaigian (2005)
• Petursdottir, Carr, & Michael (2005)
A Sample of Research on the EO,
and the Mand and Tact
• Taylor, Hoch, Potter, Rodriguez, Spinnato, & Kalaigian (2005)
• EO must be present to evoke mands (initiations to peers)
• Petursdottir, Carr, & Michael (2005).
• Mand training resulted in tacts, but tact training did not result in mands.
• There doesn’t appear to be a body of research that contradicts the separation of the
mand and tact.
• Research reviews
• Oah, S., & Dickinson, A.M. (1989). A review of empirical studies on verbal
behavior. The Analysis of Verbal Behavior, 7, 53-68.
• Sautter & LeBlanc (In press). The Empirical Applications of Skinner’s Analysis of
Verbal Behavior with Humans. The Analysis of Verbal Behavior.
Clinical Value of EO and Mand to
Children with autism
• Many children with autism have absent or defective mand repertoires
• A functional analysis of the child’s verbal behavior often reveals that the
response called a mand or a request is not under EO control, but rather S D
control, thus not, by definition a mand
Defective Mand - Ally
EO Does not evoke a mand
______________________________________________________
EO
Does not evoke a mand
Object
______________________________________________________
Intraverbal prompt
(e.g. “Sign cookie”) Evokes a response
Imitative prompt (not a mand)
(ASL sign)
____________________________________________________________________
EO Antecedent Control and Mand
Assessment and Intervention
in DTT and ABA Programs
• Lovaas, 1977, 1981, 2003 (clearly the most influential, outcome data)
• Expressive-receptive framework for language
• No mention of EO/motivation antecedent control
• Closest mand training activity is the “I Want___” program found late
in the program (between the adjective and preposition chapters).
• All language skills presented as SD control
• No focus on the fact that a single word, phrase, or sentence can be
strong in one repertoire and not another
• No functional analysis of words as behavior
EO Antecedent Control and Mand
Assessment and Intervention
in DTT and ABA Programs
• Taylor & McDonough (1996)
• Expressive-receptive framework for language
• Indirect mentioning of EO/motivation antecedent control (desired)
• Request training provided, but developmentally late (after verbs, body parts,
functions of objects. Multiple control (object present)
• More advanced requesting “asking WH questions” but described as S D rather
than EO control
• Clearly an improvement over Lovaas, but still no focus on a single response
form having multiple functions (a functional analysis)
• Most likely these authors would include EOs and manding should there be a
new version of the program, given several recent conference presentations
and publications
The Problem of the Terminology of
Traditional Linguistics
• What’s wrong with the term “request” and the other terms from linguistics
called “communication functions” (e.g., command, interactional, personal,
protesting, label, responsive, greeting, recurrence, existence, nonexistence,
rejection, denial, & location)
• Why would an applied behavior analyst be content with a classification system
of language based on response topography alone?
• That would certain not happen with a functional analysis of problem behavior.
Verbal behavior is behavior. A functional analysis is still the hallmark of our
field and should not stop at the verbal level
• Technical vernacular and etymological sanctions
• Most requests are mands, but many mands are not requests
• Why we don’t use “reward” either
• Many “requests” are often multiply controlled, but scored as correct.
The Distinction Between the
the Tact and the Intraverbal
Mands Textual
Imitation Matching Labeling
Vocal Play Receptive Intraverbal
Cooperation
Vocal Imitation
Nathan: Quick Assessment
4
Mands Textual
Imitation Matching Labeling
Vocal Play Receptive Intraverbal
Cooperation
Vocal Imitation
Matt: Quick Assessment
4
Mands Textual
Imitation Matching Labeling
Vocal Play Receptive Intraverbal
Cooperation
Vocal Imitation
Empirical Research on the Distinction
Between the Tact and Intraverbal
• Braam & Poling (1983)
• Chase, Johnson, & Sulzer-Azaroff (1985)
• Luciano (1986)
• Daly (1987)
• Lodhi & Greer (1989)
• Tenenbaum & Wolking (1989)
• Watkins, Pack-Teixeira, & Howard (1989)
• Sundberg, San Juan, Dawdy, & Arguelles (1990)
• Partington, & Bailey (1993)
• Sundberg, Endicott, & Eigenheer (2000)
• Finkel & Williams (2001)
• Miguel, Petursdottir, & Carr (2005)
Empirical Research on the Distinction
Between the Tact and Intraverbal
• Two examples of research
• Braam & Poling (1983) found that children with autism who could emit
specific responses under tact control could not emit the same response forms
under intraverbal control. Transfer of stimulus control between nonverbal S Ds
and verbal stimuli were successful.
• Miguel, Petursdottir, & Carr (2005) replicated the basic procedures from
Braam and Poling (1983) and concluded “ while participants were able to
tact…and point to the pictures….they were not necessarily able to reliably
produce thematically related intraverbal responses…(until) intraverbal training
was used”
• No body of research has emerged to show the tact and intraverbal are the same
Clinical Value of Verbal Stimulus Control
and the Intraverbal to Children with Autism
• Many children with autism have absent or defective intraverbal repertoires
• Verbal behavior evoked by verbal discriminative stimuli constitute a significant element
of human verbal interaction
• A functional analysis of the child with autism’s verbal behavior often reveals that the
response called intraverbal or conversational is not under verbal stimulus control, but
rather under nonverbal stimulus control, or EO control, thus not, by definition an
intraverbal
• Verbal stimulus control is extremely complicated, usually involving verbal conditional
discriminations (one verbal stimulus alters the evocative effect of another verbal
stimulus).
• Many common verbal errors by children with autism are related to defective verbal
stimulus control
• Many aspects of more complex VB involve multiple stimulus control consisting of tact
and intraverbal relations (e.g., “What color is that?” “What shape is that?”)
Verbal Stimulus Control and Intraverbal
Assessment and Intervention
in DTT and ABA Programs
• Lovaas (1977, 1981, 2003)
• No sections answering questions, fill-in’s, verbal categories, etc. or what could be
identified as intraverbal training.
• No mention of verbal antecedent control of verbal behavior (all under receptive
langauge)
• Taylor & McDonough (1996)
• Several aspects of the program contain intraverbal activities (answering questions,
fill-ins, etc.)
• Clearly an improvement over Lovaas, but still no focus on a single response form
having multiple functions; a structural listing vs. a functional analysis
• No focus on verbal stimulus control and verbal conditional discriminations.
• Sequence of tasks are out of order, both developmentally and based on a functional
analysis of the increasing complexity of verbal stimulus control
Is There Empirical Support for Skinner’s
Analysis of Verbal Behavior?
• What type of support is necessary and sufficient to
claim that there exists an empirical foundation for the
analysis of verbal behavior?
• Conceptual
• (e.g., Behavioral vs. cognitive explanations)
• Experimental
• (e.g., mand vs. tact; intraverbal vs. tact)
• Applied
• (e.g., A VB approach vs. a Discrete Trial approach)
Conceptual Support
• The analysis of verbal behavior involves the same behavioral
principles and concepts that make up the analysis of nonverbal
behavior
• No new principles of behavior are required
• “The emphasis is upon an orderly arrangement of well-known
facts, in accordance with a formulation of behavior derived from
an experimental analysis of a more rigorous sort” (Skinner,
1957, p. 11)
• Is there empirical support for principles and concepts that form
the basis of behavior analysis in general?
• YES! (e.g., JEAB)
Conceptual Support
Is there the necessary and sufficient empirical
support for the basic principles that provide the
foundation of Skinner’s conceptual analysis of
language?
Yes, it’s the same as behavior analysis in general
Conceptual Support
• What are the alternatives?
• Cognitive theories of language (e.g., Piaget, Brown,
Pinker)
• Biological/Genetic theories of language (e.g., Chomsky,
Lenneberg, Pinker)
• Is the necessary and sufficient empirical support available
for those theories?
• NO!
Experimental Support
• However, was Skinner’s interpretation of the basic facts
correct?
• For example…
• Is the distinction between the mand, tact, and intraverbal a valid
distinction?
• Do multiple sources of control have an additive effects?
• Are autoclitics tacts of the controlling variables of primary
responses emitted because of their special effects on listeners?
• There clearly are a number of research projects that would be
necessary to conclusively say whether the analysis has
empirical support or not (Sundberg, 1991).
Experimental Support
• What is the current status?
• There is a growing body of empirical evidence that
supports elements of Skinner’s analysis, such as distinction
between the elementary verbal operants (e.g., Sautter &
LeBlanc, 2005).
• For example…
• Is MO control different from SD control?
• Is nonverbal stimulus control different from verbal
stimulus control?
• Does tact training produce mands and intraverbals?
Experimental Support
• Does this constitute the necessary and sufficient empirical support?
• No, but it is a good start on the necessary support
• Substantially better than the empirical support for not recognizing the
distinction between the mand and tact
• Substantially better than the empirical support for the distinctions related to
traditional treatments of language
• Is there a comparable line of research available from, for example, the
elements of Piaget’s, or Pinker’s analysis of language and autism?
• The existing data on VB are not yet sufficient
Applied Support
• Is there enough empirical support for the separation of the elementary
verbal operants to use them as a basis for language assessment and
intervention for language delayed children?
• For example: The distinction between the mand, tact and intraverbal.
• Yes
• Is there stronger evidence that there is no distinction between the mand,
tact, and intraverbal?
• No
• Does the conceptual and empirical evidence obtained thus far suggest that
it is okay to ignore the mand and intraverbal?
• No
Applied Support
• Is an approach to language assessment and training that is
based on Skinner’s analysis of verbal behavior more
effective than an approach based on cognitive theories of
language?
• It is an empirical question
• However, given the evidence obtained thus far on the
separation of the verbal operants, transfer of antecedent
control between operants, and the use of multiple sources
of control to generate new operants, the distinction seems
quite valuable
Applied Support
• Is there empirical support for the various VB
teaching procedures and strategies to generate
verbal repertoires for children with autism or other
language delays?
• No
• There are many different ways to use behavior
modification procedures to teach a mand
Applied Support
• Is there enough evidence to support a behavioral approach
to language versus traditional approaches to language?
• Yes
• Clearly more basic and applied research is necessary, but
thus far the data support the analysis, and the interventions
have been successful
Conclusion
• Language is behavior and rightfully, the subject matter of
behavior analysts.
• The “verbal behavior approach” is behavior analysis.
• It makes use of the same basic components of applied
behavior analysis as the other behavioral approaches, but
employs a behavioral analysis of language as its
conceptual foundation.
• The term “VB approach” has only been used to distinguish
it from approaches that use behavioral procedures, but do
not make use of a behavioral analysis of language
Conclusion
• “The ‘VB approach’ is simply normative applied
behavior analysis with a few refinements. That is, it
incorporates all of the standard methodology of applied
behavior analysis, but it explicitly adopts Skinner's
interpretive framework for analyzing verbal
contingencies. In other words, it is a small variation on a
methodology that has an enormous empirical foundation.
The worst-case scenario is that the added framework
doesn't help. But even in that case the child is still
getting a full-fledged program of applied behavior
analysis procedures” (Palmer, 2005).
Conclusion
• “It is simply hard to believe that a set of
procedures guided only by a distinction between
receptive and expressive language can be as
sharp as one that respects all of the various types
of contingencies analyzed by Skinner” (Palmer,
2005).
Conclusion
The necessary and sufficient empirical support exists for the basic principles
that provide the foundation of Skinner’s analysis of language
The necessary and sufficient empirical support for the specific distinctions that
Skinner makes is emerging and very positive. However, much needs to be
accomplished
The necessary and sufficient empirical support for specific applications of the
analysis do not exist, and may never exist in the same form as those above
(e.g., outcome data).
It is a mistake to reject verbal behavior approaches based solely on applied
support
A substantial amount of research is still needed.
Conclusion
• Western Michigan University undergraduate curriculum in the 1970s
• WMU had/has the largest behavior analyst faculty in the world (20+)
• (Keller, Michael, Malott, Iwata, Ulrich, Hawkins, Lyon, Snapper, etc.)
• Psych 150 Introduction to Behavior Modification (Malott) & rat lab
• Psych 160 Child Development (Bijou & Baer)
• Psych 250 Abnormal Psychology (Ulmann & Krasner)
• Psych 260 Verbal Behavior (Skinner)
• Psych 350 Applied Behavior Analysis (Science and Human Behavior-
Skinner, JABA,), Applied Lab (e.g., KVMC w/ Jerry Shook)
• Psych 360 Experimental Analysis of Behavior (Operant Conditioning,
Honig, JEAB), & Pigeon Lab
• Psych 450 Research Methods (Stat book)
• Psych 460 Systems and theories (General Psych book)
Conclusion
• B.F. Skinner wrote in 1978…
marksundberg@astound.net
An Assessment of Typical
Children’s Intraverbal Behavior
• 28 typical children served as participants
• Most were from the Seattle area
• Ages ranged from 17 months old to 5 1/2 years old.
• Parents administered the assessment
• Instructions were given to the parents, including to write
down exactly what the child said following the
presentation of the verbal stimulus
Typical Children Age and Scores on
the Intraverbal Assessment
Typical Children Intraverbal
Assessment Scores
80
70 IV Score
60
50
20
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Children
Children with Autism Age and Scores
on the Intraverbal Assessment
Children with Autism
Intraverbal Assessment Scores
130
120
110
90
80
70
in Months and Intraverbal
60
Score
50 IV Score
40
30
20
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12
Children