7/21/2019 1 Pdfsam Derived Responding

1/26

the amount of nonfunctional language emitted by children with autism (Karmali, Greer,Nuzzolo-Gomez, Ross, & R ivera-Valdes, 2005).

Several basic tact targets should be distinguished and ordered with respect to curricu-lum. Initial training should focus on naming of familiar three-dimensional objects (forexample, toys, animals, clothing, foods, and body parts) and people (for example, teacherand siblings), followed by naming of two-dimensional representations of those objects

and people. Subsequent targets include features or aspects of familiar objects (such as size,color, and shape), followed by specific location ty pes (for exa mple, kitchen or playground).More advanced tact targets include actions (such as jumping, rolling, or flying) and func-tions and classes followed by relational tacts including prepositions (for example, in/outof the box, above/below, and in front of/behind) and relational descriptors (big/little, less/more, or slow/fast).

Procedures for eaching Basic acts

Procedures for transfer of stimulus control are well suited to teaching tacts (Barbera& Kubina, 2005) when echoic and mand repertoires are previously established and whenshaping and errorless learning procedures are incorporated (Partington et al., 1994).Specific procedures are described below for structured work situations and play-basedinterventions such as the Natural Language Paradigm (NLP; Koegel, ODell, & Koegel,1987).

ransfer of Stimulus ControlProcedures for transfer of stimulus control establish responding under some version

of multiple control with either echoic prompts (for example, G.I. Joe), EO control (inother words, manding), or both, and the presence of the nonverbal stimulus (in otherwords, tacting). argeting multiple controlled responses rather than pure operants mayresult in higher compliance and greater e njoyment for the learner without sacrificing speedof acquisition or overall strength of the learned repertoire (Braam & Sundberg, 1991).Once responding is established, the extraneous controlling variables are faded until onlythe nonverbal stimulus occasions responding. Te clinician should initiate training withthe fewest possible extraneous controlling variables, because fading each variable requiresadditional time and effort. Tus, a child with a strong echoic repertoire and reasonablelevels of compliance might only require echoic prompts (for example, Say boat in thepresence of a toy boat or a picture of a boat) without the need for manipulation of EOsto occasion responding. Children with more limited echoic repertoires or who consis-tently emit problem behavior in response to demand situations might benefit more frominclusion of EO control in establishing initial responding and might also perform moreconsistently in a play context.

o use transfer of stimulus control procedures to teach tacts in a structured teachingenvironment, first identify a robust verbal operant repertoire (for example, mand, echoic,or intraverbal) and base your prompting strategies on that operant. Next, have the childsit at a table and present the nonverbal discriminative stimulus (in other words, show theitem to the child). If you are using errorless prompting strategies, immediately provide the

relevant prompt (for example, if the verbal operant is echoic, the prompt might be red).Respond to accurate responses with immediate praise and, if needed, either unrelatedreinforcers or the item (in other words, mand transfer). Note that as the tact repertoireprogresses beyond simple naming, you must present the relevant intraverbal prompt (suchas What is it? or What color is it?) simultaneously with multiple exemplars of thenonverbal stimulus (for example, red ball, red squa re, and red car) and appropriate nonex-emplars (for example, blue ball, pink ball, orange ball) to ensure that the relevant stimulusfeature will control responding. See the act raining: Names, Features, and Actions programfor a sample of specific instructions, though you will need to create other specificprograms for specific tact targets. Programs for advanced tacts (such as prepositions ortacts of sensory experiences) can be found at the end of this chapter (see act raining:Prepositions and act raining: Sen sory Experiences). Tese programs often work wellwhen receptive program trials are intermixed with tact trials (see chapter 7 for additionalinformation). See the act raining Data Sheet for a model data sheet for all tactprograms.

act responses to intraverbal prompts (such as What is that? or What color isthat?) constitute a functional repertoire; however, do not expect tacts taught exclusivelyas responses to intraverbal prompts to occur as independent descriptions of the envi-ronment (for example, Teres your red ball) in conversations. Conduct training (seethe Play-Based act raining: Natural Language Paradigm program and data sheet) inplay situations with natural change agents (for example, parents) to facilitate this use oflanguage (Gillett & LeBlanc, 2007; Koegel et al., 1987; Laski, Charlop, & Schreibman,1988).

Procedurally, Natural Language Paradigm (NLP) has several steps (Charlop-Christy

et al., 1999; Gillett & LeBlanc, 2007; LeBlanc et al., 2006). Adult and child face eachother with fun, common objects and toys (such as cup, fish, boat, and car). Te adultpresents an array of objects from which the child may select (in other words, a prefer-ence assessment) and restricts item access after selection. Te adult models an appropriateplay activity and spoken phrase (such as Fish swims) using the item. Echoic responseapproximations are shaped by providing contingent item access with continued modelsof the descriptive phrase while the child plays with the toy for twenty to thirty seconds.Te adult retrieves the item and presents a different exemplar by modeling a differentphrase for the same object (for example, blue fish) or presenting a new stimulus array

for selection. Te adult inserts a five-second delayed echoic prompt while modeling theaction in order to allow the opportunity for responses without any direct vocal model. Seethe Play-Based act raining: Natural Language Paradigm program for instructionsand create your own user-friendly data sheet that allows you to track targets and whetherresponses occur before or after echoic prompts.

Responses during initial trials occur in response to a vocal model and the EO createdby restricted access (in other words, echoic and mand response features). As the procedureprogresses and the five-second delay begins to occur prior to the vocal model in responseto restricted access and the nonverbal stimulus, the mand and tact features of the childs

response become more evident. Te therapist is advised to begin incorporating neutral toonly mildly preferred stimuli over time with the goal of minimizing the EO control overresponding to ensure that the ultimate response form is a pure tact (and not an impuremand or partial tact). Te response requirement can be shifted over time in several waysso that children learn to make longer responses (in other words, two- or three-word

Derived

RelationalRespondingApplications for Learners with Autismand Other Developmental Disabilities

Three sections of skill-building

strategies to help individuals with

developmental disabilities:

2Establish prerequisite skills for normal language

2Acquire early relational operants

2Speak with meaning & listen with understanding

2Read with comprehension

2Use appropriate syntax

2Develop reasoning, problem solving & creativity

2Be more empathetic, observant & self-regulating

PRACTICAL

APPLICATIONS

OF THE MOST

CURRENT

RESEARCH

A

Progressive

Guide to

Change

EDITED BY RUTH ANNE REHFELDT, PH.D., BCBA

&YVONNE BARNES-HOLMES, PH.D.FOREWORD BY STEVEN C. HAYES, PH.D.

.

.

.

.

.

.

7/21/2019 1 Pdfsam Derived Responding

2/26

erived Relational Respondin

Applications for Learnerswith Autism and Other

Developmental Disabilities

A Progressive Guide to Change

EDITED BY

RUTH ANNE REHFELDT, PH.D., BCBA

& YVONNE BARNES-HOLMES, PH.D.

Context Press

New Harbinger Publications, Inc.

Shaping.Shaping is defined as differential reinforcement of successive approximations orproviding reinforcers for behaviors that closely resemble the behavior that the child needsto learn (Baum, 2005). Tus, in teaching the response bubble, an instructor might firstreinforce the sound buh, bah, or even beh for several trials while listen ing closely forvariable responses, reinforcing sounds that even more closely resemble the word bubble(such as bub and buhl), and placing prior responses on extinction (in other words, no

longer providing reinforcement). Tis process continues until the childs response closelymatches the target response, in this case, bubble.Harris (1975) discussed the use of shaping to teach language to nonverbal children.

She said that the five steps in using shaping to teach language are the following: First,ensure that the child is attending to the teacher, and that the child has the ability tosustain this attention for a short period of time. Se cond, Harris suggests that the therapistbegin teaching by reinforcing all of the childs vocalizations. Tird, the therapist shouldreinforce only vocalizations that occur within six seconds of the model. Fourth, the thera-pist should reinforce all vocalizations that occur within six seconds of the model that

resemble the model. Finally, the therapist should provide a novel model and repeat thesteps described previously. Te therapist should continue to quiz the child on previouslylearned words to make sure his or her skills on the newly learned word do not decreasebecause they are not being used.

ransfer of stimulus control.All of the procedures listed above can be enhanced withthe introduction of prompts to occasion the target response or a close approximation.

When prompts are used, stimulus control over the behavior is established with the bridgestimulus (echoic prompt, picture prompt, and so on) and stimulus control is gradually

transferred to the relevant EO as those prompts are removed. Selection of the prompttype should be based on evaluation of the strength of other verbal operants, with strongoperants used to support the emergence of weaker operants. Children with strong echoicrepertoires (in other words, they display ready responses to a variety of vocal models)would benefit from echoic prompts. For example, a teacher might use incidental teachingto capture a naturally occurring EO of interest in a nearby sink to instigate teaching trialsfor water (turning it on, playing, or drinking) by providing the relevant vocal models atthe right times (in other words, saying on or water while standing with his or herhand on the faucet handle, or saying drink while water is running and the teacher is

holding a glass). Children with strong tact repertoires (in other words, they display readyresponses to an object or picture) would benefit from tact prompts (Arntzen & Almas,2002). For example, a therapist might prompt a child who can tact a glass of water (sayingwater when a glass of water is present) in the presence of the EO by giving him or hersalty foods, presenting the glass of water, and responding to the childs tact water witha glass of water to establish reduction of thirst as the maintaining reinforcer, and thirst asthe controlling antecedent stimulus.

After several successful prompted trials, stimulus control can be tran sferred by delay-ing the vocal model (echoic prompt) or altering the prompt (for example, the tact prompt

changes from a picture of a glass to a faint outline of a glass, or a partial echoic prompt).A commonly used strategy for transferring sti mulus control with children with autism isinstitution of a time delay before the presentation of the prompt (Charlop, Schriebman,& Tibodeau, 1985). After successfully establishing stimulus control with the prompt,the therapist pauses before the presentation of the vocal model or prompt to allow an

opportunity for the child to respond to the EO in isolation, rather than the combined EOand echoic or tact prompt condition used in previous trials, often termed a spontaneousrequest. ime delay can be implemented as a fixed delay (for example, five or ten secondson all trials) or a gradually increasing delay across trials (for example, one second, twoseconds, four seconds, six seconds) until the child responds during the delay interval.Fixed time delay is favored for mand training with visible, tangible stimuli and aftermany mands have been mastered; graduated time delay is recommended for initial mandtraining (Charlop-Christy, LeBlanc, & Carpenter, 1999).

Te act: Importance and ypes

In common vernacular, tacts are labels or descriptions emitted by the speaker. Skinnerdefined a tact as a response that is evoked by a particular object or event or property ofan object or event (1957, p. 82). Tus, a tact is occasioned by the presence of a discrimi-

native stimulus and is maintained by social reinforcers (such as praise or continued inter-action with the listener). For example, in the presence of an apple, the responses appleor fruit would be praised, while the response orange would likely be corrected (No,thats an apple!).

Te tact differs from the mand in that the controlling antecedent is a nonverbalstimulus rather than an EO, and the reinforcer is a general social one rather than specificaccess to a preferred event or stimulus change. For example, a child who sees an appleand says apple in response to a parent query What do you call this? has emitted atact. Te hungry child who says apple and immediately consumes the provided snackhas manded for the apple. Unlike mands, which are directly beneficial to the speaker,tacts are primarily beneficial for the listener, as they provide descriptive information thatmay enhance understanding or provide specificity. A child who tells a peer, Look in thetree! (mand) may provide additional specificity by describing, Tat black kitty is goingto jump (tact). Te listener now knows exactly what to attend to and can respond in asocial interchange, Tats too high. It might get hurt! Unfortunately, generalized socialreinforcers are notoriously unmotivating for young children with autism compared toother types of reinforcers (in other words, tangibles); however, tacts are a basic buildingblock for conversation and are critical for success in general education environments.

Te tact also differs from listener behavior, which is often referred to by various termsincluding receptive discrimination, receptive labeling, or receptive vocabulary. acts refer tothe behavior of the speaker (in other words, naming or describing a critical aspect of theenvironment), while listener behavior refers to ones ability to respond in various ways tothe important aspects of the environment specified by others. In the example providedabove, the speaker is directing a listener to observe a specific aspect of the environment,the black kitty in the tree (tact), while the selection response of touching the picture ofthe black kitty in an array of pictures of different animals would be a receptive discrimi-

nation (listener behavior). Tese two repertoires are related but functionally distinct andwill not necessarily develop without the direct targeting of each. acts are beneficial inthat they can facilitate social interactions, they facilitate development of generalized con-ditioned reinforcers by pairing weak social reinforcers (such as praise) with potent pre-ferred items (Partington, Sundberg, Newhouse, & Spengler, 1994), and they may reduce

7/21/2019 1 Pdfsam Derived Responding

3/26

Procedures for eaching Mands

Te goal of mand training is to teach a child to request an item or activity in responseto the relevant EO without a prompt (for example, What do you want?). Tis can beaccomplished by ensuring that the relevant EO is present during learning trials (Sundberget al., 2002).

EO Manipulations

As stated previously, mands taught in the presence of an EO are more likely to occurwhenever this EO is present, even if the EO is present in novel environments or with avariety of people (LeBlanc, Esch, Sidener, & Firth, 2006). wo primary strategies forensuring that training occurs in the presence of the EO are to capture naturally occurringEOs for learning opportunities, and to create the EO as a specific part of the instructionalplan (Shafer, 1994). Most practitioners use both of t hese approaches at different points in

a childs day.

Captured EOs: Incidental teaching.Several studies illustrate the beneficial effects ofincidental teaching with children with autism (Fenske, Krantz, & McClannahan, 2001;McGee, Morrier, & Daly, 1999), in which the child initiates the learning trial as heor she navigates the natural environment and the therapist monitors the childs interestsand creates related learning opportunities. Incidental teaching involves five steps (Hart& Risley, 1980). First, allow the child to interact naturally with his or her environment.Second, watch the child closely to identify an opportunity for the child to request a

desired or needed item in the environment. Tird, respond to the opportunity by requir-ing that the child request the item in his or her communication modality (for example,vocal, sign, PECS, or VOCA). Fourth, use shaping or transfer of stimulus control pro-cedures (see below) to facilitate an appropriate response from the child if the child doesnot immediately emit the appropriate mand. Finally, immediately give the requested itemto the child when he or she gives an appropriate response and, if necessary, praise thechild for emitting the request (for example, say, Good job asking for the cookie). Asthe mand response becomes stronger, eliminate the praise and provide only the requesteditem. See the Captured Establishing Operations: Incidental eaching with ime Delay

program for a detailed list of steps for using incidental teaching to teach mands whichcan be used with the Mands Data Sheet. A program for teaching mands for removalof aversive stimuli can be found at the end of the chapter (see Capturing EstablishingOperations: erminating Aversive Stimuli).

Contriving EOs.Another way to ensure the presence of the relevant EO during mandtraining is to intentionally create or contrive situations in which an item or event becomesmomentarily highly reinforcing immediately before you prompt responding. Contrivingan EO essentially consists of tripping a child up so that the child must request an

item or event to ensure that a preferred or needed activity or event can occur. Similar toincidental teaching, the therapist or parent has to be continually vigilant in watching forsituations that would make teaching opportunities. However, contriving EOs differs fromincidental teaching in that the parent or therapist is arranging the environment to occa-sion the trial rather than relying on naturally occurring situations to occasion the childs

interest. One might contrive thirst as an EO by providing salty foods (thereby momen-tarily increasing the reinforcing properties of a drink) and blocking access to a drink untilthe child asks for it. One might also cover a light switch with a hand as the child entersa room in order to increase his or her motivation to request that the light be turned on.Placing preferred toys on a high shelf can set the occasion for mands as well. Finally, onemight hold a childs swing to block the forward progress until the child requests (eitherwith or without prompts), Let go! or Swing!

Several studies illustrate the beneficial effects of identifying behavioral chains andcontriving an EO by interrupting the chain at some point (Sigafoos, Kerr, Roberts, &Couzens, 1994). Interrupted chain procedures typically begin with teaching a child tocomplete a chain or targeting a previously learned chain (for example, eating cereal andmilk with a spoon, or cutting out a circle from construction paper). At some point inthe chain the instructor prevents access to a component required for the chain (such asa spoon or scissors), creating an EO relevant to a mand trial for the relevant item. Notethat these interrupted chain procedures are particularly useful for testing and targeting

pure mands for missing items because the child requests the item when no visual cue forthe item is present.Tere are four steps to the interrupted chain instructional procedure to teach mands

(Duker, Kraaykamp, & Visser, 1994). First, a behavioral chain requiring several steps(such as making a sandwich or preparing and eating cereal) that the child can do inde-pendently should be identified. Second, a step of the chain should be identified as thetargeted mand (for example, Knife, please, when all other items needed to make thesandwich are available, or Spoon, please, when all other components for serving andeating breakfast cereal are available). Tird, the EO is contrived by creating a learning

trial where the child has all the items needed to complete a task except for one item. Inthe sandwich example, the child would sit at a table with peanut butter, jelly, bread, anda plate, and the teacher or parent would say, Lets make a sandwich! Allow the child toinitiate the chain and potentially request the needed item independently before model-ing the response or using some other transfer of stimulus control technique to facilitateresponding (see below). Finally, immediately provide the requested item upon request.See the Contrived Establishing Operations: Interrupted Chains program for the stepsinvolved in using an interrupted chain procedure to teach mands and the InterruptedChains: Sample Chains list for potential chains. Use with the Mand raining Data

Sheet. A program for teaching mands for a missing item in a game-type context also canbe found at the end of this chapter (see Contrived Establishing Operations: Te Whatsin the Bag? Program).

Supplemental eaching Procedures

In addition to the EO manipulation procedures described above, general instructionalstrategies such as shaping and transfer of stimulus control procedures should be incorpo-rated to increase the effectiveness of mand training. Errorless learning procedures are also

recommended such that early trials have immediate prompts sufficient to produce thedesired response, which are subsequently delayed (in other words, time delay) or altered(in other words, partial vocal prompt) and gradually eliminated. See Green (2001) fora review of errorless learning and other stimulus control technology procedures. Tesesupplemental procedures are equally appropriate for tact training.

Publishers NoteTis publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold with

the understanding that the publisher is not engaged in rendering psychological, financial, legal, or other professional services. If expert

assistance or counseling is needed, the services of a competent professional should be sought.

Distributed in Canada by Raincoast Books

Copyright 2009 by Ruth Anne Rehfeldt and Yvonne Barnes-HolmesNew Harbinger Publications, Inc.

5674 Shattuck Avenue

Oakland, CA 94609

www.newharbinger.com

All Rights Reser ved

Acquired by Cathari ne Sutker; Cover design by Amy Shoup;

Edited by Karen Stein; ext design by racy Carlson

A copublication of New Harbinger Publicat ions and Noetic Books

PDF ISBN: 978-1-57224-874-8

Te Library of Congress cataloged the print edition as:

Derived relational responding : applications for learners with autism and other developmental disabilities / edited by Ruth Anne

Rehfeldt and Yvonne Barnes-Holmes ; foreword by Steven C. Hayes.

p. ; cm.

Includes bibliographical references and index.ISBN-13: 978-1-57224-536-5 (hardcover : alk. paper)

ISBN-10: 1-57224-536-0 (hardcover : alk. paper)

1. Autistic children--Rehabilitation. 2. Developmentally disabled children--Rehabilitation. I. Rehfeldt, Ruth Anne. II. Barnes-

Holmes, Yvonne.

[DNLM: 1. Autistic Disorder--rehabilitation. 2. Association Learning. 3. Conditioning, Operant. 4. Developmental Disabilities-

-rehabilitation. 5. Reinforcement (Psychology) WM 203.5 D598 2009]

RJ506.A9D457 2009

362.1989285882--dc22

2008052211

Contents

S e r i e s E d i t o r L e t t e r . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v i i

FOREWORD

An Applied Behavioral Psychology of Language and Cognition . . .ix

Int roduct ion and Acknowledgment s . . . . . . . . . . . . . . . . . . . . . . . . 1

PAR 1Establishing the Prerequisites for Normal Language

CHAPER 1Reinforcer Identification Strategies and eaching LearnerR e a d i n e s s S k i l l s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7Tomas S. Higbee, Utah State University

CHAPER 2

Te Assessment of Basic Learning Abilities (ABLA) and ItsRelation to the Development of Stimulus Relations in Personswith Autism and Other Intellectual Disabilities . . . . . . . . . . . . . . . 25

W. Larry Williams a nd Marianne L. Jackson, University of Ne vada, Reno

7/21/2019 1 Pdfsam Derived Responding

4/26

CHAPER 3

Observing Responses: Foundations of Higher-OrderVerbal Operants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41Dolleen-Day Keohane and Jo Ann Pereira Delgado, Columbia Universityeachers College and CABAS; and R. Douglas Greer, Columbia University

Graduate School of Arts and Science eachers College

CHAPER 4

Joint Attention and Social Referencing in Infancy asPrecursors of Derived Relational Responding . . . . . . . . . . . . . . . . 63Martha Pelez, Florida International University

CHAPER 5

Establishing Mand and act Repertoires ... . . . . . . . . . . . . . . . . .79Linda A. LeBlanc and Courtney M. Dillon, Western Michigan University;and Rachael A. Sautter, Y.A.L.E. School

PAR 2Speaking with Meaning and Listening with Understanding

CHAPER 6

Nonrelational and Relational Instructional Control . . . . . . . . . . . 111Jonathan arbox, Center for Autism & Related Disorders; Rachel S. F.

arbox, Chicago School of Professional Psychology at Los Angeles; andDenis OHora, National University of Ireland, Galway

CHAPER 7

Naming and Frames of Coordinat ion . . . . . . . . . . . . . . . . . . . . . 129Caio F. Miguel, California State University, Sacramento; and

Anna I. Petursdottir, exas Christian University

Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143

CHAPER 8

Acquiring the Ea rliest Relational Operants: Coordination,Difference, Opposition, Comparison, and Hierarchy . . . . . . . . . . 149Carmen Luciano, Miguel Rodrguez, Israel Maas, and Francisco Ruiz,University of Almeria, Spain; Nicholas M. Berens, Center for AdvancedLearning, Reno; Sonsoles Valdivia-Salas, University of Albany, New York

CHAPER 9

Applying Relational Operants to Reading and Spelling . . . . . . . . 171Deisy G. de Souza, Julio C. de Rose, and Camila Domeniconi, UniversidadeFederal de So Carlos, Brazil

CHAPER 10

Syntax, Grammatical ransformation, and Productivity:A Synthesis of Stimulus Sequences, Equivalence Classes,a n d C o n t e x t u a l C o n t r o l . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 0 9Harry A. Mackay, Northeastern University and Praxis Inc.; and Lanny Fields,Queens College and the Graduate School of the City University of New York

CHAPER 11

Extending Functional Communication TroughR e l a t i o n a l F r a m i n g . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 3 7Rocio Rosales and Ruth Anne Rehfeldt, Southern Illinois University

PAR 3Self, Reasoning, Problem Solving, and Creativity

CHAPER 12

raining Analogical Reasoning as Relational Responding . . . . . . 257Ian Stewart, National University of Ireland, Galway; Dermot Barnes-Holmes,National University of Ireland, Maynooth; and im Weil, University ofNevada, Reno

the child) and may result in idiosyncratic or incomplete signs (the sign e quivalent of poorarticulation or volume control).

Individual abilities and deficits that have an impact on failure to develop the mandrepertoire should also guide selection of a communication modality and mand trainingprocedures (incani, 2004; Wraikat, Sundberg, & Michael, 1991; Bourret, Vollmer, &Rapp, 2004). Children may fail to exhibit mands due to dearth of items with reinforcing

properties, lack of a meaningful trained mand response, limited reinforcement opportuni-ties (for example, poor articulation, and so few viable listeners), or because EOs were notpresent during training and do not occasion responding. Bourret and colleagues (2004)found that mand training is more effective when the training procedures target specificproblems. For example, one might target a lack of response by selecting a topographybased on the childs existing repertoires; vocal sounds and verbal imitation may indi-cate that spoken language should be pursued, whereas limited vocal imitation repertoirewith good fine motor and motor imitation skills may suggest sign language, PECS, orVOCA.

Te Mand: Importance and ypes

Development of a functional mand repertoire can decrease problem behavior in childrenwith autism (Durand & Merge s, 2001) by replacing inappropriate behaviors (such as tan-trums and aggression) with less effortful and more appropriate means to obtain the samefunctional reinforcers (such as attention, specific tangibles, or a break from an aversivestimulus). Tus, teaching children to mand early in their verbal training can allow themsome control over their environment (terminating aversive stimuli, for example) and maydecrease the occurrence of problem behavior if the vocal response that is taught is lesseffortful to emit than the problem behavior that is maintained by the same functionalreinforcers.

For the purposes of this chapter, mands will be divided into simple mands and mandsfor information. Initial language training should focus primarily on establishing a robustrepertoire of simple mands across different environments and with different people.

Simple mands are those basic requests for preferred items (such as toys, food, or

drink), people, changes in the environment (for example, Lets go outside), and termi-nation of unpleasant situations (such as Stop, please or taking a break). Tese mandsshould be targeted as soon as a clear preference or motive is identified, and targeted con-tinuously until a strong and spontaneous repertoire exists across environments and withmultiple people. A spontaneous repertoire means that mands occur under EO control (forexample, hunger, thirst, or boredom) rather than as the result of a prompt from anotherperson (for example, What do you want?), such that the child is able to use languageto meet his or her needs when they occur rather than waiting for someone to notice hisor her distress.

As training continues with preferred items, mands for missing preferred items shouldbe periodically tested to ensure that the relevant EO (such as hunger), rather than anonverbal discriminative stimulus (such as the sight of food), is controlling the childsrequests. For example, requests for a meal that occur only when food is visible or can be

smelled and not when the child is hungry would not be particularly functional. Mands formissing items should be gradually introduced by conducting trials with known mands inthe absence of the discriminative stimulus (in other words, the actual preferred item). Forexample, a child who likes cookies and is hungry might be given a small visible portion ofa cookie upon request followed by an opportunit y to mand for additional cookies t hat arenot visible. Certain manipulations can be employed in order to evoke a mand for an itemin the absence of the actual item. For example, placing the child in a contextually relevantenvironment (for example, bringing the child to the kitchen and orienting him or hertoward the cabinet where the cookies are located), but keeping the actual item that is tobe requested out of sight, may evoke this mand. Te PECS also includes a phase for train-ing mands in the absence of the actual item. Tis occurs in phase two of training, wherethe items and the communication book are moved farther away from the learning andtypically out of the direct line of vision (Bondy & Frost, 2002). Mands that occur whenan item is missing are considered pure mandsbecause of the certainty one can have thatthe EO, rather than a visible discriminative stimulus, is the relevant antecedent variable.

Mands for removal of aversive stimuli should be taught in natura lly occurring contextsas soon as the aversive stimuli can be readily identified and removed. Often practitionersfail to recognize the importance of teaching a specific mand for a break, for help, or tostop until problem behavior maintained by escape from aversive situations has been wellestablished and must be subsequently replaced. Often, one to two generally applicablemands for removal of aversive stimuli will suffice across many situations, while a large ofnumber of specific mands for items are required to produce functional repertoires. Mandsfor removal of aversive stimuli should be targeted exclusively through captured EO pro-grams, as opposed to contrived EO programs, to avoid creation of unnecessary distress

and increased likelihood of the therapist becoming a conditioned aversive stimulus.A mand for information is a subtype of mand (Sundberg, L oeb, Hale, & Eigenheer,

2002) that specifies certain information as the desired reinforcer (for example, Who hasmy Elmo? What time is it? Where are my keys? or What is your name?). Tereare often at least two pertinent reinforcers (Sundberg et al., 2002): the information itselfand the outcome that the information allows the person to achieve. Te child benefitsfrom the information because it allows quicker and more efficient problem solving thana trial-and-error strategy (for example, approaching each person in the room to look forthe doll) would provide.

A mand for i nformation is more complex than a simple mand because it re quiresa child to identify the necessary information and a person who might have it, and toformulate a reasonable and understandable question. Some mands for information occurin a purely social context and have only one reinforcer (in other words, the informa-tion itself). For example, Where did you go on vacation? produces information thatfacilitates ongoing conversation and social connection but does not provide any othertangible stimulus change in the environment. Mands for information should be targetedwhen a child has an extensive tact and developing intraverbal repertoire. Initial trainingshould focus on mands for concrete and powerful secondary reinforcers in addition to

information (for example, Where is my favorite toy?), while mands for purely socialinformation should be targeted later because children with autism may not strongly valuesocial information or small talk.

7/21/2019 1 Pdfsam Derived Responding

5/26

an establishing operation (EO; Michael, 1988). An EO has been defined as a change inthe environment that momentarily increases or decreases the power of a reinforcer andevokes behaviors that have previously produced access to that reinforcer (Michael, 1982).For example, hunger increases the reinforcing value of food and evokes behaviors thathave produced access to food in the past (for example, asking Mom for lunch or cooking).

A tact is a label or description that occurs in the presence of a nonverbal discriminative

stimulus or condition that results in social reinforcers. For example, in the presence ofan apple, a child who says the word apple would be praised; if the child said the wordorange, he or she would not. Much of early lang uage development consists of these t woverbal operants as children learn to ask for things that they want and describe things thatthey see. Although they are not the focus of this chapter, other elementary verbal operantssuch as echoics (or verbal imitation) and textual behavior (the term for reading), are some-times used as prompts in training mands and tacts. Tese verbal operants are conceptual-ized as functionally independent such that a child who mands cookie when he wantsone may not be able to tact, or name a cookie when he sees a picture of one, a pattern that

has been repeatedly demonstrated to occur in children with language disorders (Nuzzolo-Gomez & Greer, 2004; Sigafoos, Reichle, Doss, Hall, & Pettitt, 1990).

Applying Skinners Analysis of Languageto Instruction

Sundberg and Michael (2001) outline several benefits of using Skinners analysis of lan-guage to teach children with autism. First, Skinners identification of the functional inde-pendence of the verbal operants indicates that each one has to be independently anddirectly targeted or it may not develop. In other words, children with autism can only beexpected to develop functional mands or requests if they are taught in the presence of animportant motivational variable (for example, learning to ask for water when thirsty) andnot just as a product of training to tact (label) a glass of water (wyman, 1996). Sundbergand Michael (2001) also identify the mand as the type of operant to teach first to childrenwith autism because the mand is uniquely valuable to the child and doesnt require socialreinforcers. Tat is, a child learns to mand for things he or she desires when the desire

exists, allowing the child some control over his or her environment (for example, I askfor an item and receive it). In addition, the child receives a powerful reinforcer follow-ing a verbal response (in other words, the child receives what he or she asks for), whichcan increase the reinforcing properties of speaking, resulting in a greater likelihood oflanguage development in the future.

When one is preparing to teach mands and t acts, it is critical to assess the chi ldsexisting repertoires for two purposes. First, one must determine the current profile offunctional language in order to identify which operants should be targeted and to deter-mine which existing operants can be used to prompt weaker or emerging operants.

Second, one must determine the extent of existing prerequisite repertoires, which couldaffect your selection of topography of the target response. Each of these areas is coveredbriefly below.

Functional Language Abilities

Te Assessment of Basic Language and Learning Skills (ABLLS) is one tool basedon Skinners analysis that allows therapists to identify a childs specific language deficitsin terms of verbal operants (Partington & Sundberg, 1998). Another tool is the VerbalBehavior Milestones Assessment and Placement Program (VB-MAPP; Sundberg & Hale,

2007). Each tool evaluates Skinners verbal operants to determine whether a child has aspecific verbal operant in his or her repertoire and allows you to track development viarepeated administration. Each verbal operant is assessed under the relevant antecedentconditions and with the corresponding consequences in place to determine if the childreadily emits that verbal operant. Te information obtained from the ABLLS can informparents and teachers about how to prioritize language targets. However, generally chil-dren should learn mand (requesting) and echoic (verbal imitation) skills before movingon to tacts (labels), questions (mands for information), and conversation (intraverbals;Sundberg & Partington, 1998). Te obtained profile can also allow the teacher or parent

to identify strong repertoires (such as echoics) that can be used in teaching other operants(such as mands). (See the section ransfer of stimulus control, below.)

Response opography

Selecting a modality for responding is critically important to success and maintenancein language instruction with children with autism. Language, or verbal behavior, does nothave to be vocal (in other words, spoken) to be meaningful or functional. Verbal behavior

can also include manual signing (incani, 2004; Bartman & Freeman, 2003), selectionresponses such as picture exchange (Charlop-Christy, Carpenter, Le, LeBlanc, & Kellet,2002; Chambers & Rehfeldt, 2003) or use of voice output devices (Mirenda, 2003). Allpotential communication modalities should be considered for all children, with modalityselection based on three important factors: available audiences, practicality or portabilityof the system, and child repertoires.

Vocal responses are ultimately portable and have broad potential listener communities,assuming adequate articulation, but they have the drawback of requiring complex vocalmusculature manipulations that cannot be directly prompted. Picture exchange commu-nication systems (PECS) and voice output communication aids (VOCA) can be usedwith almost any listener community but must be portable, well-organized, and constantlyavailable. Researchers have proposed several drawbacks associated with selection-basedsystems such as PECS. Sundberg and Michael (2001) distinguish between topography-based systems, where there is a unique topographical response (such as a spoken wordor sign) for each communicated idea, and selection-based systems, where the series ofresponses (such as scanning, selection, and picture delivery) is identical for each com-municative event. Sundberg and Partington (1998) argue that selection-based systems

often require more complex skills than initial appearance would suggest, with increasedprobability of difficulties in acquisition as language concepts become more abstract anddifficult to depict visually. Sign language has the benefits of being topography based andamenable to modeling and manual guidance, but it has the drawback of a restrictedverbal community (such as a deaf signing community and those trained to listen to

CHAPER 13

Understanding and raining Perspective aking asR e l a t i o n a l R e s p o n d i n g . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 8 1Louise McHugh, University of Wales, Swansea; and Yvonne Barnes-Holmesand Dermot Barnes-Holmes, National University of Ireland, Maynooth

CHAPER 14

Establishing Empathy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301Sonsoles Valdivia-Salas, University of Albany, New York; Carmen Luciano, Universityof Almeria, Spain; Olga Gutirrez-Martinez, University of Central Barcelona, Spain;and Carmelo Visdmine, Justice Administration, Madrid, Spain

CHAPER 15

M a t h e m a t i c a l R e a s o n i n g . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1 3Chris Ninness, James Holland, Glen McCuller, Robin Rumph, Sharon Ninness,and Jennifer McGinty, Stephen F. Austin State University; and Mark Dixon,Southern Illinois University

CHAPER 16

Deve loping Se lf-Directed Ru le s . . . . . . . . . . . . . . . . . . . . . . . . . . 335Carmen Luciano, University of Almeria, Spain; Sonsoles Valdivia-Salas,University of Albany, New York; Francisco Cabello-Luque, Universigy ofMurcia, Spain; and Monica Hernandez, Universigy of Jan, Spain

CHAPER 17

eaching Flexible, Intelligent, and Creative Behavior. . . . . . . . . . 3 5 3Catriona Ooole, Carol Murphy, and Dermot Barnes-Holmes, National Universityof Ireland, Maynooth; Jennifer OConnor, ABACAS, Kilbarrack, Ireland

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375

Dear reader,

Welcome to New Harbinger Publications. New Harbinger is dedicated to publishingbooks based on acceptance and commitment therapy (AC) and its application to spe-cific areas. New Harbinger has a long-standing reputation as a publisher of quality, well-researched books for general and professional audiences.

Most existing books in the AC series focus on teaching either practitioners or

members of the general public how to develop acceptance and mindfulness skills, andhow to apply those skills to various aspects of their experience. Although theoretical con-structs and concepts are discussed to some extent in these books, they never take centerstage. Tis book is different. It explicitly deals with relational frame theory, a new behav-ioral psychology of language and cognition that has emerged on many fronts over thepast thirty years. Human behavior is qualitatively different from and more complex thananimal behavior, and many would agree that it is the complexity of human language andcognition that sets us apart from nonhuman animals. So making advances in understand-ing how language and cognition work to our advantage and to our detriment is truly

important.One of the goals of acceptance and commitment therapy is to help people move

beyond the tyranny of the mind that brings so much unnecessary suffering into the livesof virtually all of usnot just people with clinical diagnoses. Because RF provides anunderstanding of how the mind works, it is helpful for an AC therapist (and in factany other therapist) to know and understand RF. However, psychotherapy is only oneapplication of RF, and this book makes it clear that RF has much broader practicalimplications and applications. A glance at the table of contents will immediately showthe incredibly broad scope of RF. Indeed, what sets this book apart from other books

on RF is the formidable range of areas and applied topics from education and clinicalpsychology that are covered in it. Several of the chapters examine specific skills that areessential to all human functioning, such as reasoning (including mathematical reasoning),perspective taking, and establishing empathy. All chapters present a balance of theory,empirical data, and specific applications that bring theoretical concepts to lifeoften inthe form of real-life or case examples. Te authors also offer useful suggestions on how toapply RF knowledge in a variety of contexts.

As part of New Harbingers commitment to publishing books based on sound, scien-tific, clinical research, we oversee all prospective books for theAcceptance and CommitmentTerapy Series. Serving as AC series editors, we comment on proposals and offer guid-ance as needed, and use a gentle hand in making suggestions regarding the content,depth, and scope of each book.

Books in theAcceptance and Commitment Terapy Series:

Have an adequate database, appropriate to the strength of the claimsbeing made.

Are theoretically coherent. Tey will fit with the AC model and under-

lying behavioral principles as they have evolved at the time of writing.

Orient the reader toward unresolved empirical issues.

Do not overlap needlessly with existing volumes.

7/21/2019 1 Pdfsam Derived Responding

6/26

Avoid jargon and unnecessary entanglement with proprietary methods,leaving AC work open and available.

Keep the focus always on what is good for the reader.

Support the further development of the field.

Provide information in a way that is of practical use to readers.

Tese guidelines reflect the values of the broader AC community. Youll see all ofthem packed into this book. Tey are meant to ensure that professionals get informationthat can truly be helpful, and that can further our ability to alleviate human suffering byinviting creative practitioners into the process of developing, applying, and refining thisapproach. Consider this book such an invitation.

Sincerely,Georg H. Eifert, Ph.D., John Forsyth, Ph.D., and Steven C. Hayes, Ph.D.

FOREWORD

An Applied Behavioral Psychology ofLanguage and Cognition

Behavior analysis is a unique tradition in psychology. It was based on the bold idea thata set of analytic principles could be developed from work with nonhuman animals thatwould eventually enable the understanding and analysis of human complexity. Studyingthe relatively simple acts in relatively simple contexts of organisms with relatively well-

known histories was a strategy, not the final goal of the research. B. F. Skinner said hisapproach was to create a promising conception of human behavior by a research focuswhich began with simple organisms in simple situations and moved on, but only as itsgrowing power permitted, to the complexities of the world at large (Skinner, 1938 p. xiv).Behavioral psychology aspired to give a comprehensive account of psychological phenom-ena. It was just taking a bottom-up approach to getting there.

Behavior analysts did not know if this strategy would succeed. It was not a certaintyit was a hope. Tere is nothing in evolutionary processes that ensures that the behavior ofnonhumans mustprovide a good guide to the analysis of complex human behavior.

Remarkably, it often does. Indeed, the behavior analytic strategy worked almost toowell. Functional principles of behavior proved to be powerful tools with vast appliedrelevance, but they did not seem to provide a fully adequate set of tools for the empiri-cal analysis of language and cognition. Tat combination was a bad one for behavioralpsychology. Progress was good enough to keep adherents happy but poor enough to leadothers to abandon hope of ever succeeding.

By the mid 1970s the mainstream had moved elsewhere, convinced that the originaldream of a comprehensive psychology was either unattainable or unlikely to be success-ful within the behavioral tradition. Psychology focused on cognitive science. Cliniciansmoved on to clinical theories of cognition. Microtheories were the norm, and grandvisions fell out of favor.

And there it stood for a few decades. But in the last fif teen years something remark-able has occurred. A true behavioral psychology of language and cognition has begun toform. Initially it resided almost entirely in the basic behavioral laboratories, but ever sogradually it has emerged in the applied arena as well. As befitted the inductive behavioranalytic tradition, each step built on the work of the past. Little needed to be thrownaway; instead, more and more was added until, finally, it became reasonable to suppose

that behavioral psychology could successfully and empirically move on, as its growingpower permitted, to the complexities of human language and cognition.One began to see the implications of this shift first in clinical behavior analysis as

approaches such as acceptance and commitment therapy, dialectical behavior therapy, andfunctional analytic psychotherapy began to revitalize the behavioral wing of cognitive

Pelez, M., Lubin, J., McIlvane, W., & Dube, W. (2001). raining discrimination, reflex-ivity, mixed identity matching, generalized identity matching, and arbitrary matching ininfants: owards the emergence of stimulus equivalence.Presented at the first interna-tional conference of the Association for Behavior Analysis, Venice, Italy.

Pelez-Nogueras, M., Field, ., Hossain, Z., & Pickens, J. (1996). Depressed motherstouch increases infant positive affect and attention in still-face interactions. Child

Development, 67, 17801792.

Pelez-Nogueras, M., & Gewirtz, J. (1997). Te context of stimulus control in behavioranalysis. In D. M. Baer & E. M. Pinkston (Eds.), Environment and behavior. Boulder,CO: Westview Press.

CHAPER 5

Establishing Mand and act Repertoires

Linda A. LeBlanc and Courtney M. Dillon, Western Michigan

University; and Rachael A. Sautter, Y.A.L.E. School

Children with autism typically exhibit substantial speech delays during early childhood,with up to 50 percent failing to develop any speech (Charlop & Haymes, 1994). Tosewho do develop speech often engage in echolalia,repeating words or phrases heard previ-ously, or do not speak for social purposes such a s engaging in conversation (Smith, 1999).Because one of the best predictors of outcome for children with autism is the develop-ment of spontaneous language before six years of age (Szatmari, Bryson, Boyle, Streiner,& Duku, 2003), it is essential to develop effective programs to teach language to childrenwith autism. Tis chapter describes behavioral techniques that can be used to teach basiclanguage to young children with autism in accordance with B. F. Skinners analysis oflanguage.

Skinners Analysis of Verbal Behavior

As discussed in chapter 3 of this volume, Skin ner (1957) defined language or verbalbehavior as behavior reinforced through the responses provided by other people, and hedefined several verbal operants according to the relevant antecedents and consequences.He classified language according to specific functional operants (for example, antecedent-behavior-consequence groups), rather than the traditional units of language (such as wordsand phrases). Tis chapter will focus on two of Skinners elementary verbal operants, the

mand and the tact. In common terms, a mand is a request (such as May I have that?)or an imperative (such as Give it to me). Te mand tells the listener what response oritem will serve as a reinforcer, and it occurs because of a highly specific motivation called

7/21/2019 1 Pdfsam Derived Responding

7/26

higher cognition. For example, it seems likely that the emotional and social aspects ofsocial referencing form the basis of the later development of perspective taking. Tatis, reciprocal conversation, cooperative play, and displays of sympathy and empathy forothers are all social abilities that require the basics of joint attention and social referenc-ing, because without them you would not use the ongoing cues of others to determinehow they were feeling and to act accordingly. Hence, it is not surprising that individuals

with autism who present with deficits in social referencing, for example, subsequentlydevelop considerable delays in their social and emotional skills.

Concluding Comments

Joint attention and social referencing are an intricate part of the tapestry of social interac-tions that comprise normal development. Not only are they critical to the developmentof social and related emotional repertoires, but they also appear to be essential precursors

to conditional discriminations and identity matching, which are also important precur-sors to language development and its core process of derived relational responding. Tecurrent chapter described teaching strategies for establishing conditional discriminations,

joint attention, and social referencing in young and de velopmentally disabled learners.Despite the importance of these skills, such training is far from easy. But there is simplyno way around thisif language and social and emotional development are desired andpotentially within the capabilities of the learner, then the difficulties must be enduredand the teacher must generate increasingly clever and creative ways to make the train-ing work. Although empirical evidence in support of the various teaching strategies out-lined is still scant, they offer good first steps toward the establishment of these essentialbuilding blocks of human development.

References

Augustson, K. G., & Dougher, M . J. (1992). eaching conditional discrimination toyoung children. Experimental Analysis of Human Behavior Bulletin, 9, 2124.

Bakeman, R., & Adamson, L. (1984). Coordinating attention to people and objects inmother-infant and peer-infant interaction. Child Development, 55,12781289.

Carpenter, M., Pennington, B. F., & Rogers, S. J. (2002). Interrelations among social-cognitive skills in young children with autism. Journal of Autism and DevelopmentalDisorders, 32,91106.

Charman, ., Swettenham, J., Baron-Cohen, S., Cox, A., Baird, G., & Drew, A. (1997).Infants with autism: An investigation of empathy, pretend play, joint attention, andimitation. Developmental Psychology, 33, 781789.

Dawson, G., oth, K., Abbott, R., Osterling, J., Munson, J., Estes, A., et al. (2004). Earlysocial attention impairments in autism: Social orienting, joint attention, and atten-tion to distress. Developmental Psychology, 40, 271283.

Devany, J. M., Hayes, S. C., & Nelson, R. O. (1986). Equivalence class formation inlanguage-able and language-disabled children.Journal of the E xperimental Analysis ofBehavior, 46, 243257.

Dube, W. V., MacDonald, P., Mansfield, R. C., Holcomb, W. L., & A hearn, W. H. (2004).oward a behavioral analysis of joint attention. Behavior Analyst, 28,197207.

Feinman, S. (1982). Social referencing in infancy. Merrill Palmer Quarterly, 28,445470.

Gewirtz, J. L., & Pelez-Nogueras, M. (1992). Infant social referencing as a learnedprocess. In S. Feinman (Ed.), Social referencing and the social construction of reality ininfancy. New York: Plenum.

Gewirtz, J. L., & Pelez-Nogueras, M. (2000). Infant emotions under the positive-reinforcer control of caregiver attention and touch. In J. C. Leslie & D. Blackman(Eds.), Issues in experimental and applied analyses of human behavior. Reno, NV:Context Press.

Higbee, . S., & Pelez-Nogueras, M. (1998). Reinforcer identification in infants.Behavioral Development Bulletin, 7,1014.

Holth, P. (2005). An operant analysis of joint attention skills.Journal of Early and IntensiveBehavioral Interventions, 2, 160175.

Jones, E. A., Carr, E. G., & Feeley, K. M. (2006). Multiple effects of joint attention inter-vention for children with autism. Behavior Modification, 30, 782834.

Klinnert, M., Campos, J. J., Sorce, J. F., Emde, R. N., & Svejda, M. (1983). Social ref-erencing: Emotional expressions as behavior regulators in emotion. Teory, Research,and Experience, 2,5786.

MacDonald, R., Anderson, J., Dube, W. V., Geckeler, A., Green, G., Holcomb, W.,et al. (2006). Behavioral assessment of joint attention: A methodological report.Developmental Disabilities, 27, 138150.

McClannahan, L. E., & Krantz, P. J. (2006). eaching conversation to children with autism:Scripts and script fading. Bethesda, MD: Woodbine House.

Mundy, P., Sigman, M., & Kasari, C. (1994). Joint attention, developmental level andsymptom presentation in autism. Development and Psychopathology, 6,389401.

Mundy, P., Sigman, M. D., Ungerer, J., & Sherman, . (1986). Defining the social defi-cits of autism. Te contribution of non-verbal communication measures. Journal ofChild Psychology and Psychiatry and Allied Disciplines, 27, 657669.

Pelez, M., Gewirtz, J. L., Sanchez, A., & Mahabir, N. M. (2000). Exploring stimulusequivalence formation in infants. Behavior Development Bulletin, 9,2025.

Pelez, M., Gewirtz, J. L., & Wong, S. E. (2007). A critique of stage theories of humandevelopment: A pragmatic approach in social work. In B. A . Tyer (Ed.), Comprehensivehandbook of social work and social welfare: Vol. 2.Human behavior in the social envi-ronment. New York: Wiley and Sons.

behavioral therapy. But these clinical extensions were not yet point-to-point empiricalextensions of the new basic behavioral work in language and cognition. Tat would takemore time.

Tis book is the next giant step in that process. It is the first to demonstrate a com-prehensive set of applied behavior analytic training approaches for language and cogni-tion that directly addresses most of the key areas within that domain. Te chapters avoid

needless quarrels between competing factions within basic behavior analysis; theories inthis volume are treated more as useful tools than as distinctions between warring camps.Even if individual chapters largely adopt a particular perspective, considered as an entireset they give testimony to the emergence of an increasingly unified behavior analyticapproach that is now ready to walk, step-by-step, from the simplest learning tasks all theway through empathy, self, and creativity.

Tat is a notable achievement, and one that may be a first in applied psychology. Iknow of no other book that extends a single approach within basic experimental psychol-ogy into intervention programs across the full range of issues that need to be addressed in

applied work in human language and cognition.Tis book, intended for parents and a variety of professionals working with individu-als with autism and other developmental disabilities, who understand the principles ofapplied behavior analysis, begins with the earliest steps needed to establish the prereq-uisites for normal language, helping applied workers think through how to determinereinforcers; how to establish observation, attention, and simple discriminations; and howto establish a simple repertoire of Skinnerian verbal operants. Tat section is closest totraditional applied behavior analysis and it deals with issues that many cognitive scientistsdo not address because they deal with higher-functioning participants in their studies of

reading or problem solving. Te book then steps firmly and boldly into derived relationalresponding and the issues of meaning and understanding. Chapters address instructionalcontrol, naming, acquisition of relational framing, syntax, reading, and functional com-munication. Tese chapters represent important steps forward within behavior analysis,bringing together research that is reasonably well-known but also showing in the total-ity how much progress has been made. Finally, the book dances into some of the mostcomplex issues of all as it considers self, reasoning, problem solving, and creativity, withchapters on analogy, perspective taking, empathy, self-rules, mathematics, and creativity.Especially in this last section, the volume reveals how bold and cutting-edge it is. Yet even

in this final section, and throughout the entire book, all of the chapters have empiricalsupport.

What is almost startling is that this book is not mere interpretation and logical exten-sion, or a broad vision to be tested in some distant future. It is also not a volume that triesto declare by fiat that a limited empirical analysis obviates an analysis of more complexissues. We have seen such volumes before. What we have not seen before is a compre-hensive empirical book that covers the full range of applied topics that educators andclinicians can begin to use now. In its scope, practicality, and empirical base, this volumedeclares that a comprehensive applied behavioral psychology of language and cognition is

here, is real, and is moving ahead.Will this approach fully succeed as measured against t he bold goals of behavior anal-

ysis? Tat we do not know. But a volume like this is needed to help applied workerstake the steps to find out; within the inductive, technique-building tradition of behavior

analysis, each step forward is likely to create progress that is sustained, since even whenwell-crafted steps fail, they provide important information.

Applied behavior analysis is a sophisticated and vigorous area with thousands ofsophisticated and creative applied professionals. I cant wait to see what all of t he wonder-ful behavioral educators and practitioners do with this bold new approach.

Steven C. HayesUniversity of Nevada

References

Skinner, B. F. (1938). Behavior of Organisms.New York: Appelton-Century-Crofts.

7/21/2019 1 Pdfsam Derived Responding

8/26

Introduction and AcknowledgmentsIn the 1970s, Murray Sidman and colleagues made an important, albeit unplanned, dis-covery that would serve as the impetus for a prosperous research program in behavioranalysis for years to come. Sidman (1971, 1977, Sidman & Cresson, 1973) found that,after researchers explicitly taught individuals with developmental disorders and limitedlanguage skills to match dictated names to the corresponding pictures and the pictures totheir corresponding printed words, the individuals proved capable of naming the pictures,

orally reading the text, and matching words to pictures and pictures to wordsa skillseemingly indicative of reading comprehensionall in the absence of direct instruction.Sidman (1971, 1977) termed these novel, emergent relations equivalence relations, as theuntrained skills seemed to represent symbolic or referential behavior, phenomena for whicha behavior analytic explanation had previously been lacking. Tese findings, and thosethat would follow in years to come, were exciting for the field of behavior analysis, forthey identified an economic and efficient means of establishing new academic and pre-academic repertoires in individuals with significant learning challenges, as each desiredskill did not have to be directly taught. But what was perhaps even more important is that

these developments, even where they were academic, have led to significant transforma-tions in our understanding of the basic processes of human language and cognitionmaking it possible that in years to come there will be no areas of learning that cannot betaught.

In the years since Sidmans pioneering discoveries, a voluminous body of laboratoryresearch has been conducted that further elucidated the conditions necessary and suf-ficient for the establishment of stimulus equivalence, multiple stimulus relations, andindeed whole networks of derived or untrained stimulus relations. Tis research led tothe formulation of several major theories regarding the nature of relational learning (such

as relational frame theory and the naming hypothesis), which have inspired even moresophisticated types of experimental research and innovative methodologies. Of course,it is not of paramount importance that we determine which is the best theory at onelevel or another. But the best test is one that shows which theory delivers most into thelives of those who need it, either scientifically and indirectly in the form of the advance-ment of knowledge or directly in the form of the development of new educational tech-nologies. And on both counts we have much to inspire us. For example, we know thatestablishing a history of reinforced relational responding in individuals with a varietyof learning challenges is an effective and efficient means of programming for the emer-gence of such educationally relevant skills as reading and spelling (Hanna, de Souza, deRose, & Fonseca, 2004), recognizing names and faces of caregivers (Cowley, Green, &Braunling-McMorrow, 1992), requesting preferred items (Rosales & Rehfeldt, 2007), andunderstanding basic numerical concepts (Lynch & Cuvo, 1995), to name a few. Tus,

Figure 4.3. Mother signals joyful cue to infant that predicts pleasant musical sound,contingent on infant reaching for object.

Figure 4.4. Mother signals fearful cue to infant that predicts loud sound and movementof object, contingent on infant reaching for object.

What thi s research demonstrates is that t he extent to which an i nfant orients to themothers face for cues in contexts of uncertainty depends on past success in obtainingsuch information, its validity, and its utility. For training purposes, therefore, either withvery young learners or with those who are developmentally disabled, the cues of othersmust be consistently contingent on the learners object-referencing behavior and mustreliably predict environmental consequences for the learners approach or avoidance. Tisinterpretation is summarized in Figure 4.5.

Figure 4.5.A learning approach to a social referencing pa radigm.

From this perspective, social referencing is an example of social knowledge with anemotional component. Put simply, the experienced learner gains the knowledge that ifanother is smiling when a stranger approaches, reinforcement is likely; but if the otherperson is cringing, for example, reinforcement is not likely. Tus, the facial expressionof the other becomes a setting event that establishes the function of the stranger as

being discriminative for positive or negative reinforcement or aversive consequences forapproaching. In line with this analysis, it should then be possible to establish learnersresponses to the basic emotions displayed by others and how they should act on this basis.Once these have been established, it is likely that a whole array of more subtle emotionalreactions and appropriate response patterns will be trainable within the context of simpleconditioning paradigms.

Why Are Joint Attention and Social ReferencingImportant for Derived Relational Responding?

Joint attention and social referencing would seem to have an i mportant role in t heestablishment of derived relational responding, thus forming the core of language and

7/21/2019 1 Pdfsam Derived Responding

9/26

a visual path from your eyes to the treat as you place it under a cup. Tis can also betrained explicitly. Tis is to be repeated until the learner chooses the right cup.

Joint object attention.With joint object attention, the learner orients quickly or directlytoward an object once another persons attention to the object has been discriminated. Atbest, the learner should also initiate your attention once a novel object has been identified.Consider the following scenario described by Jones, Carr, and Feeley (2006). Position atoy of choice less than five feet away from the learner, activate the toy, and turn and lookat the learner while pointing to the toy and commenting upon it (for example, Look atwhat the car is doing). It should be possible to get the learner to orient toward the toywithin as little as two seconds of your comment. Ten, to improve initiation on behalf ofthe learner, have him attend to the toy for several seconds, and encourage him to pointto the object while he looks at you. o reinforce this pointing response, you can simplymodel it or physically form the learners hand to point at the object.

Mutual object orienting with gestures.Holth (2005) described the following steps forestablishing mutual object orienting with gestures. Attach five or six envelopes to a wallin a horizontal line. In view of the learner, who is seated approximately ten to thirteenfeet away from you, place an edible reinforcer in one of the envelopes. In order to accessthe snack, the learner must guide you through the envelopes. Tat is, you will begin bypointing to the envelope farthest away from the one containing the snack, and promptthe learner to guide you with simple directives such as left and right, and stop whenyou reach the correct envelope. You can also arrange the envelopes in a vertical line andinclude prompts such as up and down. Ultimately, you should be able to arrange theenvelopes in a semirandom sequence (some side-by-side and others above and below) andall direct prompts to the learner should be faded. It is interesting to note some empiricalevidence suggests that this type of intervention not only improves mutual orienting andgesturing, but is also associated with language gains (Jones et al., 2006).

Empirical Evidence

Tere is some empirical evidence to support the use of training regimes for the estab-

lishment of joint attention in children with autism. In one study, MacDonald and col-leagues (2006) investigated joint attention initiations in twenty-one typically developingchildren (ages two to four) and twenty-six children with autism. As expected, the childrenwith autism demonstrated relatively minor deficits in joint attention responding and moresevere deficits in joint attention initiation. While the majority (78 percent) demonstratedgaze shifts, 44 percent demonstrated use of gestures, and only 22 percent were capable ofrelated vocalizations. However, after one year of participation in a comprehensive treat-ment program, all of the children with autism demonstrated gaze shifts, all had gestures,89 percent could vocalize, and levels of joint attention were now commensurate with the

normally developing counterparts.A study by McClannahan a nd Krantz (2006) also demonstrated the remediation of

deficits in joint attention in three children with autism (ages two to fi ve). In this research,photographic activity schedules were used to cue learners to play with toys in threelocations a puppet theater, toy shelves, and a toy box. W hen learners initiated use of toys,

they were manually guided to point to the toy while orienting to the teacher. Across trials,manual prompts were faded (from graduated guidance to spatial fading and shadowing),and the teachers proximity was decreased gradually. Te results indicated that all threechildren learned to point and orient for attention and could do so with novel stimuli.

Te Concept of Social ReferencingAlthough numerous authors i ntegrate the concepts of joint attention and social referenc-ing, the current chapter argues that they are distinct and that joint attention essentiallyprecedes social referencing. Specifically, what social referencing addsto joint attention isthat it also involves the learner reacting to the novel stimulus in a manner that is in accor-dance with the others expression (Pelez-Nogueras & Gewirtz, 1997). Consider again theprevious example of the three-year-olds surprise when the kitten runs into the room. Aspart of her joint attention skills, the child looks up at her mother while pointing to the

kitten, but then she engages in social referencing when she sees her mother make a fearfulface and as a result avoids approaching the kitten.

As well as incorporating the component of concordant responding (in other words,using the reactions of others as discriminative stimuli for ones own responding), socialreferencing also appears to comprise an emotional component. In other words, it extendsbeyond the simple sharing of information and also facilitates the learners emotional reac-tion to stimuli. Tis emotional aspect of social referencing appears to make up a four-stageprocess that involves recognizing emotional expressions, understanding emotional expres-sions, responding to emotional expressions as cues, and altering behavior in accordance

with changes in emotional expression.Cognitive-developmental psychologists view the informational and emotional com-

ponents of social referencing as separate processes. Specifically, they distinguish betweeninstrumentalsocial referencing, which involves the learners use of knowledge from othersas indicators of how to understand stimuli (Feinman, 1982), and affectivesocial refer-encing, which involves the learners use of others emotional facial expressions to deter-mine how to feel about ambiguous events (Klinnert, Campos, Sorce, Emde, & Svejda,1983).

Learning theorists such as myself, however, have argued that both types of socialreferencing are outcomes of the same conditioning process, because the cues that conveyaffective components also contain instrumental information, and vice versa. Te studyby Gewirtz and Pelez-Nogueras (1992) provided some empirical support for this view,as well as examples of how the emotional aspects of social referencing can be explicitlytrained.

In contexts of ambiguity, we identified two originally meaningless maternal facialexpressions and then trained them with standard conditioning procedures to denoteopposite consequences for responses where infants reached for objects. Hence, one mater-

nal hand-to-face expression was trained to predict positiveauditory-kinetic consequencesof the infant reaching for ambiguous objects (see figure 4.3), while the other maternalhand-to-face expression was trained to predict negativeauditory-kinetic consequences ofthe infant reaching for ambiguous objects (see figure 4.4).

the incorporation of a technology based on derived stimulus relations into learning cur-ricula for individuals with developmental disabilities would seem to hold great promise inhelping such individuals acquire functional and meaningful goals.

Te present book serves as a compilation of instructional strategies based on decadesof basic and applied research on derived stimulus relations from prominent, world-renowned researchers who attest to different theoretical frameworks. Te book is intended

for parents and a variety of professionals working with individuals with autism and otherdevelopmental disabilities. Tese professionals include but are not limited to teachers,developmental therapists, adult service providers, speech-language pathologists, andbehavior analysts, all of whom have some basic understanding of the principles of appliedbehavior analysis.

Te book is divided into three parts. Part 1 focuses on the establishment of prerequisiteskills necessary for individuals to participate meaningfully in a curriculum based upon,or including components of, derived relational responding. Part 2 emphasizes instructionthat will lead to the production of such intermediate skills as naming, reading, spelling,

and requesting. Part 3 aims to help the practitioner establish more complex skills in learn-ers, including perspective taking and empathy, higher-order intelligence, and mathemati-cal competence. Each chapter contains a variety of practitioner tools, such as sample datasheets, step-by-step instructions, training notes, and problem-solving strategies. Te readerneed not work through the entire book for it to be of value. Some learners may be moreappropriate candidates for the strategies and techniques presented in one or more parts ofthe book only. Tus, the chapters can be used in isolation or in combination with otherchapters, depending on the particular learners educational needs. It is also not necessaryfor the reader to be committed to one particular theory regarding derived stimulus rela-

tions or verbal behavior, since the chapters represent an eclectic mix of theoretical orienta-tions. Rather, our intention is that the strategies in this book can be incorporated, if notmade the basis of, educational curricula for learners with mild or significant communica-tion and intellectual deficits due to autism, mental retardation, or other developmentaldisabilities. As a result, we hope that practitioners and their clients will benefit from thematerial presented in this book, and that future years will see the implementation of thistechnology in schools, clinics, and habilitation settings around the world.

We wish to thank Anna Neises and Char Burrell for e ditorial assistance, and wegratefully acknowledge the many contributors to this book for their thoughtful and per-

sistent work. Many of the authors who contributed to this book are our long-standingcolleagues, friends, and mentors and we have been proud to be associated with theircreativity and expertise. In this vein, we also acknowledge the many researchers, researchassistants, students, and research participants whose work over many years has been aninspiration for this book.

Ruth Anne Rehfeldt, Ph.D., BCBA Yvonne Barnes-Holmes, Ph.D.Southern Illinois University National University of Ireland,Carbondale, Illinois Maynooth Ireland

References

Cowley, B. J., Green, G., & Braunling-McMorrow, D. (1992). Using stimulus equiva-lence procedures to teach name-face matching to adults with brain injuries.Journal of

Applied Behavior Analysis, 25,461475.

Hanna, E. S., de Souza, D. G., de Rose, J. C., & Fonseca, M. (2004). Effects of delayed

constructed-response identity matching on spelling of dictated words. Journal ofApplied Behavior Analysis, 37,223228.

Lynch, D. C., & Cuvo, A. J. (1995). Stimulus equivalence instruction of fraction-decimalrelations.Journal of Applied Behavior Analysis, 28, 115126.

Rosales, R., & Rehfeldt, R. A. (2007). Contriving transitive conditioned establishingoperations to establish derived manding skills in adults with severe developmentaldisabilities.Journal of Applied Behavior Analysis, 40, 105121.

Sidman, M. (1971). Reading and auditory-visual equivalences. Journal of Speech andHearing Research, 14,513.

. (1977). eaching some basic prerequisites for reading. In P. Mittler (Ed.), Researchto practice in mental retardation: Vol. 2. Education and training. Baltimore, MD:University Park Press.

Sidman, M., & Cresson, O. (1973). Reading and crossmodal transfer of stimulus equiva-lences in severe retardation.American Journal of Mental Deficiency, 77,515523.

7/21/2019 1 Pdfsam Derived Responding

10/26

PAR 1

Establishing the Prerequisitesfor Normal Language

face lights up with surprise and pleasure. However, her next action is not to engage thekitten in play, but to look up at her mothers face while pointing to the kitten, to see if hermother had also witnessed the animals dramatic entrance. Gaze shifts may subsequentlybe combined with gestures toward the object within the visual field of the familiar face.

Behavioral researchers (including myself) have proposed that an operant (rather thanage-based) history guides the emergence of t he skills of joint attention (Dube, MacDonald,

Mansfield, Holcomb, & Ahearn, 2004; Holth, 2005). Specifically, these behaviors nor-mally result from environmental contingencies that operate during early mother-childverbal and gestural communications (Pelez, Gewirtz, & Wong, 2007). From this per-spective, gaze shifts in joint attention incorporate (1) the selective effects of environmentalstimuli that set the occasion for the response class, (2) stimuli that support joint attentionbehavioral chains in dual roles as discriminative and reinforcing stimuli, (3) the conse-quences that lead to the choice of experiencing a stimulus together with the adult versusexperiencing it independently of the adult, and (4) relevant and plausible environmentalconditioning histories. Te analysis also identifies the function of reinforcers and suggests

various classes of socially mediated stimuli that maintain joint attention behavior. Indeed,the most common function of the reinforcers appears to be face-to face interactions withan adult (Pelez-Nogueras, Field, Hossain, & Pickens, 1996). Put simply, reinforcers areinitially produced by the activity related to the stimulus in question (for example, playingwith a toy) and then increased by adult-generalized social reinforcers such as vocalizationsand smiling, gestures of approval, or demonstrations of affection while engaged. In otherwords, it is often more reinforcing for a child to play with a toy or look at a book whenthe caregiver participates in the event than it is when the caregiver is absent.

Joint Attention Deficits in Autism

Interest in the concept of joint attention has increased because of its putative role indevelopmental disabilities (Carpenter, Pennington, & Rogers, 2002), and Dawson andcolleagues (2004) have even argued that joint attention deficits alone can differentiatebetween normally developing learners and those with autism. Specifically, learners withautism appear to lack prerequisites for joint attention that include orienting to speechsounds and other social stimuli (for example, when someone points) and show more

direct evidence of deficiencies in joint attention behaviors. For example, Charman andcolleagues (1997) demonstrated that children with autism looked at a mechanical toywhen it was activated but did not exhibit ga ze switches between the toy and an adult whowas present.

Deficits in joint attention have also been associated with abnormalities in languagedevelopment (Mundy, Sigman, Ungerer, & Sherman, 1986). Specifically, in children withautism, correlations have been recorded between low frequencies of adult-object gazeswitching at twenty months, limited language gains, and diminished social communica-tion at forty-two months (Charman et al., 1997). One explanation for the relationshipbetween language and joint attention suggests that the rapid vocabulary expansion oftypical preschool development depends in part on the learners ability to determine, viaobservation of adult-attending stimuli, which object in the immediate environment isrelated to the adults speech.

According to this view, joint attention should warrant a potentially i mportant placein early intervention programs, especially those aimed at establishing critical languageprerequisites. And yet it is not often given such an important place in these programs.Indeed, despite its pivotal developmental significance, there are few effective interventionsfor ameliorating deficits in joint attention in the literature. Te section below sets out atraining sequence for this purpose that may be used with very young learners or thosewith developmental delay.

A Protocol for Esta blishing Joint Attention

Te section below contains a description of the key components of a protocol forestablishing joint attention, followed by empirical evidence to support the use of trainingregimes such as these in children with autism.

Establishing social reinforcers. Te training of joint attention critically requires theexistence of social stimuli, such as nods or smiles, as reinforcers at an early age. Tis can

be accomplished with the teacher and learner sitting face-to-face, with ten edible reinforc-ers spread across the table between them. Block any attempts to remove the reinforcersfrom the table until the learner is sitting quietly; then nod and smile before allowing thelearner to take one. It is important to emphasize that the learner is onlyallowed to take areinforcer when the teacher nods and/or smiles (in order to make these gestures functionas discriminative stimuli). In addition, you should emit an occasional verbal cue, such asyes or Look at that to further improve the learners general communication skills.

Of course, this type of training may lead to the possibility that nods and smiles func-tion as conditioned reinforcers only when treats are available but fail to do so in othersituations. Naturally, this would mean that the learner may not recognize the nods andsmiles of other adults in other contexts. But this situation seems unlikely, or at least rela-tively easily rectified.

Gaze following. When teaching a child the skill of gaze following, the teacher andlearner should again sit at opposite ends of a small table. First, show the learner a rein-forcer of choice, and then ask her to turn around while you place the reinforcer under oneof two opaque cups. Ten say, ready, and allow her to turn around again to observe thecups. Ask her to point to the cup that she thinks contains the treat. Lift the chosen cup,and if the treat is there the learner can have it. If the empty cup has been selected, simplyremove the treat and start again.

On a subsequent trial, place your face close to the cup with the treat while maintain-ing eye contact with the learner, such that she comes to rely on this cue for discriminatingthe cup that holds the treat. Continue with this type of training until the learner looksat your face and consistently chooses the cup with the treat. Next, across trials, fade outyour proximity to the cup, so that eventually the learner can choose the right cup afteronly a brief glance on your part.

Learners such as those with developmental disorders may experience difficulty simplyattending to others faces. In this case, getting the learner to attend to your face, evenwhen it is near the cup, will be difficult. In such a situation, it is possible to establish thisskill by saying the learners name, holding the treat up to your eyes, and then tracing

7/21/2019 1 Pdfsam Derived Responding

11/26

shaping the touch-screen response by modeling, prompting, and signaling appropriateresponding. A familiar assistant may perform these functions for a learner who is severelydisabled. Tree stimuli that appear on the screen are directly trained: a picture of anapple (A), a picture of a baby (D), and a sketch of a bear in a box (G). Each stimulus ispresented in a separate block of trials.

Stage 2: Fading. During this phase of training, comparison stimuli should be gradu-ally faded in, so that each trial begins to more closely resemble the MS format. Tat is,while A, D, or G appear, two comparison stimuli also gradually appear (fade in) until allthree stimuli are clearly visible and the learner can select or point to the target stimulus.Pointing is reinforced by contingent stimulation (movement and sounds coming from t hestimulus, while the mother or assistant also touches and praises the learner).

Stage 3: Discrimination tr