Document 9933

Transcript

Volume V Issue Number 1 ISSN 1932 - 4731 Table of Contents Pg.: 1: Generic Instruction versus Intensive Tact Instruction and the Emission of Spontaneous Speech - R. Douglas Greer & Lin Du Pg.: 20: Effects of Multiple Exemplar Instruction on the Transformation of Stimulus Function Across Written and Vocal Spelling Responses by Students with Autism - Carly M. Eby, R. Douglas Greer, Lisa D. Tullo, Katherine A. Baker & Rebecca Pauly Pg.: 32: .AAC Interventions: Case study of In-utero Stroke. Cindy Geise Arroyo, Robert Goldfarb, Danielle Cahill, & Janet Schoepflin Pg.: 48: Implications of Skinner’s Verbal Behavior for Studying Dementia Jeffrey A. Buchanan, Daniel Houlihan, & Peter J.N. Linnerooth Pg.: 59: Using Skinner’s Model of Verbal Behavior Analysis to study Aggression in Psychiatric Hospitals - Michael Daffern & Matthew Tonkin Pg.: 70: Evaluation of Two Communicative Response Modalities for a Child with Autism and Self-Injury - Stacy E. Danov, Ellie Hartman, Jennifer J. McComas, & Frank J. Symons Pg.: 80: Using DRO, Behavioral Momentum, and Self-Regulation to Reduce Scripting by an Adolescent with Autism - Vanessa Ann Silla-Zaleski & Mary J. Vesloski The Journal of Speech - Language Pathology and Applied Behavior Analysis VOLUME NO. 5, ISSUE NO. 1 ISSN: 1932 - 4731 Published: January 12, 2010 Publisher’s Statement The Journal of Speech-Language Pathology and Applied Behavior Analysis (JSLP-ABA) is published by Dr. Joseph Cautilli and BAO Journals. It is a peer-reviewed, electronic journal intended for general circulation in the scientific community. The mission of this journal is to provide a forum for SLP and ABA professionals to exchange information on topics of mutual interest. These topics may include, but are not necessarily limited to support for disorders of prelinguistic communication, speech perception/production, oral language and literacy, speech fluency, and voice. They may also address issues pertaining to accent reduction, culturally-based language variations, and augmentative-alternative communication. JSLP-ABA welcomes articles describing assessment and treatment efficacy research based on detailed case studies, single -subject designs, and group designs. Also encouraged are literature reviews that synthesize a body of information, highlight areas in need of further research, or reconsider previous information in a new light. Additionally, this journal welcomes papers describing theoretical frameworks and papers that address issues pertaining to SLP-ABA collaboration. All materials, articles, and information published in JSLP-ABA are peer-reviewed by the review board of JSLP-ABA for informational purposes only. The information contained in this journal is not intended to create any type of patient-therapist relationship or representation whatsoever. To receive a free subscription to JSLP-ABA, please send an e-mail to [email protected]. Include your name and the email address in the body of your e-mail; and type “subscribe-SLP-ABA” in the subject field. When your email is received, your name will be added to the subscription list. You will then automatically receive notice of publication of each new issue through an e-mail containing a hyperlink to the latest issue. All rights are reserved. The Journal of Speech – Language Pathology and Applied Behavior Analysis may be freely accessed, downloaded, and distributed free of charge. If you wish to sell our journals or charge a fee for access to our journals, you need to obtain the express prior written permission of the copyright holder. For uses requiring permission, contact Joseph Cautilli, Ph.D., BCBA. All information contained within is provided as is. The SLP-ABA journal, its publisher, authors, and agents, cannot be held responsible for the way this information is used or applied. The Journal is not responsible for typographical errors. Mission Statement The mission of the Journal of Speech-Language Pathology and Applied Behavior Analysis (JSLPABA) is to provide a forum for SLP and ABA professionals to exchange information on topics of mutual interest. These topics may include (but are not necessarily limited to) support for disorders of prelinguistic communication, speech perception/production, oral language and literacy, speech fluency, and voice. They may also address issues related to accent reduction, culturally based language variations and augmentativealternative communication. JSLP-ABA welcomes articles describing assessment and treatment efficacy data based on detailed case studies, single -subject research design, and group designs. Also encouraged are literature reviews that synthesize a body of information, highlight areas in need of further research, or reconsider previous information in a new light. Additionally, this journal welcomes papers describing theoretical frameworks and papers that address issues pertaining to SLP-ABA collaboration. JSLP-ABA is viewed as a primary source of information for speech-language pathology (SLP) professionals and professionals in applied behavior analysis (ABA) who support individuals of all ages with communicative disorders. The contents of this journal are intended to meet the interest of these professionals for information to support evidence-based practice. JSLP -ABA is also intended to serve as a vehicle to encourage collaboration between these SLP and ABA professionals. Submission Information for Authors Overview All papers must be submitted in RTF or MS Word DOC format to the Lead Editor (Dr. Joseph Cautilli) via e-mail at [email protected]. Papers may be submitted at the initiative of an author or in response to an invitation from the Lead or Associate Editors. All submissions are peer-reviewed and must be accompanied by a signed Assignment of Rights (AOR) form. A link to the AOR form is at the bottom of this page. After peer review and follow-up, all articles are copyedited. Authors have an opportunity to review and approve their manuscript prior to publication. Once approved, authors are responsible for all statements made in their work, including changes made by the copy editor prior to approval. Content To be considered for publication, articles must address topics of mutual interest to SLP and ABA professionals. These topics may include (but are not necessarily limited to) support for disorders of prelinguistic communication, speech perception/production, oral language and literacy, speech fluency, and voice. They may also address issues related to accent reduction, culturally-based language variations and augmentative-alternative communication, SLP-ABA collaboration. Articles may report original research, descriptions of theoretical frameworks, literature reviews, treatment critiques, and tutorials. Peer Review Process All submitted manuscripts are reviewed initially by the Lead Editor. Manuscripts with insufficient priority for publication will be rejected promptly. Other manuscripts will be sent to the Senior Associate Editor, who will distribute them to editorial consultants with relevant expertise. The editorial consultants will read the papers and evaluate (1) the importance of the topic addressed by the paper; (2) the paper’s conformity to standards of evidence and scholarship; and (3) the cla rity of writing style. Comments provided by the editorial consultants will then be provided to the author(s) for follow up. Formatting Requirements: To support the electronic copy-editing process, authors must honor all of the following guidelines: • The page set-up for manuscripts must be set for 1-inch margins on all 4 boarders. • All pages must be in portrait orientation. There can be no pages in landscape orientation. • Manuscripts must be typed in single -spacing using size 11 “Times New Roman” Font. • Manuscripts must be submitted as one continuous document rather than in sections or subdocuments. • Each manuscript must include 7 elements in the following order: title, name(s) of author(s), abstract, key words, body, references, author(s)’ contact information. • Do not insert pagination, headers, or footers. (These are inserted in the copy-editing process) • • • • • • • • • The use of headings is encouraged and should be structured according to the guidelines described in the Publication Manual of the American Psychological Association (5th edition). If graphics, figures and tables are used, they must be created in *.jpg or *.bmp format. No Excel graphs will be accepted. Graphics, figures, and tables, if used, may be embedded in the body of the manuscript or they may be submitted in a separate MS Word document. If the latter option is chosen, then author(s) must indicate clearly the intended location of each item (graphic, figure, table) within the manuscript so that the copy editor can make the insertions. Individual graphics, figures, and tables, when used, may not be larger than one page. The caption for a table must be printed above the table. The caption for a figure must be printed below the figure. In the references section, please use italics where APA style would allow underlining (e.g., the titles of journals and books). Author contact information must include the following 4 elements for each author: name, mailing address, phone, and e-mail. Manuscripts must be saved and submitted in MS Word “DOC” format. When there is a conflict between the requirements of APA style (see below) and the formatting rules listed here, the formatting rules will supersede the APA requirements. Manuscript Style Requirements: • • • • With the exception of the above (formatting) guidelines, authors must write their manuscripts in a style that is consistent with the Publication Manual of the American Psychological Association (APA Manual) (5th edition). A copy of this manual may be ordered at http://www.apastyle.org/ Consistent with APA style, authors must use non-sexist language. Please refer to Table 2.1 in the APA Manual for “Guidelines for Unbiased Language.” Also consistent with APA style, authors must use person-first language for referring to individuals with potentially stigmatizing characteristics. Person-first language requires an author to name the individual first, followed by descriptive information (e.g., "child with autism") rather than to use an adjectival form (i.e., "autistic child") or a nominal form (i.e., "the autistic"). As noted above: When there is a conflict between the requirements of APA style and the formatting rules listed in the above section, the formatting rules will supersede the APA requirements. General Guidelines for Preparing Abstracts: The following general guidelines must be honored to insure that JSLP-ABA will be accepted into the major psych databases. (See PsychINFO website: http://www.apa.org/psycinfo/about/covinfo.html) • • • • • • An abstract may not exceed 960 characters and spaces (approximately 120 words). Characters can be conserved by using digits for numbers (except at the beginning of sentences); by using well-known abbreviations; and by using the active voice. Begin the abstract with the most important information, but don’t repeat the title. Include only the four or five most important concepts, findings, or implications. Embed as many key words and phrases in the abstract as possible. Include in the abstract only information that appears in the body of the manuscript. For the sake of clarity, define all acronyms and abbreviations except for measurements; spell out the names of tests; use generic names for drugs (when possible); and define unique terms. • • Use the present tense to describe results with continuing applicability or conclusions drawn and the past tense to describe variables manipulated or tests applied. As much as possible, use the third person rather than the first person. Abstracts for Empirical Studies: Abstracts for empirical studies are also generally about 100 to 120 words in length. They should include the following information: • Problem under investigation (in one sentence) • Pertinent characteristics of participants (e.g., number, type, age, sex, genus and species) • Experimental method, including apparatus, data-gathering procedures, and complete test • Names and complete generic names and dosage and routes of administration of any drugs (particularly if the drugs are novel or important to the study) • Findings, including statistical significance levels • Conclusions and implications or applications Abstracts for Literature Reviews and Theoretical Articles Abstracts for review or theoretical articles are generally about 75 to 100 words in length, and they include the following information: • The topic (in one sentence) • The purpose, thesis, or organizing construct and the scope (comprehensive or selective) of the article • Sources used (e.g., personal observation, published literature) • Conclusions Thank you! The Behavior Analyst Online Journals Department ADVERTISEMENT ADVERTISING IN BAO Journals If you wish to place an advertisement in any of our journals, you can do it by contacting us. The prices for advertising in one issue are as follows: 1/4 Page: $50.00 1/2 Page: $100.00 Full Page: $200.00 If you wish to run the same ad in multiple issues for the year, you are eligible for the following discount: 1/4 Pg.: $40 - per issue 1/2 Pg.: $75 - per issue Full Page: $150.00-per issue An additional one time layout/composition fee of $25.00 is applicable In addition to placing your ad in the journal(s) of your choice, we will place your ad on our website’s advertising section. For more information, or place an ad, contact Halina Dziewolska by phone at (215) 462-6737 or e-mail at: [email protected] The Journal of Speech - Language Pathology and Applied Behavior Analysis ISSN: 1932-4731 Editorial Staff Editor In Chief Joe Cautilli, Ph.D., BCBA Senior Associate Editor Mareile Koenig, Ph.D., CCC-SLP, BCBA Associate Editors Leslie Cohen, Ph.D. Joanne Gerenser, Ph.D., CCC-SLP Elizabeth Grillo, Ph.D., CCC-SLP Caio Miguel, Ph.D. Editorial Board Christine Barthold, Ph.D., BCBA Vince Carbone, Ph.D. Jenn Cronin, M.Ed. Brian Cowley, Ph.D., BCBA Kathy Dyer, Ph.D., CCC-SLP, BCBA Anntonette Falco, M.Ed. Lori Frost, MS, CCC-SLP Cheryl Smith Gabig, Ph.D., CCC-SLP Cheryl Gunter, Ph.D., CCC-SLP James Halle, Ph.D. Giri Hegde, Ph.D., CCC-SLP Anne Holmes, Ph.D., CCC-SLP, BCBA Laura Hutt, M.S., SLP -CCC, BCBA James Luiselli, Ph.D. Hedda Meadon, Ph.D. Pat Mirenda, Ph.D., BCBA Pete Peterson, Ph.D. Anna I. Petursdottir, Ph.D. David Sidener, Ph.D. Francois Tonneau, Ph.D. Michael Weinberg, Ph.D., BCBA Mary Jane Weiss, Ph.D., BCBA Ray Weitzman, Ph.D. Notice to Readers Due to technical difficulties, it has become necessary for us to create a mirror site to access all editions of BAO journals published since the loss of our webmaster, Craig Thomas . We are working to correct all the problems with the old site and we appreciate your patience. We are currently constructing a new, fully functional BAO journal website with the features of the old BAO site. Until further notice, you may access the current and future BAO journals at www.BAOJournal.com. Please bookmark this location or put it in your browser’s “Favorites” for future access. A link on the new BAO Journal site will take you to the old BAO site where all past issues are currently accessible. Thank you for your loyal support for BAO and its journals. Cordially, Joe Cautilli and BAO Journals SLP-ABA Volume 5, No. 1 Generic Instruction versus Intensive Tact Instruction and the Emission of Spontaneous Speech R. Douglas Greer and Lin Du Abstract We isolated the effects of intensive tact instruction from increased generic instruction on the emission of spontaneous speech (pure mands, pure tacts, intraverbals) in non-instructional settings by 3 boys with ASD, using a delayed multiple probe design across participants. The teaching procedure included the replacement of 100 generic academic learn units with 100 learn units of tact instruction until mastery of 5 sets of 4 stimuli (20 pictures). Results showed a strong functional relationship between intensive tact instruction and the participants’ production of pure mands and pure tacts in non-instructional settings. Findings suggest that it is not an increase in instruction alone but the nature of intensive tact instruction that results in more verbal operants in non-instructional settings. Keywords: intensive tact instruction, verbal operants, tacts, mands, learn units Introduction The expansion of the tact repertoire appears to be critical to the acquisition of subsequent and more complex verbal developmental stages. It is likely that direct instruction in tacts must continue until children have acquired the verbal developmental cusp of Naming, which is defined as the capability to acquire the listener and speaker responses to stimuli incidenta lly (Greer & Speckman, 2009; Greer, Stolfi, & Pistoljevic, 2007; Horne & Lowe, 1996; Pistoljevic & Greer, 2006; Skinner, 1957). Skinner (1957) described six elementary verbal functions for the speaker, including (1) echoics, (2) mands, (3) tacts, (4) autoclitics, (5) intraverbals, and (6) textual responding. For the purposes of this study, we focus primarily on mands, tacts, and intraverbals. Mands are speaker or substitute speaking topographies that are emitted under conditions of deprivation or annoying conditions in the presence of a listener who then mediates the environment for the speaker (Skinner, 1957; Greer & Ross, 2008). Tacts are verbal operants occasioned by a discriminative stimulus in the environment and are reinforced by generalized reinforcers particularly attention (Tsiouri & Greer, 2003). Children with autism usually demonstrate deficits in functional communication (DSM-IV, American Psychiatric Association, 1994) , including tacts. For example, children with autism typically do not emit language without prompting or questioning by others and thus demonstrate few instances of “spontaneous speech” (Krantz & McClannahan, 1998). For this reason, studies have focused on the development of verbal skills by these children. Partington, Sundberg, Newhouse, & Spengler (1994) implemented a procedure for transferring stimulus control from verbal stimuli to nonverbal stimuli in a six-year-old girl with autism. As a result, the girl acquired tact responses to 19 stimuli. Partington et al. then proposed that the reason for the failure of children with autism to acquire tacts may be tied to the antecedent verbal behavior of others (e.g., “What is that?”) which functions to block the establishment of 1 SLP-ABA Volume 5, No. 1 stimulus control by a nonverbal stimulus. This observation is consistent with findings reported by Williams and Greer (1993). Several evidence-based protocols are available for inducing echoics and then mand and tact responses for children who have no, or limited, vocal verbal operants. Included here are (1) the stimulusstimulus pairing procedure to induce echoics (Sundberg, Michael, Partington, & Sundberg, 1996), (2) rapid motor imitatio n to induce echoics to mand and echoics to tact functions (Ross & Greer, 2003; Tsiouri, & Greer, 2003), and (3) direct echoic to mand and echoic to tact instruction (Williams & Greer, 1993). Procedures are also available to expand repertoires such as speaker immersion (Greer & Ross, 2004; Ross, Nuzzolo, Stolfi, & Natarelli, 2006; Schwartz, 1994). However, once repertoires are established, children need extensive instruction in order to expand their language functions or verbal repertoires. The primary repertoire in need of expansion is the tact repertoire because tacts are foundational to the subsequent development of more complex verbal behavior such as Naming, say-do correspondence, conversational units, and age-appropriate self-talk (Barnes-Holmes, Barnes-Holmes, & Cullinan, 2000; Greer & Ross, 2008; Greer & Speckman, 2009; Horne & Lowe, 1996). Several recent investigations suggested the importance of directly and intensely teaching the tact repertoire, especially for those children who cannot acquire tacts through spontaneous observational learning (Williams & Greer, 1993; Ross & Greer, 2003; Schauffler, & Greer, 2006; Tsiouri & Greer, 2003; Pistoljevic & Greer, 2006; Pistoljevic, 2008). The protocol used in these studies is referred to as the intensive tact procedure (ITP). Using this procedure children received at least 100 instructional trials (learn units) devoted to tacting visual stimuli, in addition to their general educational instruction. All of these studies found that the children significantly increased their “spontaneous vocal verbal behavior in social settings”. In all of the studies the vocal verbal responses that were emitted in the non-instructional settings (NIS) differed from those taught during the ITP, suggesting that ITP affected the reinforcing effects of emitting tacts. Pistoljevic and Greer (2006) tested the effects of ITP on the number of vocal verbal operants emitted in NIS (transition time, lunch, and free play) by three preschool boys with autism who, prior to intervention, emitted low numbers of pure (spontaneous) tacts and mands. In these studies, the participants received 100 additional learn units daily during the ITP phase while learn units of the generic educational programs remained constant during baseline. Results showed that the participants’ verbal operants (tacts and other verbal behaviors) increased in NIS. However, Pistoljevic and Greer noted that the increase may have occurred simply as a result of increasing the number of learn units received by the participants, regardless of whether the learn units were based on ITP or any other generic educational program. They proposed that future research should be designed to isolate the increased tact instruction from increased instruction of any kind. Schauffler & Greer’s study (2006) found that after the implementation of ITP, the audience-accurate tacts and conversational units increased significantly for two middle school students, while inaccurate tacts/conversational units decreased for one student. They also found that the participants emitted accurate tacts not taught during ITP, suggesting that ITP taught audience control. That is, rather than emitting language that was inappropriate for school settings, the students emitted audience-appropriate language. In a recent dissertation, Pistoljevic (2008) found that preschool participants who mastered several sets of tacts using the ITP not only emitted more verbal operant but also acquired Naming. 2 SLP-ABA Volume 5, No. 1 Prior research reported that by increasing the numbers of learn units presentations one may significantly increase the instructional objectives and educational standards achieved by students (Albers & Greer, 1991; Greer, McCorkle & Williams, 1989; Ingham & Greer, 1992; Kelly, 1997; Lamn & Greer, 1991; Selinske, Greer, & Lodhi, 1990). It is possible then that the results of studies of ITP could be attributed to the increase in learn units regardless of whether the learn units targeted tacts or other educational goals. None of the prior studies isolated the increased tact instruction from increased instruction of any other kind (Schauffler & Greer, 2006; Pistoljevic & Greer, 2006). Therefore, it is possible that simply increasing the numbers of generic learn units that students receive daily this will also affect the emission of pure tacts and pure mands in NIS. For example, it is possible that increasing the number of learn units to target the identification of body parts, counting, number-object correspondence, visual matching, or textual responses can lead to the same or similar outcomes as those attributed to the ITP. To assess whether an increase in spontaneous tacting is due to an increase in generic learn units or to an increase in tact-specific learn units, it is necessary to control for the number of learn units in each condition. In other words, instead of having fewer learn units in generic instruction than in ITP, baseline conditions should include similar numbers of learn units in each condition to control for the effects of instruction. In the present study, we compared the effects of generic instruction versus ITP on the emission of tacts in NIS by three participants. The number of learn units was equivalent in each condition, allowing for the isolation of the tact component that is present in ITP and missing from generic instructional programs. Method Participants Three male children diagnosed with Autism Spectrum Disorders served as participants in this study. These participants were chosen for this study because they had limited tacts in their repertoire and because they did not emit many pure tacts and mands throughout the day, especially in NIS (transition time, free play, lunch). Each participant’s verbal behavior achievement levels and standardized measures at the beginning of this study are summarized in Table 1.The table indicates that Participant A was an 8-year-old male who functioned at emergent listener/emergent speaker levels of verbal behavior. According to the Preschool Inventory of Repertoires for Kindergarten (PIRK) (Greer & McCorkle, 2003) and assessments of verbal development described in Greer and Ross (2008) which were conducted prior to the experiment, the following verbal developmental cusps and capabilities were not in his repertoire: self-talk, say-do correspondence, and Naming (the capability to acquire language incidentally). He was receiving instruction devoted to conditioning books and toys as reinforcers for observing responses and preference in free play, mands for non-visible items, and dictation. When assessed using the Preschool Language Scale- Fourth Edition (PLS-4) (Zimmerman, Steiner, and Pond, 2002), he scored 50 in total language and 50 in expressive communication. Participant B was a 5-year-old male who functioned at the emergent listener/speaker levels of verbal behavior. According to the PIRK (Greer & McCorkle, 2003) and assessments of verbal behavior (Greer & Ross, 2008), appropriate self-talk, say-do correspondence, auditory matching and spontaneous incidental learning were not present within his repertoire. His program included procedures to teach vocal direction following, to 3 SLP-ABA Volume 5, No. 1 condition books and toys as reinforcers for observing responses and preference in free play, and to teach the use of mands for accessing non-visible items. He scored 61 in expressive communication on the PLS-4. Participant C was a 3-year-old male who functioned at the emergent listener/speaker level of verbal behavior. According to the PIRK (Greer & McCorkle, 2008) conducted prior to the study, the following capabilities were not within his repertoire: self-talk, say-do, auditory matching and spontaneous incidental language learning. He was being instructed in conditioning books and toys as reinforcers for observing responses and preferences in free play, mands for non-visible items, and dictation. His scores on the PLS-4 included a 55 in auditory comprehension, 57 in expressive communication and 51 in total language. Table 1. Description of the Three Participants 1 P Age A 8 B C 5 3 Diagnosis and Level Of Verbal Capability Standardized Test Scores -Autism 3 - emergent listener/emergent speaker -Total language : 50 -Autism 3 - emergent listener/speaker - Expressive Communication: 61 -Autism 3 -emergent listener/emergent speaker -Total language : 51 PLS-4 - Expressive communication: 50 PLS-4 PLS-4 - Expressive communication: 57 2 PIRK Repertoires Echoics, mands/tacts with autoclitic frames, matching and pointing/ listener literacy, visual tracking, generalized imitation, visual instruction control, capacity for sameness, and conditioned reinforcement for voices Echoics, mands/tacts with autoclitic frames, matching and pointing/ listener literacy, visual tracking, generalized imitation, conditioned reinforcement of books, conditioned reinforcement of voices, capacity for sameness, listener literacy Echoics, mands/tacts with autoclitic frames, matching and pointing/ listener literacy, visual tracking, generalized imitation, visual instruction control, capacity for sameness, and conditioned reinforcement for voices 1 P = Participant PIRK = Preschool Inventory of Repertoires for Kindergarten (Greer & McCorkle, 2003) 3 PLS-4 = Preschool Language Scale-Fourth Edition (Zimmerman, Steiner, and Pond, 2002) 2 All participants attended school in a suburb of New York City. Participants A and B were both enrolled in the same class of a public elementary school. The setting was a self-contained classroom with five students, one teacher and four teaching assistants. All students in the classroom had varying levels of verbal behaviors. Participant C attended a privately run and publicly funded preschool using the CABAS® (Comprehensive Application of Behavior Analysis to Schooling) educational model, which has 7 classrooms that vary in terms of the students' levels of verbal behavior (pre-listener/ speaker through speakers and emerging self-editors). The method of instruction employed and the measurement of the students' responses to instruction in both settings were based on the principles of applied behavior analysis. 4 SLP-ABA Volume 5, No. 1 Setting The experiment was conducted in the children’s classrooms and throughout the school environment. In Participant A’s and B's classroom, there were 5 tables, 12 adult chairs, and 6 child chairs. The classroom had two free play areas with rugs (located at different ends of the classroom), and an extensive assortment of toys, games, children’s books, and one plastic cottage playhouse. In one play area, there was a shelf with toys, books, blocks, and plastic animals. A TV with DVD and tape player were placed on the top of a cabinet facing this play area. In Participant C’s classroom, there were 4 tables, 8 adult chairs, and 7 child chairs. The toy area was approximately 2 x 2.5 meters and was located in the corner of the classroom, sectioned off by shelves holding books and toys. The study was conducted in several pla ces at the school. During the probes for pure tacts, mands, and intraverbals, data were collected in three NIS, including the toy areas of the classrooms, at the table during lunchtime, and in the hallways during the transition from the school buses. The intensive tact instruction took place in the toy area or other places in the classroom rather than at the traditional instructional table. At this time, other students received one-to-one instruction by other instructors in the classroom or they received speech, physical or occupational therapy in the therapists' rooms. Participants A and B ate their lunch in the school's lunchroom at a large rectangular shaped table near the windows. Their lunch started from 11:30 and ended at 12:00 everyday. For Participant C, the students ate their lunch inside the classroom at a large horseshoe shaped table near the door. Probes of the lunchtime responses were conducted when the participants finished their lunch. The probes in each NIS were conducted during 5-minute segments, and the three 5-minute sessions were blocked into a single 15-minute daily session. Dependent Variable The dependent variables in this study were pure mands, pure tacts, and intraverbals emitted by the participants in NIS (transition, play, lunch). Pure tacts were defined as verbal operants that made contact with the environment or identified components of the environment when there was no verbal antecedent, and were reinforced by generalized social reinforcers (Greer, 2002). Examples of the pure tacts emitted by the participants included "My backpack", and "It is raining". Pure mands were defined as verbal operants that specified their reinforcers under associated motivational conditions when there was no verbal antecedent (Greer, 2002). Examples of the pure mands emitted by the participants included "Help me, please", "Open, please" and "Lunch box, please". Intraverbals were defined as verbal operants that did not have a point-topoint correspondence with verbal stimuli and also occurred under verbal stimulus control. For example, a participant said “hi” in response to his teacher’s greeting (“Good morning”) while transitioning from the bus to the classroom. Independent Variable: The Intensive Tact Procedure The independent variable was an intensive tact procedure (ITP), consisting of 100 extra learn units of daily intensive tact instruction (ITI). Before the treatment, each participant’s mean number of daily learn units was calculated for the prior month. During baseline, each participant received 100 extra learn units to address 5 non-tact generic educational objectives. Specifically, participant A’s and B’s learn units were increased from 300 to 400, and Participant C’s learn units were increased from 350 to 450 5 SLP-ABA Volume 5, No. 1 during baseline. When ITI began, each participant’s total number of daily learn units remained the same as they were in baseline, but 100 of the non-tact generic academic learn units were replaced with 100 tact instruction learn units. ITI targeted 20 different tacts, each in response to a pictured stimulus. The stimuli represented 5 categories (community helpers, animals, food, instruments, and transportation) with 4 stimuli in each category. To establish that the participants did not know the tacts for these stimuli prior to the initiation of ITI, each participant was probed with one trial for each stimulus picture. The instructor presented a picture and (without saying anything) waited 2-3 seconds for the participant to make a response. None of the responses were consequated during the probe. Results of the probe demonstrated that the participants did not have any of the tacts for these stimuli in repertoire. For ITI, the stimuli were divided into five sets corresponding to each of the 5 categories. Within each category set, each of the 4 stimuli were represented by 5 different pictures. For example, within the set of instruments, there were 5 pictures of pianos, 5 pictures of triangles, and so on. This resulted in 20 pictures per set. As indicated above, participant received 100 tact learn units during ITI. A correct response was defined as the emission of a tact within 2 to 3 seconds following the presentation of a stimulus picture by the instructor. For example, when the instructor, without any vocal antecedent, presented the participant with a picture of a fox, a correct response was recorded if the participant emitted the vocal response “fox” within 2 to 3 seconds. The instructor then reinforced the participant by giving him his preferred edibles or vocal praise (i.e. “You are so smart! It is a fox!”). If the participant did not emit any vocal response or if he emitted one or more incorrect responses within 2-3 seconds, it was coded as incorrect and a correction procedure was initiated in which the teacher presented the picture of the fox again, said “fox,” and required the participant to echo the word “fox”. When the participant mastered the items in one or more categories (100% accuracy in one session or 90% in two consecutive sessions), the 100 learn units of ITI were devoted to the rest of the categories that were not yet mastered, which were rotated until the participant received 100 learn units of ITI. For example, if the participant met criterion on the category of food first, he still needed to receive 100 learn units of ITI, including 80 learn units for the remaining four categories plus 20 learn units for the mastered set of food. A day of post probe was conducted after the participant met criterion on a set. This same procedure was used in prior studies of the ITP (Pistoljevic & Greer, 2006; Pereira-Delgado & Oblak, 2007). Data Collection Design. A delayed multiple probe design across participants (Horner & Baer, 1978; Pistoljevic & Greer, 2006; Schauffler & Greer, 2006) was used to compare the number of verbal operants (tacts, mands, intraverbals) emitted by each participant in NIS (transition, play, lunch) before and after mastering each set of objectives in each of two conditions: (1) baseline, during which 5 sets of non-tact generic learning objectives were targeted, and (2) ITP intervention, during which 5 sets of tacts were targeted. This resulted 6 SLP-ABA Volume 5, No. 1 in 13 sets of probe data per participant, including 3 before baseline, 5 during baseline, and 5 during intervention. The specific sequence of procedures occurred as follows: (1) Pre-intervention probes were conducted by documenting the pure mands, pure tacts, and intraverbals emitted by each participant in NIS on 3 consecutive days prior to the initiation of baseline. (2) For each participant, the mean number of daily learn units and the mean number of learn-units-to-criterion were calculated for the month prior to the initiation of baseline. (3) During baseline, the number of learn units per participant was determined by adding 100 to the mean number of daily learn units the participant had received in the previous month. The extra 100 learn units were allocated equally to target 5 sets of pre-selected non-tact generic learning objectives. (4) Whenever a participant met criterion on a set of non-tact generic learning objectives, another probe was done in the NIS to assess pure mands, pure tacts, and intraverbals. This resulted in 5 sets of probe data during baseline. (5) After a participant achieved criterion on all 5 sets of non-tact learning objectives, the ITP was initiated. The 100 additional daily learn units that had been used during baseline to target 5 sets of non-tact generic learning objectives were replaced with 100 daily learn units of ITI to target 5 sets of tacts. (6) Whenever a participant met criterion on a set of tacts during ITI, another probe was done in the NIS to assess pure mands, pure tacts, and intraverbals. This resulted in 5 sets of probe data during the ITP. Probe Sessions. During all probe sessions, pure mands, pure tacts, and intraverbals emitted by the participants were transcribed word for word and then documented by counting the occurrences of each type of operant within the 5 minute interval. During the transitioning from the school bus to the classroom, the instructor started the timer when the participant stepped off the school bus and stopped the timer after 5-minutes. During the lunchtime probes, the instructor started the timer after the participant finished eating and concluded after 5-minutes. During the toy area probes, the instructor started the timer when the participant was playing with toys in the toy area of the classroom and concluded after 5 minutes. A pen and a data collection form were used to record the data, and a timer was used to determine elapsed time. A circled capital T next to a transcribed verbal operant classified the operant as a pure tact, a circled capital M indicated a pure mand, and a circled capital I indicated an intraverbal. At the end of each day, the numbers of pure tacts, pure mands and intraverbals collected in the three 5-minute segments were combined into a single session and totaled. Baseline Sessions. During baseline, the participants were presented with learn units for 5 sets of non-tact, generic learning objectives. As indicated above, the numbers of daily learn units were increased for each participant by 100 over the mean number of daily learn units received in the prior month. This was done to control for intensity of instruction across baseline and intervention phases. Since each instructional program consisted of 20-learn units, the addition of 100 learn units was accomplished by presenting one more session in each of the 5 non-tact programs that were selected from the three participants’ daily curricular instructions. Table 2 summarizes each participant’s 5 non-tact programs. Five short-term-objectives (STOs) were chosen for each of these five programs and they were grouped into different sets in sequence. For example, Participant A had to master each of the first STOs in his five selected programs (Edmark Reading objectives, textually responding to targeted sight words, math objectives, prepositions as 7 SLP-ABA Volume 5, No. 1 autoclitics, and writing picture descriptions) before he progressed to the next objectives. A plus sign (+) was recorded when the participant emitted a correct response, and a minus sign (-) was recorded if he emitted an incorrect response or no response. Mastery of a short-term-objective (STO) was defined as 90% accuracy across 2 consecutive sessions or 100% accuracy for 1 session. When the participant attained mastery in one or more programs in one set, the 100 learn units were devoted to the rest of the programs that were not met, which were rotated until the participant received 100 learn units in those five non-tact programs. The participant was not presented with a novel set of 5 categories until he met criterion on all 5 programs in the prior set. A day of post probe was conducted after the participant met criterion on the prior set. Table 2. Five non-tact programs selected for the three participants for instruction in baseline Programs Participant A Participant B Participant C Program 1 Edmark Reading objectives Calendar Action discrimination Program 2 Textually Responding to targeted Sight Words Letters Dolch Words Program 3 Math objectives Body parts identification Body parts identification Program 4 Prepositions as autoclitics Number identification Number identification Program 5 Writing Picture Descriptions Follow vocal directions 2-step motor imitation Intensive Tact Instruction (ITI). During ITI, data were recorded on the number of correct responses to the target stimuli. A plus sign (+) was recorded when the participant emitted a correct response, and a minus sign (-) was recorded if he emitted an incorrect response or no response. Mastery of a set (category) of tact stimuli was defined as 90% accuracy across 2 consecutive sessions or 100% accuracy for 1 session (equivalent to mastery of an STO in the generic instruction phase). After achieving criterion on one of the training sets, a new set (category) of tacts was introduced. Thus the procedures were the same for the ITI as for the non-tact generic instruction in the baseline. As indicated earlier, participants received intensive tact training on 5 sets of stimuli, one set at a time. Each set represented one of 5 categories: community helpers, animals, food, instruments and transportation. Each category included 4 stimuli. For example the category of ‘instruments’ included piano, cello, triangle, and organ, and each stimulus item was represented by 5 different pictures (e.g., 5 different pictures of a piano, 5 different pictures of a cello, etc.). This resulted in a total of 20 pictures per set. For each set, the 20 pictures were rotated systematically so that a participant received 20 learn units with equal numbers of response opportunities for each stimulus item within the set. Interobserver Agreement (IOA) IOA on verbal operants during probe sessions. IOA was conducted by measuring percent agreement between the data collected during NIS by the primary instructor and the data collected simultaneously and independently by the classroom teacher or a teaching assistant. All observers used 8 SLP-ABA Volume 5, No. 1 event recording. The classroom teacher and teaching assistants were trained through observations by the supervisor and one of the experimenters. Some of the probe sessions were also videotaped so that another independent observer could view the video and record the data at a later time. IOA for pure tacts, pure mands and intraverbals across all observations of Participants A and B was 100%. The mean IOA for the same operants produced by Participant C across all observations was 90% (range = 80% -100%). IOA during baseline (while targeting non-tact generic learning objectives). IOA during baseline was also measured by percent agreement. It was assessed by two independent observers throughout 17% of sessions for Participant A, 67% of sessions for Participant B, and 21% of sessions for Participant C. Each observer used the Teacher Performance Rate Accuracy Protocol (TPRA) (Ingham & Greer, 1992) which assesses both the accuracy of the measurement of the students’ responses and the fidelity of implementation of learn units. The results for Participants A, B, and C resulted in mean IOA values of 93% (range = 85% - 100%), 98%, (range = 95% -100%), and 93% (range = 85% - 100%), respectively. IOA during the ITP (when using ITI to target 5 categories of tacts). IOA during ITP was measured via percent agreement by two independent observers throughout 19% (43/227) of sessions for Participant A, 30% (56/187) of sessions for Participant B, and 24% (32/131) of sessions for Participant C. These observers also used the TPRA (Ingham & Greer, 1992) to assesses the accuracy of the measurement of the students’ responses and the fidelity of implementation of learn units. The mean IOA for learn unit accuracy during ITP for Participants A, B, C was 98% (range =95% - 100%), 100%, and 90% (range = 90% - 100%), respectively, across all sessions. Results Table 3 shows the number of tacts, mands, and intraverbals produced by each participant during the pre-baseline probes. Figure 1 shows the number of pure tacts, pure mands, and intraverbals produced by each participant during the 15-minute probes in NIS following mastery of sets targeted during non-tact instruction (NTI) and following mastery of sets targeted during ITI. The response patterns demonstrated by each individual participant are summarized below. For participant A, Table 3 shows a total of 0 tacts, 13 mands (i.e., 4, 5, & 4 respectively), and 0 intraverbals during the pre-baseline probe sessions. Figure 1 shows the following progression for verbal operants emitted in NIS following mastery of each NTI set: 0 tacts, 8 mands, and 1 intraverbal following Set 1; 0 tacts, 19 mands, and 1 intraverbal following Set 2; 0 tacts, 7 mands, and 1 intraverbal following Set 3; 0 tacts, 7 mands, and 2 intraverbals following set 4; and 0 tacts, 8 mands, and 2 intraverbals following Set 5. In contrast, Participant A emitted the following pattern of verbal operants in NIS following mastery of each ITI set: 3 tacts, 6 mands, and 2 intraverbals following Set 1; 4 tacts, 6 mands, and 2 intraverbals following Set 2; 6 tacts, 8 mands, and 2 intraverbals following Set 3; 17 tacts, 9 mands, and 1 intraverbal following Set 4; and 15 tacts, 7 mands, and 2 intraverbals following Set 5. For Participant B, Table 3 shows a total of 0 tacts, 1 mand and 3 intraverbals across 3 pre-baseline probe sessions. Figure 1 shows the following progression for verbal operants emitted in NIS during probes following the mastery of each NTI set: 2 tacts, 5 mands, and 0 intraverbal following Set 1; 0 tacts, 2 mands, and 0 intraverbal following Set 2; 2 tacts, 4 mands, and 0 intraverbal following Set 3; 3 tacts, 0 9 SLP-ABA Volume 5, No. 1 mands, and 0 intraverbals following set 4; and 0 tacts, 2 mands, and 0 intraverbals following Set 5. In contrast, Participant B emitted the following pattern of verbal operants in probes following mastery of each ITI set: 4 tacts, 7 mands, and 1 intraverbals following Set 1; 4 tacts, 9 mands, and 3 intraverbals following Set 2; 16 tacts, 4 mands, and 2 intraverbals following Set 3; 10 tacts, 7 mands, and 2 intraverbal following Set 4; and 15 tacts, 6 mands, and 2 intraverbals following Set 5. For Participant C, Table 3 shows a total of 9 tacts (i.e., 4, 2, & 3 respectively), 2 mands , and 0 intraverbals across 3 pre-baseline probe sessions. Figure 1 shows the following progression for verbal operants emitted in NIS during probes following the mastery of each baseline NTI set: 4 tacts, 5 mands, and 2 intraverbal following Set 1; 3 tacts, 5 mands, and 4 intraverbal following Set 2; 6 tacts, 3 mands, and 1 intraverbal following Set 3; 4 tacts, 8 mands, and 1 intraverbals following set 4; and 5 tacts, 18 mands, and 1 intraverbals following Set 5. In contrast, Participant C emitted the following pattern of verbal operants in NIS during probes following mastery of each ITI set: 10 tacts, 2 mands, and 2 intraverbals following Set 1; 14 tacts, 3 mands, and 3 intraverbals following Set 2; 18 tacts, 6 mands, and 1 intraverbals following Set 3; 12 tacts, 3 mands, and 1 intraverbal following Set 4; and 10 tacts, 4 mands, and 1 intraverbals following Set 5. Table 3. Three participants’ number of verbal operants during pre baseline probes Pre-Probes Verbal Operants Tacts 1 2 3 4 Participant A 0 Participant B 0 Participant C 4 Mands 4 0 0 Intraverbals 0 0 0 Tacts 0 0 2 Mands 5 0 1 Intraverbals 0 0 0 Tacts 0 0 3 Mands 4 0 1 Intraverbals 0 1 0 Tacts 0 Mands Intraverbals 1 2 Table 4 summarizes the number of each participant’s learn units to criterion for sets 1, 2, 3, 4, and 5 during Baseline (NTI) and for sets 1, 2, 3, 4, and 5 during the intensive tact procedure (using ITI). These data show that Participant A’s learn units to criteria for programs 1 through 5 were 80, 64, 96, 84, and 84 , respectively. Participant B’s learn units were 72, 92, 92, 80, and 76 respectively; and Participant C’s learn units were 100, 80, 80, 124, and 100, respectively. During the 5 ITI sets, we see that Participant A’s learn units to criteria were 200, 136, 236, 216, and 124, respectively. Participant B’s learn units to criterion were 124, 264, 144, 128, and 88, respectively, and Participant C’s learn units to criterion were 96, 128, 88,104, and 112, respectively. 10 SLP-ABA Volume 5, No. 1 For Participants A and B, mastery of sets during ITI required significantly more learn units than did mastery of sets during NTI. Participant A required 408 learn units (mean = 81.6) to master NTI sets compared with 912 (mean = 182.4) to master ITI sets. Participant B required 412 learn units (mean = 82.4) to master NTI sets compared with 748 (mean = 149.8) to master ITI sets, and Participant C required 484 learn units (mean = 96.8) to master NTI sets and 528 learn units (mean = 149.8) to master ITI sets. Figure 1.Verbal operants emitted by three participants during 15-minute probes in non-instructional settings following the mastery of 5 sets targeting non-tact objectives 5 sets of intensive tact instruction 11 SLP-ABA Volume 5, No. 1 Table 4. Three participants’ number of learn units to criterion for each set mastered. Treatment 2 (ITP) Baseline 1 (NTI) Conditions Set 1 Set 2 Set 3 Set 4 Set 5 1 2 TOTAL Set 1 Set 2 Set 3 Set 4 Set 5 TOTAL Participant A 80 64 96 84 84 Participant B 72 92 92 80 76 Participant C 100 80 80 124 100 408 412 484 200 136 236 216 124 124 264 144 128 88 96 128 88 104 112 912 748 528 NTI = Non-Tact Instruction ITP = Intensive Tact Procedure using Intensive Tact Training (ITI) Figures 2, 3, and 4, show the responses of Participants A, B, C to their intensive tact instruction. Figure 2 shows responses for Participant A, Figure 3 for Participant B, and Figure 4 for Participant C. These data show the mastery of each set for each participant. Discussion The results of this study demonstrated a functional relationship between intensive tact instruction and the emission of pure mands, pure tacts and intraverbals by three participants in non-instructional contexts. These findings are consistent with the results of Pistoljevic & Greer (2006), Schauffler & Greer (2006), Pereira-Delgado, & Oblak (2007), and Pistoljevic (2008) in that that the intensive tact instruction facilitated the acquisition of tacts and mands. The current study also addressed the question raised by Pistoljevic and Greer (2006) of whether the increased emission of verbal operants following intensive tact instruction was simply due to an increase in the number of learn units provided to participants or to the specific features of intensive tact instruction. The results of this study show that the emission of tacts in non-instructional settings was due to the intensive tact instruction and not simply to the addition of learn units. Increasing the numbers of learn units for the non-tact programs increased the number of pure mands, but not the number of pure tacts emitted by the participants in non-instructional settings. The increase in the number of pure tacts emitted in non-instructional settings may be explained in that tacts link to the recruitment of generalized reinforcement from adults. It is also possible that intensive tact instruction enhanced the reinforcement effect of social attention, which prior research has shown to be the critical control for the emission of tacts (Tsiouri & Greer, 2003; Pistoljevic, 2008). Like prior studies of the intensive tact procedure, the specific tacts emitted by children in noninstructional settings were seldom the same tacts that were directly trained during intensive tact 12 SLP-ABA Volume 5, No. 1 instruction. Greer and Speckman (2009) argue that acquiring conditioned reinforcement for attention is Figure 2. Participant A’s instructional sessions for each category of each set of tacts 13 SLP-ABA Volume 5, No. 1 Figure 3. Participant B’s instructional sessions for each category of each set of tacts 14 SLP-ABA Volume 5, No. 1 Figure 4. Participant C’s instructional sessions for each category of each set of tacts 15 SLP-ABA Volume 5, No. 1 the strongest predictor of children’s emission of tacts. The acquisition of conditioned reinforcement for attention is basic to the development of more advanced verbal development especially the emission of conversational units and the acquisition of Naming. Why did the mands increase in the baseline during generic educational instruction? We think that with the increased daily number of learn units in generic non-tact instruction, the participants were presented with more opportunities to be reinforced and more opportunities for emitting mands for their tokens and other reinforcers, and these experiences may have resulted in generalization to noninstructional settings. However, the increased intraverbals following intensive tact instruction also showed that the participants engaged in more speaker-listener exchanges with others after the intensive tact instruction, again suggesting the enhanced reinforcement of social attention. Greer and Speckman (2009) and Greer and Ross (2008) argue that the emission of mands is much easier to obtain because of the nature of reinforcement for mands, since they specify their reinforcer. In contrast, the emission of tacts is controlled by social contingencies of reinforcement. The latter is critical to becoming truly verbal (Greer & Speckman; Barnes-Holmes, et al., 2000; Horne & Lowe, 1996) in the joining of the listener and speaker. The intensive tact procedure is an effective means to compensate for the missing language opportunities associated with children like those described in studies by Pistoljevic and Greer (2006) and Pereira-Delgado & Oblak (2007). The protocol simply ensured that the participants received frequent tact instruction. The findings in the current study were consistent with Pistoljevic and Greer (2006) , PereiraDelgado, & Oblak (2007), and Pistoljevic (2008) in that the participants used both the tacts that they had in repertoire and those that were taught, while most of the tacts that the students emitted in noninstructional settings were not the tacts that they were taught during intensive tact instruction. The intensive tact procedure may have increased the reinforcement value of generalized reinforcement for tacts. This interpretation is supported by the fact that the participants emitted tacts that were not targeted during intensive tact instruction for the most part. Thus the participants may have learned to emit the tacts as a means to recruit generalized reinforcers in the form of attention from the teachers made possible by the enhanced effects of attention as a generalized reinforcer. Several limitations of the current study should be noted. First, there are only three participants. More participants functioning at similar and different verbal behavior levels are needed to assess the generality of these findings. Second, the instruction for the intensive tact procedure was delivered in some of the same contexts (e.g., play area) where probe data were collected to assess verbal operants in noninstructional settings. This enhanced the generality of responding and differed from some prior studies where the intensive tact instruction occurred at the children’s desks. Third, responses made by some of the participants were not consistently progressive. There was an especially noticeable decrease in Participant C’s emission of pure tacts during probe 4 in comparison to probe 3 of intensive tact instruction. This might have been due to a significant change in this Participant’s environment. His school had rearranged all of the students and staff at the time he was learning Set 4 of intensive tact instruction. Moreover, based on the significant progress he had made across the semester, he had been switched to a new classroom with peers who demonstrated relatively higher verbal behavior levels. Therefore, the probes conducted after Set 4 took place in the new classroom with new peers and teachers. Together with 16 SLP-ABA Volume 5, No. 1 Participant C’s new placement, the books and toys he usually read and played with were different. What’s more important, the peers that he interacted with and emitted tacts with had changed as well. The children in the new classroom were more verbally sophisticated and had more verbal capabilities. In this new class, Participant C’s verbal behavior levels became relatively lower compared to his new friends. His decrease in the numbers of pure tacts following the final set may be a reflection of this change. According to Kantor (1958), the founder of Interbehaviorism, any interaction of events occurs in a context rather than in isolation. Perhaps, the social control for tacts is contextually related to one’s acquaintances. This needs to be assessed empirically. Despite its limitations, the results of this study replicated those of prior studies regarding the effects of the ITP on “spontaneous speech”. Moreover, they suggest that it is the tact training and not an increase in instruction that was the source of the increase in spontaneous vocal verbal behavior. The fact that the tacts emitted in the non-instructional settings were not those taught is also consistent with prior findings. We speculate that either the procedure acts as an establishing operation for tact operants or that the attention of other individuals is conditioned as a reinforcer for the emission of tacts. The current findings, together with prior studies on the intensive tact instruction, suggest that the effect of the procedure is relatively robust. Children like these and children like those in the prior studies are likely to increase their emission of “spontaneous” verbal operants in non-instructional settings. At the very least they acquire more tacts, and until they acquire Naming this is their only way to expand their verbal repertoire. References Albers. A, & Greer, R. D. (1991). Is the three term contingency trial a predictor of effective instruction? Journal of Behavioral Education, 1, 337-354. American Psychiatric Association (1994). Diagnostic & Statistical Manual of Mental Disorders (4th Ed). Washington, DC: Author. Barnes-Holmes, D., Barnes-Holmes, Y. & Cullinan, V. (2000). Relational frame theory and Skinner's Verbal Behavior: A possible synthesis. The Behavior Analyst, 23, 69-84. Greer, R. D., McCorkle. N. P. & Williams, G. (1989). A sustained analysis of the behaviors of schooling. Behavioral Residential Treatment. 4, 113-141. Greer, R.D. (2002). Designing teaching strategies: An applied behavior analysis system approach. San Diego, CA: Academic Press. Greer, R. D., & McCorkle, N. P. (2003). CABAS® curriculum and inventory of repertoires for children from preschool through kindergarten, 3rd edition. Yonkers, NY: CABAS®/Fred S. Keller School. (Publication for use in CABAS schools only). Greer, R. D., & Ross, D. E. (2004). Verbal behavior analysis: A program of research in the induction and expansion of complex verbal behavior. Journal of Early Intensive Behavior Intervention. 1(2), 141-165. http://www.jeibi.com/JEIBI-1-2.pdf. 17 SLP-ABA Volume 5, No. 1 Greer, R. D. & Ross, D. E. (2008). Verbal Behavior Analysis: Inducing and Expanding New Verbal Capabilities in Children with Language Delays. New York, NY: Allyn and Bacon. Greer, R. D. & Speckman, J. (2009). The integration of speaker and listener responses: A theory of verbal development. The Psychological Record, 59, 449-488. Greer, R. D., Stolfi, L., & Pistoljevic, N. (2007). Emergence of naming in preschoolers: A comparison of multiple and single exemplar instruction. European Journal of Behavior Analysis, 8(2), 109-131. Horne, P. J., & Lowe, C. F. (1996). On the origins of naming and other symbolic behavior. Journal of Experimental Analysis of Behavior, 65, 185-241. Horner, R. D., & Baer, D. M. (1978). Multiple probe technique: A variation on the multiple baseline. Journal of Applied Behavior Analysis, 11, 189-196. Ingham, P., & Greer, R. D. (1992). Changes in student and teacher responses in observed and generalized settings as a function of supervisor observations. Journal of Applied Behavior Analysis, 25, 153-164. Kantor, J. R. (1958). Interbehaviora l Psychology: A sample of scientific system instruction. Principia Press: Bloomington. Kelly, R. L. (1997). A functional analysis of the effects of mastery and fluency on maintenance. (Doctoral dissertation, Columbia University, 1997). Abstract from UMI Proquest Digital Dissertation [on-line]. Dissertation Abstracts Item: AAT 9616720. Krantz, P. J., & McClannahan, L. E. (1998). Social interaction skills for children with autism: A script fading procedure for beginning readers. Journal of Applied Behavior Analysis, 31, 191–202. Lamn, N. & Greer, R. D. (1991). A systematic replication of CABAS in Italy. Journal of Behavioral Education, 1, 427-444. Partington, J. W., Sundberg, M. L., Newhouse, L., & Spengler, S. M. (1994). Overcoming an autistic child’s failure to acquire a tact repertoire. Journal of Applied Behavior Analysis, 27, 733–734. Pereira-Delgado, J. A. & Oblak.M. (2007). The Effects of Daily Intensive Tact Instruction on the Emission of Pure Mands and Tacts in Non-Instructional Settings by Three Preschool Children with Developmental Delays. Journal of Early and Intensive Behavior Intervention, 4, 392-411. Pistoljevic, N. (2008). The effects of multiple exemplar instruction and intensive tact instructional histories on the acquisition of naming in preschoolers. Unpublished dissertation. Columbia University Graduate School of Arts and Sciences. Pistoljevic, N. & Greer. R. D. (2006). The Effects of Daily Intensive Tact Instruction on Preschool Students’ Emission of Pure Tacts and Mands in Non-Instructional Setting. Journal of early and intensive behavior intervention, 1, 103-120. Ross, D. E. & Greer, R. D. (2003). Generalized imitation and the mand: Inducing first instances of speech in young children with autism. Research in Developmental Disabilities, 24, 58-74. 18 SLP-ABA Volume 5, No. 1 Ross, D. E., Nuzzolo, R., Stolfi, L., & Natarelli, S. (2006). Effects of speaker immersion on independent speaker behavior of preschool children with verbal delay. Journal of Early and Intensive Behavior Intervention, 3. 1, 135-150. Retrieved February 22, 2006, from http://www.behaivor-analyst-online.org Schauffler, G. & Greer, R. D. (2006). The effects of intensive tact instruction on audience accurate tacts and conversational units. Journal of Early and Intensive Behavioral Interventions,3, 120-132. Schwartz, B. Z. (1994). A comparison of establishing operations for teaching mands to children with language delays. (Doctoral dissertation, Columbia University, 1994). Abstract from UMI Proquest Digital Dissertations [on-line]. Dissertations Abstracts Item: AAT 9424540. Selinski, J., Greer, R. D., & Lodhi, S. (1991). A functional analysis of the comprehensive application of behavior analysis to schooling. Journal of Applied Behavior Analysis, 24, 108-118. Skinner, B.F. (1957). Verbal Behavior. Acton, MA: Copley Publishing Group. Sundberg, M. L., Michael, J., Partington, J. W., & Sundberg, C. A. (1996). The role of automatic reinforcement in early language acquisition. The Analysis of Verbal Behavior, 13, 21-37. Tsiouri, I. & Greer, R. D. (2003). Inducing vocal verbal behavior through rapid motor imitation training in young children with language delays. Journal of Behavioral Education, 12, 185-206. Williams, G., & Greer, R. D. (1993). A comparison of verbal-behavior and linguistic communication curricula for training developmentally delayed adolescents to acquire and maintain vocal speech. Behaviorology, 1, 3146. Zimmerman, I. L., Steiner, V. G., & Pond, R. E. (2002). Preschool language scale (4th ed.). San Antonio, TX: Psychological. Authors Note This experiment was conducted as part of the second author’s doctoral coursework under the supervision of the first author. We appreciate the contributions and cooperation of the children during the conduct of this research. Author contact information R. Douglas Greer e-mail: [email protected] Box 76 Teachers College Columbia University New York, NY 10027 Lin Du Department of Health and Behavior Teachers College Columbia University New York, NY 10027 Email: [email protected] 19 SLP- ABA Volume 5, Issue No. 1 Effects of Multiple Exemplar Instruction on Transformation of Stimulus Function Across Written and Vocal Spelling Responses by Students with Autism Carly M. Eby, R. Douglas Greer, Lisa D. Tullo, Katherine A. Baker, and Rebecca Pauly Abstract Transfer of stimulus function (TSF) involves the acquisition of an untaught response to a stimulus which previously evoked only a single taught response. We tested the effects of multiple exemplar instruction on the TSF across vocal and written spelling responses of 3 elementary students with autism. Participants were taught to spell 4 words (Set 1) either vocally or graphically. Untaught responses were probed. Then, participants were taught to spell 4 different words (Set 2) in the opposite topography. Following mastery, untaught responses to Set 1 were again probed. Finally, 4 novel words (Set 3) were taught in a single-response-topography (saying or writing) and the untaught topography was probed. Results showed correct untaught responses to Set 1 words and eme rgence of set 4. Keywords : spelling, multiple exemplar instruction, independence of verbal operants, transformation of function Introduction According to Skinner (1957), speaking and writing are different kinds of verbal operants that initially must be “separately conditioned” (p. 191). Skinner (1957) proposed that words that occur in written form are functionally independent of the vocal form of the same word. He refers to this as the “same response in different media” (p.191). On this topic, Skinner wrote: But speaking and writing are obviously different kinds of behavior, which utilize different parts of the body in different ways. Where we could paraphrase “the same word used in different ways” as “the same response in different types of operant,” here we must attempt to bridge the gap between spoken and written behavior either by pointing to something common to the occasions upon which the behaviors occur or among the effects which they have upon the listener and reader. But common controlling variables, action either prior to the behavior in the stimulation occasion or after the behavior as part of the event called reinforcement, will not permit us to get from one form of the response to the other. The two forms of behavior must be separately conditioned. (p. 191) Consistent with Skinner’s theory, several types of verbal operants are initially independent including: mands and tacts (e.g., Lamarre & Holland, 1985; Hall & Sundberg, 1987; Twyman, 1996; Williams & Greer, 1993), listener and speaker responses (Horne & Lowe, 1996), and written and spoken responses (Greer, Yuan, & Gautreaux, 2005). Recent research reported the effectiveness of multiple exemplar instructional methods in joining separate verbal operants to a single controlling stimulus, such as with listener and speaker responses (Horne & Lowe, 1996; Fiorile & Greer, 2007; Greer, Stolfi, Chavez-Brown, & Rivera-Valdez, 2005; Pistoljevic, 2008), and with written and spoken responses (Greer, Yuan, et al., 2005). Similarly, Nirgudkar (2005) and Nuzzolo-Gomez & Greer (2005) joined single responses to the control of different establishing operations for mand and tact operants as a function of multiple exemplar instruction (MEI) across establishing operations. According to Greer & Ross (2008) MEI) (also known as multiple exemplar training or MET) has been described in the literature in two ways. In one use of the term (also called general case teaching), 20 SLP- ABA Volume 5, Issue No. 1 MEI refers to a tactic in which different examples of the same stimulus are presented when teaching generalization or abstraction (e.g., Hughes & Rusch, 1989; Sprague & Horner, 1984). For example, teaching textual responses to phonemes across multiple exemplars results in the emission of accurate textual responses to novel words (Engelmann & Carnine, 1982). In another usage, MEI refers to a tactic used to bring independent operants (e.g., writing, spelling aloud, and textually responding) under joint stimulus control for novel stimuli by rotating different responses to a single stimulus in instructional sets (Greer, Yuan, et al., 2005; Fiorile & Greer, 2007; Greer, Stolfi, et al., 2005). That is, “the manipulation of initially independent response topographies with the same stimulus may generate the joint stimulus control such that a single stimulus can evoke both responses” (Greer, Yuan, et al. p. 100). In the study described herein, the latter description of MEI applies. Greer, Yuan, et al. (2005) tested the effect of MEI across written and spoken responses in instructional sets on the acquisition of untaught spelling responses in the vocal or written form, for four young children, using a delayed multiple probe design. After an initial test showed that the participants could not spell three sets of five words, the participants were taught to spell Set 1 words in one response form (either vocal or written), and after achieving mastery in the taught form, probes were conducted on the untaught form. The probes showed that the participants could not emit the untaught responses. Participants were then taught to spell an instructional set of words (Set 2) to mastery, rotating the responses across writing and saying letters. Following the MEI intervention (with Set 2 only) probes were again conducted on the untaught response forms in Set 1 and the results showed that the untaught forms emerged. Lastly, participants were taught to spell Set 3 words using single response training, and then probes were conducted on the untaught form. Correct responses to the untaught forms emerged. The results of the Greer, Yuan et al. (2005) study showed that correct responding of the untaught form for Set 1 was low for all four participants prior to the MEI intervention. Following MEI, all four participants demonstrated correct responses in the untaught form in 80% to 100% of probe trials. In addition, when Set 3 was introduced, all four participants demonstrated correct responses in the untaught form in 80% to 100% of probe trials. These results were replicated with a second group of Kindergarten students diagnosed with autism spectrum disorders (ASD) who performed academically at or above grade level. The participants in the Greer et al. study were kindergarteners who were either described as having language delays (Experiment 1) or as performing at or above grade level (Experiment 2). These procedures have not yet been tested with participants with more limited levels of verbal behavior. Therefore, we tested the effects of multiple exemplar instruction across saying and writing on the acquisition of untaught spelling responses with elementary-aged children who were diagnosed with ASD and who had more limited verbal repertoires. Method Participants Three 7-year old males were selected for this study because, based on baseline probes, they did not demonstrate joint stimulus control of spelling across written and spoken functions and they could not spell any of the words used in the experiment in either topography. All of the participants had a diagnosis of ASD. The participants’ verbal capabilities were assessed using the International Curriculum and Inventory of Repertoires for Children from Pre-school through Kindergarten (AIL/PIRK) (Greer & McCorkle, 2008), a curriculum-based assessment that includes assessments of children’s verbal developmental cusps and capabilities as identified in Greer and Ross (2008) and Greer and Speckman (2009). See Waddington and Reed (2009) for a description and evaluation of the PIRK. Table 1 includes a more detailed description of the participants. Participant 1 was a 7.9-year old male. His verbal capabilities and cusps included: basic listener literacy, the teacher’s presence results in instructional control, following of two-step directions, generalized match-to-sample responding, and the 21 SLP- ABA Volume 5, Issue No. 1 presence of mands and tacts with autoclitic s (Greer & Ross, 2008). In addition, Participant 1 read at a first grade level. Participant 2 was a 7.8- year old male. His verbal capabilities and cusps included: basic listener literacy, basic mand and tact operants, following of multi-step directions, and generalized match-to sample responding. He read at a Kindergarten level. Participant 2 had a speech disorder that affected his articulation of consonant sounds. For example, when reading the word, “Cat,” he often omitted both the “c” and “t” sounds, thus reading the word as “ah” or “gah.” The participant had an echoic repertoire at the time of the study and targeted echoic responses were shown to improve with the application of learn units as a method of instruction. Participant 3 was a 7.0-year-old male. He followed one-step instructions and emitted some mands and tacts in the context of instruction. He performed at a Kindergarten level for math and a first grade level for reading. His verbal cusps and capabilities included generalized motor imitation, match-tosample responding across the senses, echoic -to-mand, echoic -to-tact, and transformation of establishing operations across mands and tacts. Participant 3 did not have the topography of handwriting in his repertoire at the time of the study, however he could use a computer keyboard to type words, therefore during the study all writing responses for this participant were typed. Table 1. Description of Participants Part Age Diagnosis & Level of Verbal Ability Standardized Test Scores PIRK Verbal Behavior Developmental Cusps and Capabilities 1 - 7.9 - male -ASD - Listener/Speaker - Writer/Reader - 1st grade reading level - Preschool Language Scale-4: Auditory Comprehension: SS=53, 3%; Expressive Communication: SS=50; Total Language: SS=50 - Receptive One-Word Picture Vocabulary Test: SS=84, 14% -Mands/Tacts with autoclitic frames -Following vocal verbal directions -Conversational units emitted -Matching and pointing/Listener repertoires 2 -7.8 - Male -ASD -Listener/Speaker -Emergent Reader/Writer -Vineland-II Adaptive Behavior Scales: Adaptive Behavior Composite: 61 - Receptive One-Word Picture Vocabulary Test: SS:62:1% - Expressive One-Word Picture Vocabulary Test: SS:<55;<1% -Test for Auditory Comprehension of Language3: 53; <1% -Mands with autoclitic frames -Tacts -Following vocal verbal directions -Matching and Pointing/Listener repertoires - ASD - Listener/Speaker - Emergent Reader/Pre Writer - Reads on Kindergarten level - Receptive One-Word Picture Vocabulary Test: Total A.E. 5.3 - Expressive One-Word Picture Vocabulary Test: A.E. 5.2 -Beery-Visual Motor Integration-V: SS=76, 5% - Mands/Tacts - Following vocal verbal directions Transformation of establishing operations for mands and tacts 3 - 7.0 - male 22 SLP- ABA Volume 5, Issue No. 1 Setting The participants were selected from a self-contained classroom within a public school located in a suburb outside of a large metropolitan area. The classroom was characterized by its comprehensive application of applied behavior analysis strategies and tactics to teach the range of curricula. All participants had a history of instruction that consisted of learn units (Greer & McDonough, 1999; Albers & Greer, 1991) that made up all of their daily academic, verbal behavior, self-management, and problem solving instruction. A learn unit consists of a series of interlocking operants for a student and teacher, in which there is one potential three-term contingency for the student and two or more for the teacher. The learn unit begins with the student emitting an attending response, which is the first antecedent for the teacher. The teacher presents an antecedent stimulus to the student (functioning as both teacher behavior and student antecedent). The student responds to the teacher antecedent (functioning as both student behavior and teacher consequence). Following the student’s correct or incorrect response the teacher immediately delivers a consequence (reinforcement for a correct response or correction for an incorrect response) and records the student response (functioning as both student consequence and teacher behavior). The student attends to the correction or reinforcement and actively responds to any corrections (functioning as teacher consequence and completing the learn unit). Instructional sessions took place in the participants’ classroom. During all sessions, other students in the class received instruction from other professionals at the same time. Both the student and the teacher sat at either a horseshoe shaped table or a desk, in child-sized chairs, facing one another. Materials included data sheets, black pens, and prosthetic reinforcers. Student materials included pencils and lined paper used for elementary school-age children’s writing instruction. Participant 3 used a Dell™ Computer keyboard for his written responses. Dependent Variable The dependent variable was correct responses to no feedback probe trials for untaught response topographies, either written or spoken. Prior to the baseline probe for Set 1, the participants were taught to mastery either the written or spoken response form for Set 1 using learn units. During instruction, the teacher said, “Spell ___” (for Participant 1) or “Write ___” (for Participant 2) and reinforcement and corrections were provided, as described in the Independent Variable section. The instruction “Spell” was used to evoke a spoken response for the letters of the words and the instruction “Write” was used to evoke a written response. Response Definitions The target behaviors were written and spoken responses to vocal dictation by the teacher. The definition of a correct spoken spelling response was, given the antecedent, “Spell ____,” the student would say the letters of the word in the correct order so that there was point-to-point correspondence to the conventional spelling of the word. For example, in response to the teacher-delivered antecedent, “Spell boat,” the student would say “b-o-a-t.” Examples of an incorrect spoken spelling response included, saying the letters in the wrong order, omitting or adding incorrect letters, spelling a different word, or no response. The definition of a correct written spelling response was, given the instruction, “Write ____,” the student would write the letters of the word in the correct order according to the conventional spelling of the word. For example, given the teacher instructional antecedent, “Write jump,” the student would write jump. Examples of an incorrect written spelling response included, writing the letters in the wrong order, writing a word other than the target, omitting or adding incorrect letters, writing illegibly, or no response. Each of the participants was assigned three sets of four words. The three sets of words for each participant are shown in Table 2. The words were two-, three-, four-, and five-letter words from a list of high-frequency words (Eldredge, 1995). 23 SLP- ABA Volume 5, Issue No. 1 Table 2. Word Sets for Participants 1, 2, and 3 Participant 1 2 3 Set 1 Words where, away, three, I Vocal Instruction Written Probe help, jump, not, here Written Instruction Vocal Probe seven, carry, try, gave Written Instruction Vocal Probe Set 2 Words help, jump, not, here MEI: Vocal and Written Instruction where, away, three, I MEI: Vocal and Written Instruction why, upon, myself, your MEI: Vocal and Written Instruction Set 3 Words Little, find, make, fear Written Instruction Vocal Probe every, warm, live, take Vocal Instruction Written Probe buy, does, the, mad Vocal Instruction Written Probe Data Collection. An experimenter collected data after each response using pencil and paper. Some sessions were video recorded using a digital video camera with an audio recorder. This was used for interobserver agreement purposes. Data were collected on untaught written and spoken spelling responses that served as the dependent varia ble (Set 1 and Set 3 words). Data were also collected on the responses to learn unit instruction used in the MEI intervention for both written and spoken topographies for Set 2 words, as well as the taught topographies for Set 1 and Set 3 words. Correct responses were recorded by marking a plus (+) on the data form and incorrect responses were recorded by marking a minus (-) on the data form. During the intervention phase, if one response type was mastered before the other, we continued to present it in the rotation as an antecedent, however they were no longer recorded or graphed. That is, if a participant met criterion for vocally spelling the words in his training set, but did not yet meet criterion in the written topography for those same words, both topographies continued to be presented and reinforced on a continuous schedule until criterion was met in both response topographies. Interobserver and Interscorer Agreement Independent observers recorded correct and incorrect responses in 42% of probe sessions either during the sessions or after viewing the video recordings. The percentage of agreement for correct and incorrect responses was 100% across all sessions in which there was an independent observer. We determined the percentage of agreement by dividing the total numbers of point-by-point observer agreements by the total numbers of agreements plus disagreements, and then multiplying by 100. Procedure Design. The design was a multiple probe design across participants. Sessions were time lagged to control for maturation and history. The order of response type, written or spoken, was counterbalanced across participants, such that Participant 1 received vocal response instruction on Set 1 words and Participants 2 and 3 received written response instruction on Set 1 words. The conditions were presented in the following sequence: (a) Pre-experimental probes of written and spoken responses to all three sets of words, demonstrating that the students could not spell the words in either topography; (b) single response instruction in either spoken or written responses to Set 1 words, (c) probes for untaught responses to Set 1 words, (d) multiple exemplar instruction across vocal and written responses for an instructional set of words, (e) probes for untaught responses to Set 1 words, (f) single response instruction in either spoken or written responses to a new set of words (Set 3), (g) probes for untaught responses to Set 3 words General Procedure. The participant and experimenter were seated at a table with limited distractions. The experimenter gained the participant’s attention before delivering each antecedent by saying his name and/or waiting for the participant to make eye contact or emit a ready response, such as the participant saying, “I’m ready.” Depending on the condition, the experimenter would deliver one of 24 SLP- ABA Volume 5, Issue No. 1 two antecedents. In the written condition, it was “Write the word ____.” In the vocal condition, it was “Spell the word ___.” Instructional sessions consisted of 20 learn units. Probe sessions consisted of 20 trials with no feedback (no reinforcement or corrections). Pre-Experimental Probe Conditions (Phase 1). A pre-test was conducted in which each of the three sets of words was presented under probe trial conditions prior to the baseline. This consisted of five presentations of each of the four words (i.e., 20 trials) for each of the three sets of words (Sets 1, 2, and 3), and for each of the response topographies (i.e., vocal and written). In the written condition, it was “Write the word ____.” In the vocal condition, it was “Spell the word ___.” No programmed reinforcement or corrections occurred. The experimenter delivered the relevant antecedent, waited 5 sec for a response, and then presented the next trial. There were 20 massed trials for vocal responses followed by 20 massed trials for written responses. None of the students emitted any correct responses in either the written or spoken response forms to any of the three word sets. Single Response Instruction – Set 1 (Phase 2). Learn units were implemented to teach a set of spelling words in either the written or vocal topography. Table 2 shows the words that were selected for each participant and the topographies that were taught and/or probed during each phase. Participant 1 received instruction on the vocal topography only for the words where, away, three, and in. Participant 2 received instruction on the written topography only for the words help, jump, not, and here. Participant 3 received instruction on the written topography only for the words seven, carry, try, and gave. Sessions consisted of 20 learn units, such that every word was presented five times per session. The order of word presentation was pre-determined so that the words were not presented in the same order each time. The experimenter gained the participant’s attention and delivered an antecedent. For Participant 1 the antecedent was, “Spell the word ___.” For Participants 2 and 3 it was, “Write the word ___.” Correct responses were followed by the delivery of reinforcement in the form of verbal approval (e.g., “that’s right”), tokens, and for Participant 2 and 3, small pieces of preferred foods. The difference in reinforcement operations was based on the participant’s prior instructional history. If the participant emitted an incorrect response or did not begin to respond within 5 sec of the antecedent, a correction procedure followed. A correction procedure consisted of the experimenter re-stating the antecedent, providing a model response, and having the participant emit the correct response. Corrections were not reinforced according to learn unit protocol. Criterion for mastery was 90% correct across two consecutive sessions or 100% in one session. Following mastery of Set 1 spelling words, a probe was conducted on the topography that was not taught. Post Single Response Instruction Probe (Phase 3). The same set of words from Phase 2 was presented, except the antecedent changed in order to evoke the untaught response topography. Thus, Participant 1 was asked to “Write the word ___.” Participants 2 and 3 were asked to “Spell the word ___.” No programmed reinforcement or corrections occurred. The experimenter delivered the antecedent, waited 5 sec for a response, and then presented the next tria l. Sessions consisted of 20 trials. Multiple Exemplar Instruction – Set 2 (Phase 4). We implemented multiple exemplar instruction across vocal and written topographies of spelling as an instructional tactic to teach Set 2 words in both topographies. Table 2 shows the words that were selected for each participant. Participant 1’s words were help, jump, not, and here. Participant 2’s words were where, away, three, and in. Participant 3’s words were why, upon, myself, and your. Instruction consisted of learn units. The experimenter presented an antecedent to the attending student, depending on the condition. In the spelling condition, the antecedent was “Spell the word ___”. In the writing condition, the antecedent was “Write the word ___”. The order of presentation of words was counterbalanced across response topographies and words so that the same word was never presented in two consecutive learn units. An example of an instructional sequence was: “Write help,” “Spell jump,” “Write not,” “Spell here”. Participants were given 5 sec to respond. Correct responses were followed by the delivery of reinforcement in the form of verbal approval (e.g., “that’s right”), tokens, and for Participant 2 and 3, small pieces of preferred foods. Incorrect responses or no response within 5 sec were followed by a correction operation in which the teacher re-stated the 25 SLP- ABA Volume 5, Issue No. 1 antecedent, modeled the correct response, and the student repeated vocally or wrote the correct response, depending on the condition. Corrected responses were not reinforced. Sessions consisted of 20 vocal learn units and 20 written learn units presented in an alternating fashion, thus a total of 40 learn units per session. Mastery criterion during instruction was 90% across two consecutive sessions or 100% in one session for both response types. Post Multiple Exemplar Instruction Probe (Phase 5). The procedures in this condition were identical to those described in Phase 3. Post MEI Single Response Instruction – Set 3 (Phase 6). The procedures in this condition were similar to those described in Phase 2, however a new set of words was introduced and the response topography that was not taught (i.e., the probe topography) in the first single response instruction phase was taught in this phase. Thus, Participant 1 received instruction in the written topography only for the words little, find, make, and fear. Participant 2 received instruction in the vocal topography only for the words every, warm, live, and take. Participant 3 received instruction in the vocal topography only for the words buy, does, the, and mad. Post Single Response Instruction Probe (Phase 7). The same words from Phase 6 were presented, except the antecedent changed in order to evoke the untaught response topography. Thus, Participant 1 was asked to “Spell the word ___.” Participants 2 and 3 were asked to “Write the word ___.” No programmed reinforcement or corrections occurred. The experimenter delivered the antecedent, waited 5 sec for a response, and then presented the next tria l. Sessions consisted of 20 trials. Fidelity of Treatment. Fidelity of treatment was measured using the Teacher Performance Rate and Accuracy scale (Ingham & Greer, 1992) that simultaneously assessed both the accuracy of the measurement of the students’ responses and fidelity of treatment. Fidelity of treatment observations were conducted during 19% of the instructional sessions and the mean percentage of agreement was 92% (range, 84% to 100%). Results The results for the acquisition of spelling responses for Participants 1, 2, and 3 are displayed in Figure 1. The results for the pre- and post-instructional probes for Participants 1, 2, and 3 are displayed in Figure 2. In the pre-experimental probes, none of the participants emitted any correct responses to the three word sets, in either the vocal or written response form. Following single response instruction on Set 1 words, Participants 1, 2, and 3 emitted 70%, 55%, and 0% correct untaught responses to Set 1 words, respectively. Following MEI with Set 2 words, correct untaught responses to Set 1 increased to 95%, 90%, and 90% for Participants 1, 2, and 3, respectively. Following instruction on a single topography for Set 3 (a novel set of words), correct untaught responses to Set 3 were 95%, 100%, and 100% for Participants 1, 2, and 3, respectively. Participant 1 achieved criterion for vocal responses to Set 1 words after two sessions (i.e., 40 learn units). He achieved criterion for vocal and written responses during the MEI training condition with Set 2 words after two sessions (i.e., 80 learn units). Finally, he achieved criterion for written responses to Set 3 words after three sessions (i.e., 60 learn units). Participant 2 achieved criterion for vocal responses to Set 1 words after six sessions (i.e., 120 learn units). He achieved criterion for vocal and written responses during MEI with Set 2 words after six sessions (i.e., 240 learn units). Finally, he achieved criterion for written responses to Set 3 words after four sessions (i.e., 80 learn units). Participant 3 achieved criterion for vocal responses to Set 1 words after three sessions (i.e., 60 learn units). He achieved criterion for vocal and written responses during MEI with Set 2 words after four sessions (i.e., 140 learn units). Finally, he achieved criterion for written responses to Set 3 words after three sessions (i.e., 60 learn units). 26 SLP- ABA Volume 5, Issue No. 1 Figure 1. Correct vocal and written responses to learn unit presentations during: (a) baseline instruction in single response forms, (b) multip le exemplar instruction across written and vocal topographies, (c) single response form instruction for Participants 1, 2, and 3. 27 SLP- ABA Volume 5, Issue No. 1 Figure 2. Correct untaught spelling responses by Participants 1, 2, and 3 following: (a) baseline instruction in single response forms, (b) multiple exemplar instruction across written and vocal topographies, (c) single response form instruction for Participants 1, 2, and 3. Discussion The results of this study replicated previous findings of Greer, Yuan, et al. (2005). Prior to the study, none of the participants could spell any of the words that were presented, either in written or vocal forms. The number of correct untaught spelling responses increased as a function of the MEI tactic for all three participants. Thus, MEI across saying and writing was an effective intervention for all three participants that resulted in the transformation of stimulus function across written and vocal spelling 28 SLP- ABA Volume 5, Issue No. 1 responses for the initial untaught response sets. Moreover, when a novel set of words was introduced and taught in one response topography, the untaught topography emerged for all three participants. This transformation of function was not present for any of the participants prior to the intervention, but was present following the intervention. Greer, Yuan, et al. (2005) first characterized this phenomenon as the acquisition of joint stimulus function. They proposed that the multiple exemplar training, in which participants were taught vocal and written responses to a subset of words in an alternating fashion, “produced joint stimulus function for both responses to novel words taught as a single response” (p. 111). According to Greer et al. (2005), this could be characterized as a higher order class of behavior, as described by Catania (1998), or as a relational frame as described by Barnes-Holmes, Hayes, Barnes-Holmes, & Roche (2001). According to Catania (1998), a higher-order class of behavior is “an operant class that includes within it other classes that can themselves function as operants, as when generalized imitation includes all the component imitations that could be separately reinforced. Higher-order classes may be a source of novel behavior” (p. 392). Greer, Yuan, et al. (2005) suggested that the transformation of stimulus function observed in their study could perhaps be explained in terms of a derived relation (BarnesHolmes, Barnes-Holmes, and Cullinan, 2000; Barnes-Holmes, et al., 2001) between saying the letters and writing them, such that “once individuals have derived relations between saying the letter and writing the letter, rotated experiences for a subset of exemplars can result in the emission of untaught response forms” (Greer et al., 2005, p. 112). Although the presence of the derived relation was not tested, Greer et al. suggested that it was probable that their participants had certain prerequisites that allowed the fundamental derived relations to be present. The strongest results in the current study were shown by Participant 3, who emitted zero correct untaught responses prior to the intervention and following the intervention increased by 90% on Set 1, then emitted 100% correct untaught responses to the novel Set of words. Both Participants 1 and 2 did demonstrate some joint speaker-writer behavior prior to the intervention, however neither responded at criterion level (90%). Participant 1 responded correctly in 70% of probe trials prior to the intervention, and Participant 2 responded correctly in 55% of probe trials prior to the intervention. Although these preexperimental levels of responding indicated that the joining of speaker and writer responses was not fully present for these participants, we recognize that they may have had degrees of this “cusp” prior to receiving the intervention. However, the multiple exemplar instructional experience that they received in this study functioned to strengthen or perhaps induce the joining of saying and writing for Participants 1 and 2. For Participant 3, the results showed that the capability of jo ining saying with writing was not present prior to the intervention and emerged as a result of the intervention. It is likely that the acquisition of phonemic control for letter sounds and the joining of the sounds of letters with saying the letter names and the writing of the letters is the source of the derived relation between saying and writing. While direct phonemic instruction can assist this development, in many cases exemplar experiences appear to lead to the abstraction of phonemic control without direct instruction as when children who are taught to read with the whole word approach derive phonemic control. Thus, the participants in this experiment will likely require more exemplary experiences or direct instruction in phonemes and exceptions for saying and writing letters in order to have complete transformation of stimulus function, however the basic instantiation of the transformation of stimulus function has been established. This and the prior experiments reported by Greer, Yuan et al. (2005) increases the probability that multiple exemplar experiences or instruction leads to the joining of saying and writing. If other laboratories can replicate these procedures it would appear that what Skinner (1957) identified as the missing controlling variables for the joining of saying and writing have been identified 29 SLP- ABA Volume 5, Issue No. 1 References Albers, A. E., & Greer, R. D. (1991). Is the three-term contingency trial a predictor of effective instruction? Journal of Behavioral Education, 1, 337-354. Barnes-Holmes D., Barnes-Holmes Y., & Cullinan, V. (2000). Relational frame theory and behavior: A possible synthesis. The Behavior Analyst, 23, 69-84. Skinner’s verbal Barnes-Holmes, Y., Hayes, S. C., Barnes-Holmes, D., & Roche, B. (2001). Relational frame theory: a postSkinnerian account of human language and cognition. Advances in Child Development and Behavior, 28, 101-138. Catania, A. C. (1998). Learning (4th edition). New Jersey: Prentice Hall. Eldredge, J.L. (1995). Teaching decoding in holistic classrooms. Englewood Cliffs, NJ: Prentice Hall. Engelmann, S., & Carnine, D. (1982). Theory of instruction: Principles and applications. New York: Irvington. Fiorile C.A, & Greer, R.D. (2007). The induction of naming in children with no echoic-to-tact responses as a function of multiple exemplar instruction. The Analysis of Verbal Behavior, 23, 71-88. Greer, R.D. & McCorkle, N.P. (2008). CABAS® international curriculum and inventory of repertoires for children from pre-school through kindergarten (AIL/PIRK), 3rd edition. Yonkers, NY: CABAS®/Fred S. Keller School. Greer, R. D. & McDonough, S. H. (1999). Is the learn unit a fundamental measure of pedagogy? The Behavior Analyst, 22, 5-16. Greer, R. D. & Speckman, J. (2009). The integration of speaker and listener responses: A theory of verbal development. The Psychological Record, 59, 449-488. Greer, R. D., Stolfi, L., Chavez-Brown, M., & Rivera -Va ldez, C. (2005). The emergence of the listener to speaker component of naming in children as a function of multiple exemplar instruction. The Analysis of Verbal Behavior, 21, 123-134. Greer, R. D., Yuan, L., & Gautreaux, G. (2004). Novel dictation and intraverbal responses as a function of multiple exemplar instructional history. The Analysis of Verbal Behavior, 21, 99-116. Greer R.D. & Ross D.E. (2008) Verbal behavior analysis: Inducing and expanding complex communication in children with language delays. Boston: Allyn & Bacon. Hall, G.A., & Sundberg, M.L. (1987). Teaching mands by manipulating conditioned establishing operations. The Analysis of Verbal Behavior, 5, 41-53. Horne, P. J. & Lowe, C. F., (1996). On the origins of naming and other symbolic behavior. Journal of the Experimental Analysis of Behavior, 65(1), 185-241. Hughes, C., & Rusch, F. R. (1989). Teaching supported emp loyees with severe mental retardation to solve problems. Journal of Applied Behavior Analysis, 22, 365-372. Ingham, P. I. & Greer, D. R. (1992). Changes in student and teacher responses in observed and generalized settings as a function of supervisor observations. Journal of Applied Behavior Analysis, 25(1), 153-164. Lamarre, J., Holland, J.G. (1985). The functional independence of mands and tacts. Journal of the Experimental Analysis of Behavior, 43, 5-19. Nirgudkar, A.S. (2005). The relative effects of the acquisition of naming and the multiple exemplar establishing operation experience on the acquisition of the transformation of establishing operations across mands and tacts. (Doctoral dissertation, Columbia University, 2005). Abstract from UMI Proquest Digital Dissertations [on-line]. Dissertations Abstract Item: AAT 3159751. Nuzzolo-Gomez, R., & Greer, R. D. (2004). Emergence of untaught mands or tacts with novel adjective-object pairs as a function of instructional history. The Analysis of Verbal Behavior, 24, 30-47. Pistoljevic, N. (2008). The Effects of Multiple Exemplar and Intensive Tact Instruction on the Acquisition of Naming in Preschoolers Diagnosed with Autism and Other Language Delays. Unpublished doctoral dissertation, 30 SLP- ABA Volume 5, Issue No. 1 Columbia University Teachers College. Abstract from UMI Proquest Digital Dissertations [on-line]. Dissertations Abstracts Item: AAT 3317598. Skinner, B.F. (1957). Verbal Behavior. Engelwood Cliffs, NJ; Prentice Hall. Sprague, J. R., & Horner, R. H. (1984). The effects of single instance, multiple instance, and general case training on generalized vending machine use by moderately and severely handicapped students. Journal of Applied Behavior Analysis, 17(2), 273-278. Twyman, J. (1996). The functional independence of impure mands and tacts of abstract stimulus properties. The Analysis of Verbal Behavior, 13, 1-19. Waddington, E. M, & Reed, P. (2009). The impact of using the “Preschool Inventory of Repertoires for Kindergarten” (PIRK®) on school outcomes of children with Autism Spectrum Disorders. Research in Autism Spectrum Disorders, 3, 809-827. Williams, G., & Greer, R. D. (1993). A comparison of verbal-behavior and linguistic communication curricula for training developmentally delayed adolescents to acquire and maintain vocal speech. Behaviorology, 1, 3146. Authors’ contact information: Carly M. Eby Box 76, Teachers College, Columbia University 525 West 120th St. New York, NY 10027 (212) 678-3880, [email protected] R. Douglas Greer Box 76, Teachers College, Columbia University, 525 West 120th St. New York, NY 10027 (212) 678-3880, [email protected] Lisa D. Tullo Box 76, Teachers College, Columbia University 525 West 120th St. New York, NY 10027 Katherine A. Baker Box 76, Teachers College, Columbia University, 525 West 120th St. New York, NY 10027 (212) 678-3880 [email protected] Rebecca Pauly Box 76, Teachers College, Columbia University 525 West 120th St. New York, NY 10027 (212) 678-3880. 31 SLP- ABA Volume 5, Issue No. 1 AAC Interventions: Case study of In-utero Stroke Cindy Geise Arroyo, Robert Goldfarb, Danielle Cahill, & Janet Schoepflin Abstract A case study design was used to examine the progression of Alternative Augmentative Communication (AAC) interventions including Picture Exchange Communication System (PECS) and the Speech Generating Devices (SGDs) 7-Level Communication Builder and Dynamo . The participant was a preschooler (CA=4:5) who presented with a history of hearing loss and, based on MRI/CT examination, a possible in-utero hypoxic-ischemic event. The goals of reducing maladaptive behaviors, improving receptive language skills , and using an AAC system to request, comment, and respond were achieved. The participant transitioned from using PECS, to using a fixed display and ultimately to using dynamic-display speech generating devices (SGDs). Naturalistic teaching strategies including graduated prompting, mandmodel, and fading were used. Long term follow-up revealed the participant’s present ability to communicate effectively without the support of AAC. Keywords : Augmentative and Alternative Communication (AAC), Speech Generating Device (SGD), inutero-stroke, fixed display, dynamic display, naturalistic teaching. Introduction Fetal stroke may follow an ischemic (thromboembolic) or hemorrhagic event which occurs between 14 weeks of gestation and the onset of labor resulting in delivery (Ozduman, et al., 2004). In ischemic strokes, cerebral blood flow falls below a level necessary to maintain nerve cell integrity and neurological function. Hemorrhagic strokes occur secondary to intra-cranial bleeding. A perinatal stroke is similar to an ischemic stroke, but it occurs between 28 weeks of gestation and 7 days of age. Poor understanding of etiologies of fetal and perinatal strokes have led to an estimate of 1 in 4,000 live births, with the true incidence probably being higher. Diagnosis usually includes ultrasound measurement of the fetal cranium and, more recently, fetal magnetic resonance imaging (MRI) and computed tomography (CT), which provide better definition of the injury to the fetal cerebrum. Patients often remain undiagnosed, as clinical outcomes of surviving infants may not be present until later in the first year of life (Ozduman, et al., 2004). Reports of in-utero strokes began some 30 years ago, based on autopsy data, and continue to follow a case report format, viewing living brains. An early review of nearly 600 infants examined at autopsy (Barmada, Moossy & Shuman, 1979) indicated the presence of cerebral infarcts (necrosis in an area of brain tissue, caused by an obstruction, usually a thrombosis or an embolism) in about one in twenty (5.4%) instances. Neonates delivered at term were more likely to be brain-damaged than premature infants, where multiple smaller infarcts had occurred. Surprisingly, for the infants who survived, disorders associated with focal neurological deficits were not predominant. Rather, the clinical features tended to include such autonomic disturbances as prolonged apnea and episodic seizures, and, in those infants with less severe complications, hemiplegia, mental and motor retardation, and recurrent seizures. As recently as 25 years ago, few infants surviving stroke were reported in the literature (Ment, Duncan & Ehrenkranz, 1984). Among infants who died during the first months of life, necrotic foci (areas of cell death) were generally located in border zones between vascular territories (Lou, 1983). This suggests that the in-utero brain is fragile, offering the brain poor protection against perfusion pressure, with normal birth causing a decrease in oxygen or mild hypotension sufficient to abolish autoregulation (the process by which organs maintain their own blood supply). Inadequate pressure, caused by reduced blood flow to the brain, can cause ischemia. Neonatal ischemia in surviving infants was seen as decisive in development of atrophic encephalopathy, where brain cells decrease in size, with the resultant clinical picture of motor and cognitive dysfunction. 32 SLP- ABA Volume 5, Issue No. 1 Etiologies of fetal and neonatal strokes are reported both as ischemic (thromboembolic) in “a significant number of these events” (Chalmers, 2005, p. 333) and hemorrhagic, with intracranial hemorrhage occurring in “approximately 40% of infants of less than 32 weeks’ gestation” (Huang, Chen, Tseng, Ho, & Chou, 2006, p. 135). One cause of the discrepancy may be the use of ultrasonography (US) for prenatal detection of fetal strokes. The use of MRI to supplement US findings may contribute to diagnostic accuracy and improve prediction of postnatal neurodevelopmental prognosis (Elchalal, et al., 2005). Confirmation of diagnosis by MRI or CT scan has implicated the brain’s ventricles, which produce the cerebrospinal fluid (CSF) needed to surround and protect the brain. Periventricular venous infarction may result in a focally enlarged lateral ventricle (Takanashi, Barkovich, Ferriero, Suzuki, & Kohno, 2003; Takanashi, Tada, Barkovich, & Kohno, 2005), or in hydrocephalus following hemorrhage in the choroids plexuses, which manufacture CSF in the ventricles (Huang, et al., 2006). Motor outcomes, after follow-up of more than five years, included leg hemiparesis and spasticity, if the basal ganglia were involved; non-motor disorders associated with cortical involvement included cognitivebehavioral impairments, visual deficits, and epilepsy (Kirton, Deveber, Pontigon, Macgregor, & Shroff, 2008). Another avenue of investigation in neonatal stroke is mutation in the factor V gene (factor V Leiden mutation), which is the most common cause of familial thrombosis, an inherited deficiency of antithrombin III (Thorarensen, Ryan, Hunter, & Younkin , 1997). Although not a risk factor for ischemic stroke in adults, the factor V Leiden mutation may be associated with in-utero cerbrovascular disease and hemiplegic cerebral palsy. Anticoagulants taken by the mother may be a factor in a fetal hemorrhagic stroke. Although heparin , which inhibits the activity of thrombin in coagulation of the blood, does not cross the placenta and cannot cause a fetal stroke, anti-epileptic medications may be associated with a decrease in vitamin K-dependent coagulation factors (Ozduman, et al., 2004). Varied outcomes and uncertainties regarding long-term prognosis following fetal strokes have been reported in the literature. Ozudman, et al., (2004) reported that 55% of the 22 children with a history of fetal stroke in their study were handicapped at follow-up ages of 3 months to 6 years. Sreenan, Bhargava, & Robertson (2000) reported that two-thirds of children in their study suffered from mental retardation, visual impairment, motor disabilities, or seizure disorders. Children with a history of fetal stroke may have good short-term outcomes but also the possibility of later onset of seizures, cognitive deficits, and sensory impairments (Roach et al., 2008; Sran & Baumann, 1988). Although motor deficits such as hemiplegia and asymmetries often are associated with neonatal infarction, signs of neuromotor impairment affecting speech, cognition, and behavior may also be evident at early school age. Neonatal or postnatal clinical evaluations have not always been predictive of outcomes; rather the extent of the damage as evidenced on the MRI is usually a better predictor (Mercuri, et al., 2004). Improvements in the diagnosis and understanding of the neuropathology underlying strokes in children have resulted in increased attention to implementing appropriate therapeutic interventions (Hartman, Lunney, & Serena, in press). Alternative and Augmentative Communication Management of the speech, language, and communication disorders which affect survivors of inutero stroke can require augmentative and alternative communication (AAC) strategies, particularly when the clinical picture includes cerebral palsy. Augmentative communication is operationally defined here as a system which supports or enhances currently existing language and communication abilities. Alternative communication refers to a system which replaces the communication of non-vocal individuals (Nicolosi, Harryman & Kresheck, 2005). According to the American Speech-Language-Hearing Association (ASHA, 2005, p. 1): AAC refers to an area of research, clinical, and educational practice. AAC involves attempts to study and when necessary, compensate for temporary 33 SLP- ABA Volume 5, Issue No. 1 or permanent impairments, activity limitations, and participation restrictions of individuals with severe disorders of speech-language production and/or comprehension. Modes of AAC Unaided AAC: Unaided AAC methods, such as sign language or gestural cueing systems, require no external device. American Sign Language (ASL) is a complex visual-spatial language that is used by the Deaf community in the United States and the English-speaking parts of Canada (Humphries & Padden, 2004). It is a linguistically complete and natural language. ASL encompasses hand gestures, facial expression, and the use of the space surrounding the signer to aid in the description of places and persons. Many signs represent ideas and are therefore iconic, us ing a visual image to represent a specific idea (Riekehof, 1987). A number of manual sign systems, including ASL, also have been used by individuals with severe communication disorders, but no hearing impairment (Beukelman & Mirenda, 2005). Unaided AAC requires a certain level of motor control to produce signs or gestures. This method of AAC has the advantage of speed, portability, and access to a wide number of messages, but it also has limitations. Signs require a certain level of fine motor dexterity, and there is a restricted set of potential listeners, as not everyone understands sign language (Wilkinson & Hennig, 2007). Aided AAC - Light Technology: Aided AAC involves an external component to communicate, using symbols or voice output. Light technology involves little to no technology (e. g., electronic output), but requires external aids of some sort. These may include alphabet boards, communication/ picture books/boards and communication programs such as the Picture Exchange Communication System (PECS). PECS was developed by Frost and Bondy (1998; 2002) as an augmentative and alternative communication method using operant-based procedures. Designed for children with autism and related developmental disabilities, it is a self-initiating and functional communication system that is rapidly acquired. PECS begins with the exchange of simple icons and builds sentence-like structures. It emphasizes a request function before the child responds to or comments about simple questions. An independent validation of PECS (Charlop-Christy, Carpenter, Le, LeBlanc, & Kellet, 2002) used a multiple-baseline design with three children with autism. All three children met the learning criterion for PECS, with concomitant increases in verbal language, as well as ancillary increases in socialcommunicative behaviors and decreases in problem behaviors. Light technology does not provide voice output, and therefore requires a communication partner to interpret the messages that the AAC user selects. Aided AAC- High Technology: Communication devices classified in the category of high technology may consist of a standalone device with voice output or a computer operating with communication software. A device designed specifically for communication is called a dedicated communication device, although it may be able to interface with a computer and perform environmental control functions. Computer communication devices are typically not considered to be dedicated devices, because communication is just one of the many software functions that can be accessed. The voice output provided by high technology has advantages over light technology, especially when the user is communicating in situations such as a classroom. Changes in technology continue to occur at a rapid pace with progressive changes in memory capacity, processing speeds, and battery life in high technology devices. The ability to integrate new technology such as digital cameras, a variety of software programs, scanners, and the ability to interface with the internet have resulted in a wider application of use and functions for individuals with severe communication impairments. Types of displays: There are two primary methods of displaying symbols for communication on AAC devices: fixed (or static) and dynamic displays. In fixed displays, pages or overlays containing symbols are set up on a board or on a simple voice output communication aid. Fixed displays usually require another person to change the pages/overlays if the AAC user cannot do so independently. These 34 SLP- ABA Volume 5, Issue No. 1 simple voice output devices generally operate via recorded voice. The content of the message can be recorded by a parent, professional, or another child , and can be easily re-recorded. Overlays can be designed around a specific topic or activity, offering opportunities for commenting and requesting using that vocabulary (e.g., house overlay, art activity). The advantages of these more simple fixed-display AAC devices include the ability to record a variety of voices, music, etc.; ease of recording; and relatively inexpensive cost. Disadvantages include the fact that the user is dependent on others to create and change the symbol overlays; only a limited number of symbols are available at one time, and this limits the number of possible messages and communicative interactions (Wilkinson & Hennig, 2007). Dynamic displays operate via communication software running on an electronic (aided) device, often a modified computer. Symbols can speak words/phrases/sentences via digitized voice. There can be links to different pages of symbols, activated via a navigation button on the device. The AAC user is not reliant on someone else to change an overlay to access additional vocabulary or messages (Wilkinson & Hennig, 2007). In either type of display, symbols should be organized in ways that promote efficient and effective communication (Beukelman & Mirenda, 2005). One of the more frequently used strategies involves organizing vocabulary according to event schemes, routines, or activities (Drager, Light, Speltz, Fallon, & Jeffries, 2003). Each display includes symbols for the vocabulary items that are relevant to the activity or routine (e.g., vocabulary for participation in morning circle). In this configuration, the use of single-meaning symbols can also support more complex linguistic functioning through symbol combinations and the use of sight words paired with the symbols (Wilkinson, Romski, & Sevcik, 1994). In addition to enhancing participation, schematic and activity displays can promote the use of multiword linguistic structures and facilitate receptive language growth and the development of syntactic skills (Beukelman & Mirenda, 2005). Roles of AAC as a Communication Mode: The primary role of AAC is to enhance or augment the expressive language skills of individuals who have severe communication impairments. There are no particular prerequisite cognitive skills or receptive language levels that need to be met before introducing AAC. The range of AAC options that are available makes it possible to address a variety of language impairments. Another role of AAC is to enable the user to express a range of communication functions across different environments and with a variety of communication partners. AAC may also serve to reduce challenging behaviors such as aggression, self-injurious behaviors, or behaviors resulting from frustration. PECS training with the population of individuals with autism has often resulted in a decrease in problem behaviors and an increase in verbal speech (Charlop-Christy, et al., 2002). Students with severe communication deficits have been taught to use assistive devices in everyday environments, resulting in decreased levels of problem behavior (Durand, 1999). AAC may provide a bridge to later linguistic development through the use of orthographic or other generative symbols (Wilk inson & Hennig, 2007). It is a common misconception that AAC may replace the possibility of speech as a mode of communication. The use of AAC may enhance existing speech skills among children with developmental and intellectual disabilities (Millar, Light, & Schlosser, 2006; Romski & Sevcik, 1996). The predominance of evidence supporting the mutual benefits of AAC to enhance speech development, as well as the acknowledged value of multimodal communication, result in reduced use of cognitive prerequisites as inclusion or exclusion criteria for services, particularly in younger populations (Wilkinson & Hennig, 2007). The behavioral paradigm of contingent reinforcement applies to AAC intervention for children with expressive speech and language delays. If the antecedent event is a symbol presented with the spoken word, as in high-technology AAC, and the consequent event is receipt of the labeled item, both the AAC mode and speech production should increase in frequency. A meta -analysis supported using 35 SLP- ABA Volume 5, Issue No. 1 AAC to facilitate production of natural speech as well as the development of communicative competence and language skills (Millar, et al., 2006). A number of teaching strategies have been associated with AAC interventions. In graduated prompting, the goal is to use a least-to-most cuing hierarchy (natural cue, expectant pause, general point & pause, and model), fading cues as soon as possible (Beukelman & Mirenda, 2005). In naturalistic, or milieu teaching, the emphasis is on teaching functional language skills in the context of common activities or routines. In this method, the facilitator initially provides verbal, gestural cues, modeling, or physical prompts to assist the individual to make requests. Requests are then followed by consequences that are functionally related (e.g., obtaining requested object/action) (Goodman & Remington, 1993; Kaiser, Yoder, & Keetz, 1992). The mand-model procedure has been effective in enhancing communication skills by obtaining the child’s focus, then delivering a mand (non-yes/no request or command), providing an interval for a response, and providing a model of the desired response, if needed (Venn, Wolery, Fleming, DeCesare, Morris, & Cuffs, 1993). Case Study In the present paper we use a case study design to examine in depth a specific individual in specific situations in order to illustrate important principles that might be overlooked in examining group data. Case study research also permits evaluation of phenomena that occur rarely and that may provide exceptions to generally accepted rules. Among weaknesses of case study research are the limitations of generalization and the increased likelihood of subjective biases on the part of the investigators. However, the factors that threaten internal validity in experimental research, especially history (events occurring between the first and second measurements in addition to the experimental variable) and maturation (changes in the subjects themselves that cannot be controlled by the experimenter and whose effects are attributed, incorrectly, to the experimental treatment), may be the substance of the case study approach (Schiavetti & Metz, 2006). This case study focused on the progression of AAC interventions for a young non-verbal child with a history of possible in-utero stroke and maladaptive behaviors. Often when young children have no consistent means of communication, they may express their wants and needs in socially unacceptable ways. AAC systems can help replace these maladaptive behaviors and often foster the development of natural speech (Cress & Marvin, 2003; Goldstein, 2002; Mirenda, 2003; Romski & Sevcik, 2005). Through this case research, we hope to support the evidence for the importance of individualized modality selection when making intervention recommendations. There has been an increase in the theoretical arguments for the use of AAC with young children but little research to inform clinical practice in this area (Mirenda, 2003). Decisions concerning appropriate AAC interventions must be made with considerations for the individual learners, in specific contexts to meet individual needs (Beukelman & Mirenda, 2005). Method Participant The child at the core of this case study, CM, a white, middle -class female, was age 4;5 (years; months) at the initiation of treatment discussed in this article. CM presented with a history of profound hearing loss in her right ear and a high frequency hearing loss in her left ear, as well as significant delay in speech skills, language skills, pragmatic skills, motor skills , and attention skills, which inhibited her participation in age-appropriate activities. According to parental report, CM had been receiving speechlanguage therapy focusing on American Sign Language (ASL) training. She also received physical therapy, occupational therapy, and parent training through the Committee on Pre-school Education. 36 SLP- ABA Volume 5, Issue No. 1 According to parental report, CM was using a Phonic Ear Solaris binaural FM hearing system with headset receivers in her school setting. FM hearing systems are personal wireless systems that utilize transmitters and receivers that are small enough to be worn on a person’s body. Generally, they are used to compensate for a hearing loss. This Phonic Ear FM system features a receiver that is approximately the size of a deck of cards and can be connected to a hearing aid or used with a head set, ear buds, or other accessories. Audiological Findings: CM was first seen for audiological testing at Adelphi University’s Hy Weinberg Center for Communic ation Disorders on May 10, 2005 (CA= 5:7). The evaluation continued for a total of eight sessions, over a period of two months. Multiple sessions were needed because of her tantrums and refusal to comply with assessment procedures. CM was accompanied to each test session by her mother, who served as informant. Otologic history indicated that CM passed her neonatal hearing screening, bilaterally. At about 3 ½ years of age, CM reportedly contracted scarlet fever. In August of 2003 (CA=3:10), sound field test results from another facility revealed a mild hearing loss in at least one ear. Type A tympanograms were obtained, bilaterally, consistent with normal middle ear function in each ear. Subsequent auditory brainstem response (ABR) testing (CA=4:0) yielded results consistent with a mild hearing loss at 2000 Hz in the left ear and a profound hearing loss in the right ear. Type A tympanograms were again obtained, bilaterally. An FM unit was recommended in November of 2003 for use during group language-based activities in the classroom and for individual speech sessions. On the initial test date at the Adelphi University facility, spondee recognition threshold (SRT) testing and tympanometry were completed. An SRT of 10 dB HL was obtained in the left ear; no response was obtained when testing the right ear. Type C tympanograms were obtained, bilaterally, indicating significant negative pressure in each ear. On subsequent test dates (4 sessions), sound field warble tone thresholds were obtained at levels between 10 dB HL and 25 dB HL for octave frequencies 500 Hz through 4000 Hz, suggesting adequate hearing through the speech frequencies in the better ear. Tympanometric testing reflected normal middle ear function on the last test date. Bone conduction testing was attempted in two subsequent sessions, but was abandoned, due to CM’s rejection of the instrumentation. In her final testing session, an SRT of 15 dB HL was obtained in the left ear and an SRT of 25 dB HL was obtained in the right ear via bone conduction. Type A tympanograms were obtained, bilaterally. At age 5:9, CM participated in transient evoked otoacoustic emission (TEOAE) testing. Left ear responses were obtained at frequencies from 1000 to 4000 Hz. Right ear responses were obtained at 2800 Hz and 4000 Hz. EOAEs reflect active cochlear processes and the presence of TEOAEs suggest hearing levels to be no poorer than 30 dB HL. The right ear responses were clearly unexpected, given the previously reported profound hearing loss. It is quite possible that these responses were artifacts, resulting from the equipment or poor probe placement. However, it is also possible that there are some surviving outer hair cells in the 2800 Hz to 4000 Hz cochlear region. A similar case was reported by Prieve, Gorga , and Neely (1991), in which an adult with a bilateral severe-to-profound hearing loss was found to have EOAEs in a restricted frequency in one ear. Unfortunately, CM did not return to this Center following this last test date, and the finding in this case could not be confirmed with a different measurement system (as was done in the Prieve, et al. study). An alternative conclusion, based on neurologic and audiometric findings, is that part of the behavior that had been attributed to a profound hearing loss may have resulted from an auditory processing deficit secondary to in-utero CVA. MRI Results: Following oral sedation with chloral hydrate, an MRI study of the brain was conducted when CM was age 1:11. T1, T2, and FLuid-Attenuated Inversion-Recovery (FLAIR) imaging 37 SLP- ABA Volume 5, Issue No. 1 were incorporated, encompassing all three imaging planes. The second author (RG) added ultraviolet digital enhancement to the image in Figure 1. The MRI study demonstrated scattered abnormalities within the white matter of both hemispheres, as seen on a CT examination. There were hyper-intense signals on the FLAIR and T2 images within the periventricular white matter, most notable within the peri-trigonal white matter. In addition, there were multi-focal, more discrete areas of abnormal signal throughout the frontal, parietal, and temporal lobes, bilaterally. These were scattered within the centrum semiovale. When evaluating the white matter, there was an age -appropriate pattern of myelination, landmarks achieved. There was mild relative prominence of the lateral ventricles, without evidence for acute hydrocephalus. The major vessels were grossly patent, with no gross malformation. The skull base appeared intact. While there is normal immature periventricular white matter seen on T2 imaging in children, the abovementioned findings are more pronounced. The hyper-intense signals on the FLAIR and T2 images within the periventricular white matter support a consideration of an in-utero stroke, where periventricular venous infarction may result in a focally enlarged lateral ventricle . Compared to the CT examination, the abnormality revealed in the MRI is more generalized but less pronounced. Figure 1. MRI of CM’s brain. Settings PECS training and other AAC interventions were conducted at a university speech and hearing center in a suburb of New York City for 10 weeks, 2 times per week for 30-minute durations. The therapy room contained a small table and two chairs positioned facing each other. Sessions were able to be viewed via a camera system with the capability of video-taping. The room was free of visual distractions and external auditory distractions were minimal. A graduate student in Speech-Language Pathology was the primary interventionist, implementing the six phases of PECS as recommended by Frost & Bondy (2002) and later introducing Boardmaker communication boards, the 7-Level Communication Builder, and the Dynamo. A clinically certified speech-language pathologist/university professor supervised 100% of the therapy sessions and a certified speech-language pathologist with extensive AAC experience provided recommendations for the introduction of AAC materials. CM was not receiving any concurrent speech-language intervention in another setting during the time she was seen at the university clinic. 38 SLP- ABA Volume 5, Issue No. 1 Materials PECS Materials : Individual cards using colored symbols from The Picture Communication Symbols Combination Book (Mayer-Johnson Company, 1994) were used as recommended by Frost and Bondy (1994) during Phases I and II of PECS training. These pictures were compiled into a PECS communication book during Phases II, III and IV. In Phase IV a sentence strip was added using a Velcro strip at the bottom of the binder with the carrier phrase “I want.” Boardmaker: Boardmaker is a graphic database that contains upward of 3,000 picture communication symbols. Each symbol may be translated into various languages and may be printed with or without text on the symbol. Boardmaker symbols were used to construct supplemental communication boards with a schematic or topic orientatio n and they were also used as overlays for the 7-Level Communication Builder. 7 – Level Communication Builder (manufacturer: Enabling Devices): This is a self-contained speech generating device (SGD) using recorded speech and requiring paper overlays that need to be changed to correspond with each of the 7 levels. It allows the user to use 1, 2, 4, 8, or 16 different messages per level, giving a possible 112 messages (in the 16 window setting). Dynamo (manufacturer: DynaVox): The Dynamo is a small, portable SGD using digitized speech. It has a dynamic black and white screen display and allows access to several levels. Clinical Intervention Baseline behaviors including sitting with trunk alignment, eye contact, and ability to transition across activities without tantrums were identified to be targeted as goals, and were calculated across the 10-week period. Individual picture cards using colored symbols from The Picture Communication Symbols Combination Book were introduced and PECS (Frost & Bondy, 2002) training was initiated. CM learned PECS in the six phases recommended by the authors. Phase I, How to Communicate, had the goal of training the motor response after initiation. Accordingly, CM was taught to pick up, reach, and release the stimulus picture. For the two-person prompt procedure, one clinician, who stood in front of the child, served as her communication partner, while a second clinician acted as a physical prompter behind the child, aiding in the motor response. The clinician in front used an empty hand to provide information about where to place the picture, and opened once CM reached for the picture or initiated an action. The empty hand was not used as a prompt. A consequent event (secondary reinforcement) was delivered within 0.5 sec of CM’s response, and generally took the form of social praise. Phase II, Distance and Persistence, had the goal of persistence across obstacles. CM was trained to increase the distance she traveled for the picture, or to deliver the picture to her communication partner or to her PECS book. CM was also trained to gain the attention of the clinician whose back was turned, to carry her PECS book, and to request action during activities. Phase III, Discrimination, had the goal of demonstrating ability to select from all pictures in the PECS book. In a two-way discrimination task, CM had to select one of two choices, only one of which resulted in reinforcement. When provided with multiple choices of preferred items, CM’s actions had to match her request, and she learned to demonstrate correspondence between picture and item. Phase IV, Sentence Structure, had the goal of building toward commenting. The carrier phrase, “I want” was combined with a symbol of a desired item on the sentence strip of CM’s PECS communication book. After combining the symbols on the sentence strip, CM was expected to give it to 39 SLP- ABA Volume 5, Issue No. 1 her communication partner (clinician) and the clinician then read the strip. The clinician added attributes, such as colors, to lengthen the sentence strip. Phase V, Responsive Requesting, had the goal of responding to the question, “What do you want?” In the first step, the clinician simultaneously asked a question (which served as a cue), and pointed to the sentence starter or carrier phrase (which served as a prompt). The clinician gradually increased the interval between cue and prompt until CM was able to “beat” the prompt. Phase VI, Commenting, had the goal of spontaneous commenting on the environment. CM was prompted to respond to the question, “What do you see?” and then to comment by combining the symbol for “I see” with an object symbol on her sentence strip. Successful completion of each phase was attained using a criterion of 80% unprompted successful trials over two successive sessions. Communication boards using symbols from Boardmaker software were introduced concurrent with the last phases of PECS training to motivate communication during preferred activities. Words were always included with picture symbols to enhance pre-literacy skills (see Fig. 2). A schematic or topic orientation was used, for example, a board with symbols for materials needed for an art activity. Topic strategies allow beginning communicators to initiate and establish topics of communication using various types of basic symbols. Initial training on a fixed display is often desirable to achieve symbol recognition (Beukelman & Mirenda, 2005). These same (fixed) overlays were transferred onto the 7-Level Communication Builder, an SGD with recorded voice output. Finally, the same overlays were programmed onto the dynamic display on the SGD Dynamo. Support for a transition to an SGD with a dynamic display comes from observations that speed and accuracy of use with fixed displays diminishes after 8-9 practice sessions (Hochstein, McDaniel, & Nettleton, 2004; Reichle, Dettling, Drager, & Leiter, 2000). The natural branching capabilities of the dynamic screens were used to promote phrase and sentence construction within specific activities. The use of the SGD to support more complex linguistic functioning through symbol combinations and the use of sight words paired with the symbols was the primary focus to support pre-literacy skills. The same criterion of 80% unprompted successful trials over two successive sessions was used to achieve esta blished goals. During training with these AAC modes, the mand-model procedure (Venn, et al., 1993) was initially used. CM’s attention was obtained via verbal/visual prompting, a mand was delivered (non-yes/no request or command), an interval for response was given, and a model was provided, if necessary. Intervention then incorporated more of a graduated prompting strategy utilizing natural cues in context, expectant pauses, and modeled pointing to symbols , if necessary, fading cues as soon as possible. Long-term follow-up was conducted by interviewing the parent four years after the presented therapy protocol was completed. Results Receptive Language : By the conclusion of the 10-week period, CM reached established criterion levels (80%) for following two-step directives and receptively identifying colors, nouns, verbs, adjectives and prepositions by pointing to the symbol representing the words on fixed and dynamic displays. The ability to point to letters of the alphabet when named and some sight words was emerging. Acquisition of PECS: CM advanced through the PECS phases rapidly and met criterion (80% correct) for each phase. 40 SLP- ABA Volume 5, Issue No. 1 Figure 2. PECS stimuli. Expressive Communication: At the initiation of therapy, CM exhibited significantly delayed expressive language skills, using some signs/gestures and sound combinations to communicate. Her limited communication skills may have been contributing to her problem behaviors (e.g. tantrums). During the course of therapy, the following variables were calculated: number of responses using AAC modes; number of requests using AAC modes; number of word approximations/words used per session. CM increased her responses (e.g. to a mand-model) using AAC modes from 3 out of 15 to 14 out of 15 trials over the 10-week period. Her requests increased from 1 out of 6 trials to 19 out of 20 trials (see Fig. 3). CM’s use of word approximations/words increased from 2 in the initial session to more than 20 in the final session (see Fig. 4). Using PECS and Boardmaker boards, CM was able to label transitive verbs, independently initiate a conversational exchange, and request action. By the last therapy session, CM had attempted the following words and/or word approximations: go, happy, house, radio, kitchen, cake, table, fruit, bath, computer, pizza, Dad, banana, TV, apple, pool, crab, ice, one, three, and five and the phrases ‘I go’ and ‘I want.’ Behaviors : Appropriate sitting behavior, defined by postural (trunk) alignment, remaining in the chair and eye contact with the clinician increased over the 10-week period from 0% to 75% of observed intervals. CM’s ability to achieve and maintain seating posture and eye contact became more spontaneous, requiring fewer verbal/physical cues from the clinician. Tantrums, secondary to CM’s difficulties with transitioning across activities were observed in 90% of trials at baseline, reducing to 30% at the conclusion of PECS training and to 0% after the AAC devices (SGDs) were introduced (see Fig. 5). Long-term follow-up was conducted via interview with the parent when CM entered the 4th grade in a public school setting. According to parental report, CM is no longer using her SGD or any other form of AAC. Her vocal speech has flourished, and CM is clearly understood by most listeners. She is receiving speech-language therapy five times a week in her school setting and once a week privately. CM is awaiting the repair of her FM system, which she continues to use in her classroom settings. According to the results of her New York State 3rd grade assessment tests, she scored above average in the English portion and met the learning standard in the mathematics portion. She is also adjusting well to pushing into mainstream 4th grade settings. 41 SLP- ABA Volume 5, Issue No. 1 30 Responses/Requests 25 20 15 10 5 0 1 2 3 4 5 6 7 8 9 10 Sessions Fig. 3. Responses/Requests Using AAC Weeks 2-5: PECS/Communication Boards Weeks 6-7: 7 Level Communication Builder Weeks 8-10: Dynamo Word Approximations/Words 25 20 15 10 5 0 1 2 3 4 5 6 7 8 9 10 Sessions Fig. 4. Number of word approximations used spontaneously. 42 SLP- ABA Volume 5, Issue No. 1 100 90 80 Tantrums 70 60 50 40 30 20 10 0 1 2 3 4 5 6 7 8 9 10 Sessions Fig. 5. Percentage of tantrums observed during transitions across activities. Discussion The child in this case study (CM) had previously been diagnosed with a profound hearing loss. Subsequent audiological assessments using TEOAE testing reflected active cochlear processes. These surprising results may suggest the possibility of some surviving outer hair cells in the cochlear region. Alternatively, since CM’s MRI results reflect the possibility of an in-utero stroke, auditory processing deficits may account for the inconsistencies in performance. The importance of thorough diagnostic and medical evaluations, especially for the young, non-verbal child is apparent. Bax, Tydeman, and Flodmark (2006) found white-matter damage of immaturity, including periventricular leukomalacia (PVL) to be the most common finding in brain MRI scans of children with diagnoses of cerebral palsy. The authors recommended that all children with cerebral palsy should have an MRI scan to determine the extent of the damage, with the assurance that cranial MRI is a procedure that is safe to use with the pediatric population. Strokes occurring between 28 weeks gestation and four weeks postnatally are seen in at least 1 in 4,000 live births per year. These children may be at risk for long-term learning, language and behavior diffic ulties (McBride, 2003). The destruction of cerebral white matter or extrapyramidial tracts may play a prominent role in disturbances of motor control, including speech (Paneth, Korzeniewski, & Hong, 2005). Over the course of therapy, AAC interventions were introduced to CM to increase her communication skills. Steadily, improvements in sitting behaviors and eye contact were observed. At the same time, CM’s problematic behaviors (e.g., tantrums) decreased significantly and she was able to transition across activities with more ease. Brady (2000) reported successful use of an SGD to request by a five-year-old child with autism, resulting in increased comprehension of object names. The ability to 43 SLP- ABA Volume 5, Issue No. 1 use SGDs to communicate has been associated with a concurrent reduction in the frequency of problem behaviors such as tantrums (Durand, 1999). Parents and professionals may be reluctant to initiate AAC interventions , because of the misconception that AAC may inhibit the emergence of natural speech production (Beukelman, 1987; Silverman, 1995). Individuals with developmental disabilities are often perceived to have the tendency to rely on methods other than speech, such as manual signs or picture communication boards (Glennen & DeCoste, 1997). However, natural speech is certainly the most efficient and expedient means of communication. Children typically chose the easiest, most efficient mode of communication, if it is within their capabilities (Millar, et al., 2006). Support also comes from the belief that AAC provides an immediate and consistent model, along with reinforcement for individuals with developmental disabilities, especially when there are visual stimuli and voice output (Blischak, 2003; Romski & Sevcik, 1996). It has also been proposed that AAC interventions serve as a mechanism for individuals with severe speech impairments, to bypass the cognitive and motor demands of speech production (Romski & Sevcik, 1996). The increases in CM’s communication skills, including requesting, responding, and production of words/word approximations concurrent with the use of AAC strategies, support the conclusion that AAC can enhance communicative competence and language skills. Millar, et al. (2006) found that 94% of the participants in their meta-analysis review demonstrated an increase in speech production during or following at least one in a range of AAC interventions. They concluded that the gains in speech production were observed shortly after the introduction of AAC interventions, supporting the theory of automatic reinforcement. Too often, AAC is viewed as a last resort for individuals with developmental disabilities. Evidence supports the early introduction of AAC to facilitate communicative competence and language skills , and the development of natural speech (Millar, Light, & Schlosser, 1999). The implementation of AAC can set the stage for further language and communication development during the preschool and early school years (Romski & Sevcik, 2005). Clinical Implications The results of this case research support the importance of individualized modality selection when implementing AAC strategies. Implementation of PECS and SGDs using mand-model procedures and graduated prompting resulted in a reduction in problematic behaviors and a subsequent improvement in receptive language and expressive communication. Delaying the introduction of AAC strategies when behaviors and verbal performance suggest high risk for speech/language impairment can be detrimental to a child’s long-term speech and language development (Cress & Marvin, 2003). AAC should not be viewed as a last resort but rather as an early course of intervention that can provide a foundation for the development of verbal language and communicative competence. References ASHA (2005). Roles and responsibilities of speech-language pathologists with respect to alternative communication: Position statement. Barmada, M. A., Moossy, J., & Shuman, R. M. (1979). Cerebral infarcts with arterial occlusion in neonates. Annals of Neurology, 6, 495-502. Bax, M., Tydeman, O. & Flodmark, O. (2006). Clinical and MRI correlates of cerebral palsy. JAMA, 296: 1602-1608. Beukelman, D. & Mirenda, P. (2005). Augmentative and alternative communication: Management of rd severe communication impairments (3 edition). Baltimore: Brookes. 44 SLP- ABA Volume 5, Issue No. 1 Brady, N. (2000). Improved comprehension of object names following voice output communication aid use: Two case studies. Augmentative and Alternative Communication, 16, 197-204. Chalmers, E. A. (2005). Perinatal stroke—risk factors and management. British Journal of Haematology, 130, 333-343. Charlop-Christy, M. H., Carpenter, M., Le, L., LeBlanc, L. A., & Kellet, K. (2002). Using the picture exchange communication system (PECS) with children withautism: Assessment of PECS acquisition, speech, social-communication behavior, and problem behavior. Journal of Applied Behavior Analysis, 35, 213-231. Cress, C. & Marvin, C. (2003). Common questions about AAC services in early intervention. Augmentative & Alternative Communication, 19, 254-272. Drager, K., Light, J., Speltz, J., Fallon, K. & Jeffries, L. (2003). The performance of typically developing 2½-year-olds on dynamic display AAC technologies with different system layouts and language organizations. Journal of Speech, Language and Hearing Research, 46, 298-312. Durand, V.M. (1999). Functional communication training using assistive devices : Recruiting natural communication of reinforcement. Journal of Applied Behavior Analysis, 32, 247-267. Elchalal, U., Yagel, S., Gomori, J. M., Porat, S., Beni-Adani, L., Yanai, N., & Nadjari, M. (2005). Fetal intracranial hemorrhage (fetal stroke): Does grade matter? Ultrasound Obstetrics and Gynecology, 26, 233-243. Frost, L. A., & Bondy, A. S. (1998). The picture exchange communication system. Seminars in Speech and Language, 19, 373-389. Frost, L. A., & Bondy, A. S. (2002). The picture exchange communication system training manual (2 ed.). Newark: Pyramid Education Products. nd Goldstein, H. (2002). Communication intervention for children with autism : A review of treatment efficacy. Journal of Autism & Developmental Disabilities, 32, 373-396. Goodman, J. & Remington, B. (1993). Acquisition of expressive signing : Comparison of reinforcement strategies. Augmentative & Alternative Communication, 9, 26-35. Hochstein, D., McDaniel, M. & Nettleton, S. (2004). Recognition of vocabulary in children and adolescents with cerebral palsy : A comparison of two speech coding schemes. Augmentative and Alternative Communication, 20, 45-62. Huang, Y. F., Chen, W. C., Tseng, J. J., Ho, E. S., & Chou, M. M. (2006). Fetal intracranial hemorrhage (fetal stroke): Report of four antenatally diagnosed cases and review of the literature. Taiwan Journal of Obstetrics and Gynecology,45, 135-141. nd Humphries, T., Padden, C. (2004). Learning American Sign Language (2 ed.). Boston: Pearson Education, Inc. Kirton, A., Deveber, G., Pontigon, A. M., Macgregor, D., & Shroff, M. (2008). Presumed perinatal ischemic stroke: Vascular classification predicts outcomes. Annals of Neurology, 63, 413-415. Lou, H. C. (1983). Perinatal cerebral ischaemia and developmental neurologic disorders. Acta Pediatrica Schandinavia Supplement, 311, 28-31. McBride, G. (2003). Neonatal Stroke : More Questions than Answers. Neurology Today, 3 (2), 14 – 17. Millar, Light, & Schlosser (2006). The impact of augmentative and alternative communication intervention on the speech production of individuals with deve lopmental disabilities : A research review. Journal of Speech, Language, and Hearing Research, 49, 248–264. 45 SLP- ABA Volume 5, Issue No. 1 Ment, L. R., Duncan, C. C., & Ehrenkranz, R. A. (1984). Perinatal cerebral infarction. Annals of Neurology, 16, 559-568. Mercuri, E., Barnett, A., Rutherford, M., Guzzetta, A., Haataja, L., Cioni, G., Cowan, F. & Dubowitz, L. (2004). Neonatal cerebral infarction and neuromotor outcome at school age. Pediatrics, 113, 95100. Mirenda, P. (2003). Toward functional augmentative and alternative communication for students with autism : Manual signs, graphic symbols & voice output communication aids. Language, Speech & Hearing Services in Schools, 34, 203-216. th Nicolosi, L., Harryman, E., & Kresheck, J. (2005). Terminology of communication disorders (5 ed.). Baltimore: Williams & Wilkins. Ozduman, K., Pober, B. R., Barnes, P., Copel, J. A., Ogle, E. A., Duncan, C. C., & Ment, L. R. (2004). Fetal stroke. Pediatric Neurology, 30, 151-162. Paneth, N., Korzeniewski, S. & Hong, T. (2005). The role of the intrauterine and perinatal environments in cerebral palsy. NeoReviews, 6, 133-140. Pierrat, V., Cneude, F., Duquennoy, C., & Lequien, P. (1996). [Cerebral infarction: Ultrasonic diagnosis and semiologic peculiarities in premature newborn infants] [Article in French]. Archives of Pediatrics, 3, 137-140. Prieve, B.A., Gorga, M.P., & Neely, S.T. (1991). Otoacoustic emissions in an adult with severe hearing loss. Journal of Speech and Hearing Research, 34, 379-385. Reichle, J., Dettling, E., Drager, K. & Leiter, A. (2000). Comparison of correct responses and response latency for fixed and dynamic displays : Performance of a learner with severe developmental disabilities. Augmentative and Alternative Communication, 16, 154-163. nd Riekehof, L. L. (1987). The joy of signing (2 ed.). Missouri: Gospel Publishing House. Roach, E.S., Golumb, M.R., Adams, R., Biller, J., Daniels, S., deVeber, G., Ferriero, D., Jones, B., Kirkham, F., Scott, R.M. & Smith, E.R. (2008). Management of stroke in infants and children. Stroke, 39, 2644-2691. Romski, M. & Sevcik, R. (1996). Breaking through the speech barrier: Language development through augmented means. Baltimore : Brookes. Romski, M. & Sevcik, R. (2005). Augmentative communication and early intervention, Myths & realities. Infants & Young Children, 18, 174-185. th Schiavetti, N., & Metz, D. E. (2006). Evaluating research in communicative disorders (5 ed.) Boston: Peason, Allyn and Bacon. Sran, S.K. & Baumann, R.J. (1988). Outcome of neonatal strokes. American Journal of Disabled Children, 142, 1086-1088. Sreenan, C., Bhargava, R.. & Robertson, CM. (2000). Cerebral infarction in the term newborn: Clinical presentation and long-term outcome. Journal of Pediatrics, 137, 351-355. Takanashi, J., Barkovich, A. J., Ferriero, D. M., Suzuki, H., & Kohno, Y. (2003). Widening spectrum of congenital hemiplegia: Periventricular venous infarction in term neonates. Neurology, 61, 531533. Takanashi, J., Tada, H., Barkovich, A. J., & Kohno, Y. (2005). Magnetic resonance imaging confirms periventricular venous infarction in a term-born child with congenital hemiplegia. Developmental Medicine and Child Neurology, 47, 706-708. 46 SLP- ABA Volume 5, Issue No. 1 Thorarensen, O., Ryan, S., Hunter, J., & Younkin, D. P. (1997). Factor V Leiden mutation: An unrecognized cause of hemiplegic cerebral palsy, neonatal stroke, and placental thrombosis. Annals of Neurology, 42, 372-375. Venn, M., Wolery, M., Fleming, L., DeCesare, L., Morris, A., & Cuffs, M. (1993). Effects of teaching preschool peers to use the mand-model procedure during snack activities. American Journal of Speech-Language Pathology, 38-46. Wilkinson, K. & Hennig (2007). The state of research and practice in augmentative and alternative communication for children with developmental/intellectual disabilities. Mental Retardation and Developmental Disabilities Research Reviews, 13, 58-69. Wilkinson, K., Romski, M. & Sevcik, R. (1994). Emergence of visual-graphic symbol combinations in children with mental retardation using an augmented communication system. Journal of Speech and Hearing Research, 37, 883-896. Author Information Cindy G. Arroyo D.A., Assistant Professor Communication Sciences & Disorders Adelphi University 401 Franklin Avenue Room 2442 Garden City, New York 11530 Phone: (516) 877 - 4768 e-mail: [email protected] Robert M. Goldfarb Ph.D., Professor and Program Director Communication Sciences & Disorders Adelphi University 401 Franklin Avenue Room 2442 Garden City, New York 11530 Phone: (516) 877-4785 e-mail: [email protected] Janet Schoepflin Ph.D., Associate Professor and Program Chair Communication Sciences & Disorders Adelphi University 401 Franklin Avenue Room 2442 Garden City, New York 11530 Phone: (516) 877-3343 e-mail: [email protected] Danielle Cahill M.S. Adelphi University 401 Franklin Avenue Room 2442 Garden City, New York 11530 Phone: (516) 877 - 4770 e-mail: [email protected] 47 Implications of Skinner’s Verbal Behavior for Studying Dementia\\ Jeffrey A. Buchanan, Daniel Houlihan, & Peter J.N. Linnerooth Abstract Persons with dementia experience continual declines in a number of abilities. Language abilities are particularly hard hit and become increasingly impaired as the underlying disease progresses. These language impairments make verbal communication very challenging for family and professional caregivers. As a result, caregivers may inadvertently punish verbal behavior, thereby exacerbating the deterioration of verbal repertoires. Although the topography of language impairments associated with dementia have been well described, less empirical work has been conducted concerning how to minimize these impairments and their deleterious effects. In 1957 B.F. Skinner outlined his conceptualization of language and cognition in his book Verbal Behavior. This paper will explore the implications of Skinner’s Verbal Behavior for studying communication impairments associated with dementia. Keywords : elderly; dementia; verbal behavior; communication Overview of Dementia “Dementia” is a generic term that describes the progressive decline in a number of cognitive abilities such as attention, memory, language, perception, and reasoning that interferes with daily functioning (APA, 2000). Dementia can also result in behavioral changes such as wandering or aggression, declines in self-care skills, and mood disturbances such as depression or anxiety. Dementia can be caused by a number of different, irreversible causes (e.g., Alzheimer’s disease, vascular disease, Pick’s disease) as well as reversible causes (e.g., vitamin B12 deficiency, medication overdose). Dementia is a serious public health concern in the United States. If one considers all causes of dementia, approximately 6-10% of individuals over the age of 65 suffer from dementia, with Alzheimer’s Disease (AD) accounting for approximately 65% of all cases (Hendrie, 1997). Age is the number one risk factor for developing dementia, which is particularly concerning given the rapidly aging population of the United States. It is projected that 13 to 15 million Americans could suffer from AD alone by the year 2050 if no cure is found (Hebert, Scherr, Bienias, Bennet, & Evans, 2003). Language deficits associated with dementia. Language deficits (i.e., aphasia) associated with dementia can take on various forms/topographies, including receptive and expressive language deficits. Language deficits in the early stages of dementia are characterized by pronounced difficulties with speech production as opposed to comprehension (Levine, 2006). Common early stage deficits include word finding deficits (i.e., anomia), poor spontaneous writing, indefinite references (“it”, “those”, or “thing”), repetition of words or ideas, and difficulty understanding complex language such as metaphors or analogies (APA, 2000; Kempler, 1991; Orange, 2001). As the disease progresses deficits such as empty content, inappropriate word substitutions, difficulties following multi-step commands, poor topic maintenance and inappropriate topic shifts, reduced reading comprehension, frequent digression from conversational topics, reduced verbal fluency, difficulties with turn-taking and producing fewer utterances per conversational turn (APA, 2000; Kempler, 1991; Orange, 2001; Levine, 2006) become apparent. In the final stages of the disease, the patient may engage in echolalia, produce continuous streams of nonsensical language, or become mute (Orange, 2001). These language deficits have important everyday implications for patients, and caregivers, who have described communication deficits as one aspect of dementia with which it is particularly challenging to cope (Orange, 1991). The many implications of language deficits associated with dementia will be further described in the sections below. 48 Skinner’s Conceptualization of Verbal Behavior and Its Implications for Dementia Skinner (1957) distinguished and functionally defined 6 types of verbal behavior including the tact, mand, echoic, textual, intraverbal and autoclitic. To say that these are defined functionally means the categories are delineated according to the reinforcement relations that shaped and/or maintain them as operant responses, not according to any topographical property of the word(s). For example, the word “water” could be an example of any of the categories, depending upon the conditions that evoke the response. Shouting “Water!” whenever a river or lake is visible is a tact whereas the same, “Water!” emitted after an hour in warm sun, even if perfectly similar in tone, inflection, or all formal properties, is a mand, for it is controlled very differently, and has a much different ultimate function. With this critically important difference between the Skinnerian versus traditional analyses (that of grammarians, for example) in mind, the 6 categories are defined in the following sections along with a discussion of a seventh variable – the audience. The sections below will cover the following information: 1) a short overvie w and definition of each verbal operant. 2) Examples that illustrate how each verbal operant may be relevant in the context of dementia, with a particular emphasis on the potential negative implications of a breakdown in verbal functioning. Examples involving both higher and lower functioning patients will be provided to emphasize that individuals with dementia vary in their degree of functional impairment and challenges experienced by patients and caregivers differ at different points in the disease process. 3) When available, examples of empirical literature relevant to the particular verbal operant will be presented. The discussion of empirical literature is not meant to be exhaustive, but illustrative of the work that has been done examining verbal behavior in persons with dementia. Tact Tacts (verbal behavior that contacts the environment) occur under the control of a (usually nonverbal) antecedent stimulus. The discriminative stimulus that controls a given tact might be an event (“It is lunchtime”), an object (“Hamburgers are on the menu”), or a property of either (“My hamburger is too rare”). Tacting is frequently likened to the conventional term “informing,” (Baum, 2005) but Skinner (1957) is careful to point out 2 things: First, a given tact is simply a bit of verbal behavior made more likely by the presence of a certain stimulus (1957, p. 82). In other words, the antecedents controlling it must drive our analysis, not any idea about the “meaning” of what is said. Second, as a functional class of verbal behavior, tacting encompasses a diversity of forms, including announcing, proclaiming, stating and naming (1957, p. 186). The control of tacting by prior stimuli is the essential property to grasp. For example, the third author is always amused when his elderly grandparents visit him in a new city. They spend their entire time on any car ride emitting tacts evoked by commonplace stimuli (e.g. “There’s a hamburger restaurant just down the street!”). Presumably, the novel setting for the “golden arches” evokes the tact. No such effect would be seen in a longtime resident, and a person familiar with the community even ends up hard-pressed to reinforce such tacts. Dementia clearly impairs tacting throughout the disease process. For example, in the early stages of dementia, even familiar objects (e.g., family members, friends, utensils, animals) inadequately control tacting. This is often referred to as “anomia”, or the inability to name objects or people accurately. As the disease progresses, the individual has particular difficulty tacting private stimuli such as the urge to use the restroom, thirst, pain, or aversive emotional states such as fear or anger. The negative effects of impaired tacting are evident. For example, forgetting a grandchild’s name can be very upsetting for the child as well as the patient. Another common example in long-term care facilities involves the inability to tact physical pain. Patients may engage in “disruptive behaviors” such as loud moaning or physical aggression when they cannot say “I’m in pain” or “that hurts.” Prolonged suffering, loss of dignity, and inadequate care are but a few consequences of this breakdown in the ability to tact. 49 Few empirical studies have attempted to improve tacting in dementia patients, but some intriguing findings exist. For example, Cameron Camp and his colleagues have investigated the effectiveness of a procedure called spaced retrieval (SR) to improve tacting in persons with dementia. SR is a shaping paradigm that involves giving a person practice at successfully recalling information over successively longer time periods (Camp, 1998). SR has been found to assist dementia patients in learning the names of people (Camp & Schaller, 1989) and objects (Abrahams & Camp, 1993). SR appears to represent a simple, portable, and teachable intervention that can reestablish the ability to name specific objects or people over relatively long periods of time (e.g., a week or more) in persons with mild to moderate dementia. Another line of empirical work designed to improve tacting involves the use of memory books that consist of a series of bound pages with a picture, and sometimes a description, of a person, place, or symbol that is meaningful to the patient. Bourgeois and her colleagues (1992, 1993; Hoerster, Hickey, & Bourgeois, 2001) investigated the effects of an external communication aid called personalized memory books for improving communication between dementia patients and caregivers. These studies have demonstrated positive outcomes in terms of tacting including more novel, detailed, and factual statements during conversations and more on-topic verbalizations. Additional benefits include better turn-taking; reductions in negative caregiver interactions; caregivers asking fewer questions; and conversation that is more focused on the patient (Hoerster, et al.). Mand Mands (as in “demand” or “command”) are verbal operants whose likelihood of occurrence and form are controlled by: 1) a given state of deprivation or aversive stimulation and; 2) a specific type of reinforcement (Pear, 2001; Skinner, 1957). Unlike tacting, manding has no particular relation to an antecedent stimulus. Mands such as “Please turn on the light,” typically produce a reinforcer (the listener turns on a light) appropriate to the state of deprivation or aversive stimulation (e.g., perhaps the speaker cannot see to read). Most mands specify the appropriate reinforcer (Baum, 2005) and many specify the behavior of the listener as in the preceding example. The speaker then often provides generalized conditioned reinforcement of the listener’s behavior, such as praise, or a “thank you.” Like tacting, manding occurs in a diversity of forms such as requests for assistance, gesturing or advice given to others. For example, a nursing home resident who cannot hear the dayroom T.V. might mand “Could you please turn up the volume?” a behavior reinforced by the health care aide (the listener) adjusting the sound. As the ability to mand in a precise manner deteriorates as the disease progresses, difficulties between caretakers and patients commonly occur. For instance, the first author witnessed a situation where a patient rolled his wheelchair to the nurse’s station and began moaning and pointing to his foot. The nursing assistant, confused by the patient’s behavior, sternly asked him to quiet down. The patient subsequently began moaning and yelling louder while continuing to point at his foot. As the situation escalated, a more experienced nursing assistant came by, moved the patient’s foot onto the footrest of his wheelchair, and the patient consequently stopped moaning. This incident represents a larger issue commonly encountered in long-term care facilities, namely that severely impaired patients may engage in “socially inappropriate” behavior (e.g., loud moaning) or vague gestures or verbalizations (e.g., pointing to one’s foot, one-word utterances such as “foot” or “there”) that function as imprecise mands. Staff that has little shared history with the patient consequently respond ineffectively (e.g., reprimand, speak louder, ask for clarification), which results in aversive interactions with staff as in the example above. For those patients with more severe language impairments, external aids such as picture books may help supplement mands in that individuals can also point to pictures of desired objects, activities, or people. In the example above, a simple picture of a foot sitting on a footrest (along with other pictures of commonly desired objects) was attached to the patient’s wheelchair and resulted in reduced staff-patient conflicts. 50 Echoic Echoic responses are controlled by specific auditory stimuli, and feature a“point-to-point physical similarity” with that stimulus (Pear, 2001, p. 377). Echoics are likely the earliest verbal response to be learned (e.g. an infant’s “Mama!” is shaped from diffuse babbling to a perfect similarity to his or her mother’s usage). Like tacts, echoics are shaped primarily through generalized conditioned reinforcement. For example, a new nursing home resident might emit the echoic “319,” when told “Your room number is 319” and be reinforced with “That’s correct!” Two points regarding echoics that are possibly important for the subject of this paper. First, echoics are useful in building tacts and mands (Pear, 2001), although it is unclear if this is possible in individuals with dementia. Henry and Horne (2000) have demonstrated that contingent reinforcement can strengthen echoic behavior in persons with severe dementia. Future research will be needed, however, to determine if rehabilitation efforts that involve strengthening echoic behavior in persons with dementia will provide the substrate for rebuilding more complex manding and tacting repertoires. Second, echoics may be mistaken for tacts (that is, mistaken for “understanding” or misinterpreted as indicating high likelihood of compliance). For example, a family member might be preparing to leave her (early Alzheimer’s) mother’s home just as mom is finishing a cooking task. The daughter in this situation might mand “Mom, when you are done, remember to turn off the stove!” to which Mom replies “Turn off the stove, right!” The daughter discovers the stovetop still hot the next morning. What has occurred here? We might hypothesize that “Mom’s” response was not a tact. In such a situation, a verbally intact individual would likely respond to the daughter’s mand with a chain of pr ivate tacts (“The stove is on,” “This sauce will be done after 2 more minutes of stirring,” and “I need to turn off the stove at that time.”) and essentially reinforce the mand with an autoclitic describing the likely strength of the appropriate response (see Skinner, 1957, p. 315) indicating “Yes, I will definitely turn off the stove.” The mother in this example seemed to respond appropriately, but in reality simply responded with an echoic (“turn off the stove”) and, almost reflexively, added a socially appropriate intraverbal (discussed below), the “right!” that ended her sentence. Thus, what the daughter believes to be mom’s “strong intention” to safely extinguish the stove is nothing more than the confluence of two well-trained verbal operants (repeating what we are told, and doing so politely) divorced from any connection to what should be easily tacted as the dangerous stimulus in the environment. The example above shows how the ability to engage in echoic behavior can actually cause communication problems between caregivers and patients, particularly with professional caregivers who have little to no shared history with the patient. These communication problems are at best frustrating and at worst dangerous. Unfortunately, confusing echoics with other verbal operants is particularly likely given that the ability to engage in echoic responses is intact late into the dementing process. In the case described above, Skinner’s analysis might be useful for enhancing the safety and independence of the patient. It would suggest, for example that the daughter could gain greater control over her mother’s behavior by amplifying her vocal mand. She could, for example, post on the stove a written mand: “MOM, TURN OFF THE STOVE,” perhaps even including a photo of herself looking concerned and pointing toward the relevant control. Or she could devise some way that a heated burner might more effectively control the mother’s tact (“Oh, my, the stove is on”) and/or nonverbal behavior (actually turning it off). Stoves often provide only a weak SD to occasion turning off a burner, thus the impaired user might be assisted by using a larger light or an alarm sound. Intraverbal 51 Intraverbal responses are evoked by prior verbal stimuli, much as tacts are evoked by environmental antecedents (Pear, 2001). Unlike echoic responses, intraverbals have no formal correspondence with the evoking stimulus. But like both echoics and textuals, intraverbals are maintained via generalized conditioned reinforcement. Thus, the stimulus and response may be either vocal or written, or any combination and the analysis remains the same. A simple example of what Skinner (1957, pp. 71-72) refers to as “trivial” intraverbals, are the “answers” evoked by common social “questions” such as the intraverbal response “I am fine,” in response to the vocal verbal stimulus “How are you today?” Persons with dementia typically maintain the ability to respond appropriately to these common social questions (it is often said that patients maintain “social graces”) in the early stages of the disease, making early detection a difficult task, even for those who know the patient well. However, intraverbal is not necessarily to be equated with trivial, and Skinner goes on to note many situations in which complex conversations or answers (e.g. “the facts of science” p. 72) are mainly intraverbally controlled. Lack of such control can even have diagnostic importance, such as when we ask a mental status exam question such as “Complete this sentence, ‘Right as _____’.” We might assess further if we receive the wrong answer from a patient whose response (“rain”) should have been well established intraverbally through years of training within our particular verbal community. Textual Textual (as in reading a text) responses are vocal responses controlled by non-auditory stimuli. These SDs may be visual or tactile (i.e. Braille) stimuli and may have a diversity of forms (e.g. words, pictures, symbols). But all simply set the occasion for a vocal response. Like echoic behavior, much of a texting repertoire is explicitly evoked and reinforced, with generalized conditioned reinforcers, in “educational” settings (school) or relationships (parent-child). Here the listener reinforces the speaker’s vocal responses if they have the correct relationship to the textual stimulus. Fortunately the ability to engage in textual responses remains intact early in the dementing process (Orange, 2001), thus behavior may be controlled by more complex stimuli such as lists, calendars, or notes produced by others or themselves. Therefore, analyzing textual control, and making practical use of textual stimuli in populations such as patients with dementia is relevant and useful. This is illustrated above with our “turn off the stove” example. Bourgeois and colleagues (1997) demonstrated that repetitive questions could be reduced in community-dwelling persons with dementia by having family members prompt patients to read cards that contained the answers to frequently asked questions. Even as the underlying disease progresses into the moderate stages, patients can respond effectively to simple, frequently-occurring words (Cummings, Houlihan, & Hill 1986) or pictures depicting objects or actions. The first author worked with a family in which the father with dementia left his dirty clothes on the bathroom floor. A large sign on the top of the hamper reading, “PUT CLOTHES HERE” resulted in near elimination of this behavior. Hussian (1988) showed how stimulus enhancement techniques such as making words or pictures larger, more colorful, or in prominent, noticeable places may be particularly helpful in reducing challenging behavior in persons with dementia that is due to insufficient stimulus control such as attempting to leave protective environments or inappropriate voiding. Whether textual (written by others) or self-textual (e.g. a reminder note to one’s self) the influence of written or pictorial SD s may long outlast similar echoic stimuli, and thus control behavior more effectively. And, as Skinner further notes, the massive reinforcement history for responses such as reading makes it likely that textual stimuli will attract attention. Autoclitic Autoclitic are verbal responses controlled by the speaker’s own previous verbal behavior. They allow the speaker to create longer pieces of verbal behavior that are “intelligible,” that is, that function to allow the listener to take “effective action” (Skinner, 1957, p. 314). Pear (2001) interestingly describes 52 autoclitics in a manner useful for the topic of the current paper. He casts them as verbal escape or avoidance responses that are reinforced by modifications in the listener’s response. An extension of Pear’s (2001, p. 377) pediatric example may illustrate. A geriatric patient might mand “Please give me my medication,” but then quickly add the autoclitic “the blue tablets,” to ensure that the listener does not dose him with the white tablets and blue capsules he has recently taken. The autoclitic phrase is here controlled by the deficient mand, and will be negative ly reinforced by avoidance of the overdose that the listener might otherwise have supplied. The Audience Relation and the Negative Audience As language deficits become evident, there will be at first an insidious effort to alter how one interacts with that person. Although some small adjustments might be helpful in facilitating communication (i.e., slowing down, using fewer words to communicate), many other adjustments (e.g., corrections and criticisms, or ignoring the individual) may actually serve to punish verbal behavior. These ‘tell-tale’ signs of change in communication patterns coincide with what Skinner termed the “Negative Audience” which is, “an audience in the presence of which verbal behavior is punished” (Skinner, 1957 p. 178). Changes in interactions with the verbal community alter patterns of reinforcement with very established histories. The impact of these contingencies on communication and their relative strength is established over a lifetime, and do not simply emerge as relevant when dementia begins to erode these established communication patterns. We start out our lives with our parents celebrating every word we utter. Later on, adolescence presents a rich tapestry of communication opportunities with the adolescent needing to effectively communicate with everyone from the very young (e.g., nieces and nephews) to the very old (e.g., grandparents) (Williams & Garrett, 2002). With age comes a narrowing of the channels of communication with a cohort that grows progressively smaller over time through attrition. Unfortunately, with the dwindling of an audience there also comes the potential for a concurrent decline in verbal behavior. Also, fewer opportunities for meaningful communication concomitantly increase the meaning of social opportunities and the reinforcement they provide (see Houlihan, Rodriguez, Levine, & Kloeckl, 1990). In fact, results of the Geriatric Reinforcer Survey (Houlihan et al.) show that what elderly residents of nursing homes find most rewarding is social contact and conversation with family and friends. Despite this need for socialization, many changes in the verbal community actually produce reductions in social interaction. Initially, the remaining audience is one often made up of family and friends that are most familiar to the dementia sufferer. These individuals may no longer reinforce verbalizations with smiles and head nods, but instead replace them with looks of concern, frustration or disinterest (Skinner, 1957). These audiences may effectively punish verbal behaviors in the sense of socially conveying a visible preference for silence. Because recognition is a memory system that relies on information already established in memory to match or compare, it generally outlasts recall which involves a search of memory for something that is often more recently established. This is to say that people are generally better with faces than names. Izard has shown convincingly that facial expressions are our earliest and most established forms of expression and communication (Izard & Ackerman, 1997). Smiles, nods, and continuation of conversations serve as setting events for positive social behaviors. Looks of disinterest or statements like “I told you that already,” or “I know you’re not that stupid,” serve to punish prosocial behaviors and negatively reinforce social withdrawal or dependence. The meaning of those expressions is amongst some of the last things lost to memory. People are sometimes even more callous or blunt with individuals with memory problems because they figure that they will just forget the negative statements anyway. Although those familiar with the patient may initially serve as a negative audience in the manner described above, the situation often continues as impairment worsens and the patient requires placement 53 in long-term care facilities (e.g., nursing homes, assisted living facilities). In long-term care, the patient is being cared for by individuals who share no history with the patient and who are paid little, have little training, and have many patients to care for in a short amount of time. This combination of factors can understandably lead to poor staff-patient communication. For example, recent studies by Williams and colleagues (2009; Cunningham & Williams, 2007) have shown quite clearly that a disrupted pattern of communication known as elderspeak (i.e., infantilizing speech similar to that used with small children) exists in staff-patient relationships in nursing homes and that elderspeak increases the probability of resistance to care (e.g., hitting, saying no, crying) by dementia patients. Kitwood (1990a) refers to these negative changes in patterns of communication as the “malignant social psychology.” The result can be social withdrawal, which is understandable in that many of these patients have lengthy histories of being immersed in an environment that generally provided rich schedules of reinforcement (both positive and negative) for their spoken language. As Skinner (1957) has noted, a rich schedule of reinforcement is also a schedule that is most easily extinguished (Neisworth, 1985). Over time, the graduated banishment, first psychologically and eventually physically, of the dementia sufferer from the verbal community may occur and the end can be marked by a nearly total deprivation of sustaining human contact. Overall, it appears that at a point in your life when you most crave and value human social contact, sources of social contact begin to diminish and the type and quality of communication changes. Even those fortunate enough to maintain their cognitive abilities into old age might find themselves on a schedule of extinction with few opportunities for discussion with others and their ideas often being devalued due to their age. The net result of diminished opportunities for social interaction, instances of elderspeak such as “let’s go potty,” combined with negative visible or verbal reactions to flawed statements, is to take our memory impaired loved ones away from us well before the disease does. Kitwood (1997) suggests that this malignant social psychology (i.e., negative audience) might “…even serve to accelerate the advance of neurological degeneration” (p. 51). Some empirical work has focused on interventions targeting specific repertoires of the verbal community with the goal of improving communication with patients, reducing problematic behavior (e.g., resistance to care) and improving mand compliance. These studies demonstrate how important the verbal community is in contributing to the functioning of dementia patients as well as the importance of specifying behaviors of the verbal community that either increase or decrease the likelihood of evoking effective behavior from persons with cognitive impairment. For example, Gentry and Fisher (2007), using an ABAC design, compared two different types of listener repair responses on the verbal behavior of three dementia patients. Repairs were either direct (i.e., the listener provided corrective feedback to the patient when an error is made) or indirect (i.e., the listener restated his/her understanding of what the patient said). Results indicated that indirect repairs were associated with more words spoken, longer speech duration, fewer topic changes, and fewer incomplete interactions compared to direct repairs. Recently studies have begun to investigate the relationship between patient mand compliance and the type of mands delivered by caregivers during personal care tasks (Buchanan, Christenson, & Houlihan, 2008). Preliminary results show that alpha commands (i.e., commands that are clear, concise, and feasible) produce significantly better mand compliance than do beta commands (i.e., commands that are vague, ambiguous, or do not give the individual an opportunity to comply). In addition, commands that involved one step, that were stated directly, that were repeated exactly, and that clarified previous commands were more effective in producing compliance. On the other hand commands in the form of questions, commands that involved more than one step, and commands using first-person plural pronouns (“we need to go to the bathroom”) produced greater rates of noncompliance. Practical Implications 54 An important practical implication of Skinner’s functional analysis is that it provides a framework for conceptualizing intervention targets (e.g., tacting or manding repertoires) that could potentially help maintain independence, preserve dignity, and reduce burden on caregivers. The 7 verbal operants provide valuable a system for categorizing communication problems and, as many of the examples described throughout this paper illustrate, can even be prescriptive in the sense of suggesting intervention strategies. As an additional example, consider a familiar scenario that occurs in long-term care facilities where an individual with dementia is physically or verbally aggressive during activities of daily living (ADLs) such as bathing. Commonly a patient will begin by protesting to getting wet or having clothes taken off with either vague statement (saying “no” “why”, or “stop”) or seemingly benign behaviors (e.g., heavy sighing, tightening muscles, or pulling away from caregivers). These behaviors often function as mands that essentially communicate something like, “Please back away from me because I’m scared, in pain, cold, embarrassed, etc…” For various reasons (e.g., ADLs must be completed, the patient’s mands may be imprecise or incomplete, the caregiver is attending to another caregiver and not the patient), these mands are often ignored by caregivers. Patients may respond by augmenting their mands with cursing, threatening, or hitting caregivers in order to escape or otherwise change the aversive situation. Caregivers may respond by altering their behavior in a variety of ways (e.g., provide a brief break, end the task, re-establish eye contact with the patient, talk directly to the patient as opposed to another caregiver, apologize, change the water temperature) that can serve to reinforce aggression. Ideally, a caregiver could reflect upon this situation and identify that a mand was overlooked early in the bathing process and consider other ways to approach bathing next time by, for example, proceeding more slowly, changing water temperature when asked, talking directly to the patient about familiar topics, or periodically giving the patient a break contingent upon more appropriate behavior. This approach is likely to be far more fruitful than simply labeling the patient as “aggressive” or “violent.” At best these labels result in caregivers becoming resigned to the “fact” that the patie nt is, and always will be, aggressive and aversive interactions continue. At worst these labels result in placing the patient on psychotropic medications that often are ineffective and have harmful side effects such as sedation or confusion. Future Research Implications Skinner’s analysis may also provide a useful guide for future research on communication difficulties in persons with dementia. Although the discussion above points to numerous areas for future empirical inquiry in our opinion, four particularly important targets for future research with dementia patients include: 1) Improving tacting. The inability to name objects or people is a prominent deficit early in the disease process and can be very distressing for patients and their families, particularly when the names of loved ones cannot be produced. 2) Improving manding. The ability to clearly express needs verbally is impaired throughout the disease process. Unsuccessful attempts to communicate needs can produce frustration and social withdrawal in both patients and their caregivers and leave the patient’s needs unmet. Eventually, behavioral disturbances such as loud, repetitive vocalizations, pacing or physical aggression may develop as an alternative means for expressing needs. 3) Improving mand compliance. Although mand compliance is a listener repertoire, and thus not technically verbal behavior, Henry and Horne (2000) appropriately note that one must learn appropriate listener behavior in order to function as a member of the verbal community. Also, because individuals with advanced dementia require assistance to complete personal care such as dressing or bathing, mand compliance is critical for the successful completion of these tasks. Developing ways to improve mand compliance may result in less stressful interactions during personal cares and fewer injuries to caregivers and patients. 4) Altering the behavior of the verbal community to better support the patient’s independence. This may include behaviors such as improving delivery of mands or altering communication styles so as to encourage verbal behavior instead of inadvertently punishing it. Summary 55 Perhaps the most valuable contributions of Skinner’s analysis for studying dementia are its focus on three major themes. First, Skinner conceptualizes any behavior (not just vocal behavior, but written, gestural, etc…) that exerts its effect through the actions of another person as being “verbal.” This point is worth emphasizing because many take “verbal behavior” to be equivalent with “speaking” when in fact verbal behavior can involve gesturing, unintelligible vocalizations, or writing. Conceptualizing verbal behavior as more than just speaking can help caregivers interpret a variety of different behaviors as being verbal in nature and serving a social function. This emphasis on function versus topography of behavior is the second important theme in Skinner’s analysis that is beneficial for studying dementia. Skinner’s focus on function and environmental causation encourages theorists to examine the social context in which verbal behavior occurs and the transactional influence of the patient’s verbal behavior on the verbal community, and the verbal community’s response on the patient’s verbal repertoire. In general, those studying dementia have placed less emphasis on understanding the social context in which verbal behavior occurs, and how the verbal community can contribute to maintaining existing repertoires or, conversely, exacerbate existing deficits and cause excess disability. Because conditions like Alzheimer’s disease cause progressive and irreversible impairment, changing the behavior of the patient’s verbal community may be more realistic than teaching the patient new repertoires or reestablishing lost ones (Gentry & Fisher, 2007). Furthermore, as many of the examples described previously illustrate, the focus on function may also help one better understand that seemingly disruptive behaviors (e.g., loud moaning, calling out, or repeating questions) can serve important social communicative functions and should not necessarily be eliminated through the use of psychotropic medications. The situation above concerning physical and verbal aggression during ADLs provides a good example of this point. Simply eliminating behavior through the use of medication could be considered unethical in that medications exacerbate the loss of verbal behavior in individuals who are already losing it as a consequence of their disease. In addition, an important social need would potentially be left unmet. Third, Skinner’s analysis insists on fully accounting for the environmental contingencies that affect verbal behavior versus ascribing such behavior to a ghostly (Baum, 2005) inner agent. This contrasts with how impairments in verbal behavior related to dementia have been traditionally conceptualized. Research in the communication disorders literature has thoroughly described specific topographies of verbal deficits associated with different stages of the disease process (see Kempler, 1991 for an excellent example), but has been less fruitful in terms of understanding their causes and in designing interventions for deficits in verbal behavior. Furthermore, because Alzheimer’s disease and other dementias have biological correlates (e.g., brain atrophy, declines in neurotransmitters, toxic accumulations of twisted proteins such as beta amyloid), language disturbance is often attributed primarily to biological effects of the disease. Attributing deterioration solely to biological processes has stifled theoretical and empirical inquiry into environmentally-based causes and interventions, although some notable exceptions exist (e.g., Kitwood, 1997). Skinner’s formulation could potentially move theorists away from exclusive reliance on internal causes of behavior. It places the analysis in the environment, which is currently much more manipulable given that medications for dementia (e.g., cholinesterase inhibitors) have only modest benefits (Whitehouse, 2008, pp. 117-119). Skinner’s Verbal Behavior is replete with ideas that can be applied to the study of dementia and only a portion of Skinner’s overall analysis was addressed in the current paper. As Giles (1999) notes, there is an inherent need to better understand through research the injustices put on the elderly as they struggle to communicate effectively. It is hoped that the examples and empirical work described throughout this paper will inspire the reader to explore this literature further and lead to the creation of ideas for empirical work in this under-developed, yet critically important area. References 56 Abrahams, J.P. & Camp, C.J. (1993). Maintenance and generalization of object name training in anomia associated with degenerative dementia. Clinical Gerontologist, 12, 57-72. American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders, text revised (4 th ed.). Washington, DC: Author. Baum, W. M. (2005). Understanding Behaviorism: Behavior, Culture and Evolution. Malden, MA: Blackwell Publishing. Bourgeois, M. S. (1992). Evaluating memory wallets in conversations with persons with dementia. Journal of Speech & Hearing Research, 35(6), 1344-1357. Bourgeois, M. S. (1993). Effects of memory aids on the dyadic conversations of individuals with dementia. Journal of Applied Behavior Analysis, 26(1), 77-87. Bourgeois, M.S., Burgio, L.D., Schulz, R., Beach, S., & Palmer, B. (1997). Modifying repetitive verbalizations of community-dwelling patients with AD. The Gerontologist, 37, 30-39. Buchanan, J.A., Christenson, A., & Houlihan, D. (2008, October). An analysis of command types and compliance during personal cares between nursing home staff and residents with cognitive impairment. Paper presented at the meeting of the Mid-American Association of Behavior Analysis, Champaign, IL. Camp, C.J. (1998). Memory interventions for normal and pathological older adults. In R. Schulz, G. Maddox, & M.P. Lawton (Eds.), Annual Review of Gerontology and Geriatric, volume 18. New York: Springer Publishing Company. Camp, C.J. & Schaller, J.R. (1989). Epilogue: Spaced-retrieval memory training in an adult day-care center. Educational Gerontology, 15, 641-648. Cummings, J.L., Houlihan, J.P., & Hill, M.A. (1986). The pattern of reading deterioration in dementia of the Alzheimer’s type: Observations and implications. Brain and Language, 29, 315-323. Cunningham, J. & Williams, K.N. (2007). A case study of resistiveness to care and elderspeak. Research and Theory for Nursing Practice: An International Journal, 21, 45-56. Gentry, R.A., & Fisher, J.E. (2007). Facilitating conversation in elderly persons with Alzheimer’s Disease. Clinical Gerontologist, 31, 77-97. Giles, H. (1999). Managing dilemmas in the ‘silent revolution’: A call to arms! [1999 ICA presidential address]. Journal of Communication, 49, 170-182. Hebert, L.E., Scherr, P.A., Bienias, J.L., Bennett, D.A., & Evans, D.A. (2003). Alzheimer’s disease in the U.S. population: Prevalence estimates using the 2000 Census. Archives of Neurology, 60, 1119-1122. Hendrie, H.C. (1997). Epidemiology of Alzheimer’s disease. Geriatrics, 52, S4-S8. Henry, L.M. & Horne, P.J. (2000). Partial remediation of speaker and listener behaviors in people with severe dementia. Journal of Applied Behavior Analysis, 33, 631-634. Hoerster, L., Hickey, E. M., & Bourgeois, M. S. (2001). Effects of memory aids on conversations between nursing home residents with dementia and nursing assistants. Neuropsychological Rehabilitation, 11, 399-427. Houlihan, D., Rodriguez, R., Levine, H., & Kloeckl, J. (1990). Validation of a reinforcer survey for use with geriatric patients. Behavioral Residential Treatment, 5, 129-136. Hussian, R.A. (1988). Modification of behaviors in dementia via stimulus manipulation. Clinical Gerontologist, 8, 37-43. Izard, C.E., & Ackerman, B.P. (1997). Emotion and self concepts across the life span. In K.W. Schaie & M.P. Lawton (Eds.), Annual review of gerontology and geriatrics, 17, (pp. 1-26). New York: Springer. Kempler, D. (1991). Language changes in dementia of the Alzheimer’s type. In R. Lubinski (Ed.). Dementia and communication (pp. 98-114.). Philadelphia, PA: B.C. Decker, Inc. 57 Kitwood, T. (1990a). The dialectics of dementia: With particular reference to Alzheimer’s disease. Ageing and Society, 10, 177-196. Kitwood, T. (1997). Dementia reconsidered: The person comes first. Buckingham, UK: Open University Press. Levine, R. (2006). Defying dementia: Understanding and preventing Alzheimer’s and related disorders. Westport, CT: Praeger. Neisworth, J. T., Hunt, F.M., Gallup, H.R., & Madle, R.A. (1985). Reinforcer displacement: A preliminary study of the clinical application of the CRF/EXT effect. Behavior Modification, 9, 103-115. Orange, J.B. (1991). Perspectives of family members regarding communication changes. In R. Lubinski (Ed.). Dementia and communication (pp. 168-186.). Philadelphia, PA: B.C. Decker, Inc. Orange, J.B. (2001). Family caregivers, communication, and Alzheimer’s disease. In M.L. Hummert & J.F. Nussbaum (Eds.). Aging, Communication, and Health: Linking Research and Practice for Successful Aging (pp. 225-248). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Pear, J. J. (2001). The Science of Learning. Ann Arbor, MI: Edwards Brothers. Skinner, B.F. (1957). Verbal Behavior. Acton, MA: Copley Publishing Group. Whitehouse, P.J. (2008). The myth of Alzheimer’s: What you aren’t being told about today’s most dreaded diagnosis. New York, NY: St. Martin’s Griffin. Williams, A., & Garrett, P. (2002).Communication evaluations across the lifespan: From adolescent storm and stress to elder aches and pains. Journal of Language and Social Psychology, 21, 101-126. Williams, K.N., Herman, R., Gajewski, B., & Wilson, K. (2009). Elderspeak communication: Impact on dementia care. American Journal of Alzheimer’s Disease & Other Dementias, 24, 11-20. Author Contact Information Jeffrey Buchanan, Ph.D. Minnesota State University, Mankato 23 Armstrong Hall Mankato, MN 56001 Phone: 507-389-5824 e-mail: [email protected] Daniel Houlihan, Ph.D. Minnesota State University, Mankato 23 Armstrong Hall Mankato, MN 56001 Phone: 507-389-6308 e-mail: [email protected] Peter Linnerooth, Ph.D. Minnesota State University, Mankato 23 Armstrong Hall Mankato, MN 56001 Phone: 507-389-6217 e-mail: [email protected] 58 Using Skinner’s Model of Verbal Behavior Analysis to study Aggression in Psychiatric Hospitals Michael Daffernand Matthew Tonkin Abstract An adaptation of Skinner’s verbal behaviour model is proposed as a framework for analyzing the aggressive behaviour of hospitalized psychiatric patients. Few behavio ur analytic studies have addressed verbal aggression and none draw upon Skinner’s model. These few studies suggest that the aggression of psychiatric patients is multiply determined and multi-functional. In this context, aggression occurs primarily to express anger. It is preceded proximally by aversive interpersonal interactions and often protects status. More recent approaches to the asses sment of function are broader and more instructive for observers of aggressive behavior than Skinner’s original conceptualisation. These approaches eliminate those operants (textual, transcriptive) that lack relevance to the study of psychiatric inpatient aggression and they differentiate the functions of other operants (mands, tacts ) to reflect the functions of aggression demonstrated by these patients . Keywords: Verbal behavior, aggression, psychiatry, hospital Introduction Skinner’s (1957) conceptualization of verbal behavior has its roots in a series of lectures first presented at the University of Minnesota in the late 1930s, and subsequently at Harvard University, and the University of Chicago. In these lectures Skinner proposed a new theory of verbal behavior, which differed from existing explanations of verbal behaviour that tended to account for language in terms of its underlying meaning. For Skinner, existing explanations were scientifically inadequate because they failed to identify measureable and controllable variables that were causally linked to observable behavior (Skinner, 1957). Skinner, therefore, rejected the traditional formulation of verbal behavior in favor of a new, behaviorally derived formulation. Skinner’s new formulation accounted for verbal behavior in terms of operant conditioning. He suggested that verbal behavior (1957) could be understood (and therefore controlled) by gaining an appreciation of the antecedents and consequences of specific utterances, which was best done using functional analysis. So, like other operant behavior , verbal behavior (1957) was conce ptualised as a function of the speaker's c urrent environment and their behavioral history. It is these factors that Skinner argued were central to the scientific investigation and understanding of verbal behavior (1957). Skinner’s functional analysis identified four antecedents (stimuli) and two consequences (reinforcers) that function to control verbal behavior. The four antecedent variables were: (1) some state of deprivation or aversive stimulation, (2) some aspect of the envir onment, (3) other verbal behavior, and (4) one’s own verbal behavior. The two consequences of verbal behavior were: (1) something related to the state of deprivation/aversive stimulation, and (2) social/educational consequences (Frost & Bondy, 2006). These six variables combine to form six basic functional (stimulus-response) relationships, which specify the antecedent conditions and subsequent consequences of verbal behavior. These six basic units or verbal operants are the building blocks of verbal behavior, and can either stand alone as simple forms of language or can combine to create more advanced forms of verbal behavior (Sundberg & Michael, 2001). The first verbal operant explained by Skinner (1957) was that of the mand. The term mand was used to refer to verbal behaviors that occur in response to a state of deprivation/aversive stimulation and which tends to produce some beneficial or desired outcome related to that state . In other words a mand is a verbal operant that specifies its reinforcer. A typical example is a request for water by one who is thirsty. 59 Another verbal operant identified by Skinner (1957) is the tact. Tacts are verbal behavior evoked by an object/event or some property of an object/event (Skinner, 1957). That is, the presence of a given object (the stimulus) increases the probability of a given response (the tact), which in turn is generally reinforced (e.g. praise). For example, presenting a picture of a cat to a child may evoke the response “cat”, which is reinforced by praise from the parent. Additional verbal operants are the echoic, intraverbal, textual, and transcriptive. These are similar to tacts in that they are made as a response to some external stimulus. The difference, however, is that intraverbal, echoic, textual and transcriptive operants occur in response to a verbal discriminative stimulus (S D), whereas the tact is made in response to a non-verbal stimulus. Echoic verbal operants refer to verbal behaviors that are simple imitations of the verbal behavior of others. Intraverbal operants also occur in response to an S D but are not echoic in nature. Common examples of intraverbal behaviors are answering a question or filling in a blank, such as when a child says “farm” after hearing “Old MacDonald had a…” (Frost & Bondy, 2006). Conversational responses are also common forms of intraverbal operants. Textual verbal behavior is controlled by the written word (a form of non-auditory verbal stimulus). For example, reading aloud the word “Cat” that is written on a page. Here the written word is the stimulus, saying the word aloud is the response, and typically the reinforcement is social approval (Skinner, 1957). Transcription can either be in the form of “taking dictation,” where the stimulus is auditory and the response written, or it can be in the form of “copying text” where both the stimulus and the response are in written form. Transcriptive verbal operants receive ‘many special educational and economic reinforcements’ (Skinner, 1957, p. 71). More detailed analyses of Skinner’s conceptualisation are presented by Frost and Bondy (2006), and by Sundberg and Michael (2001) Existing applications of Skinner’s Verbal behavio r Although Skinner’s (1957) book, Verbal behavior , was essentially theoretical, in that he did not present any experimental data, it was practice-oriented. Despite this, it took over 40 years before Skinner’s terminology and theory were applied in a practical manner (Sundberg & Michael, 2001). However, several applications were reported prior to this, including Zoellner’s (1969) method for teaching English composition to College students, Skinner’s (1981) application to professional writing, and Sloane, Endo, and Della -Piana’s (1980) suggestions for facilitating creativity. More relevant to the current article, perhaps, was the application of Skinner’s tacts, mands and intraverbal responses to the study of verbal communication among a group of emotionally-disturbed adolescents (Salzinger, 1958). This study developed a framework for categorising the verbal behavior used in letters to friends and relatives. Skinner’s theory was useful in this context because it provided a structured system for breaking down verbal behavior into its constituent parts, which enabled the researchers to conduct a more fine -grained analysis and comparison with ‘normal’ individuals than would otherwise have been possible. Also relevant are Glenn’s (1983) analysis of client maladaptive behavior in clinical situations (e.g. lying, denial, demanding) and Burns, Heiby, and Tharp’s (1983) analysis of auditory hallucinations, both of which adopted a verbal behavior perspective. Although interesting , these applications of Verbal behavior (1957) did not give rise to any consistent, formal programs that could be applied to the treatment or assessment of clinical problems. Indeed, it was not until the last decade or so that a body of research by applied behavior analysts (see below) has enabled Skinner’s theory to be applied to the treatment of individuals with communication deficits. Much of this research has focused on developing language acquisition programs for children with autis tic spectrum disorders, such as the aptly named ‘Verbal Behavior’ program (Sundberg, Michael, Partington & Sundberg, 1995). The most prominent deficit in autistic children tends to be their use of language to interact effectively with others (Paul & Sutherland, 2005). The Verbal Behavior program, therefore, adopts two main aims: (1) To improve the maturity of language forms; and (2) To enable a more effective use of communication. In order to achieve these aims the program provides a carefully sequenced curriculum for teaching language, which works through a series of increasingly more complex language goals. 60 The five language goals of the program (ordered in terms of complexity) are: Echoes (to facilitate the use of imitation); Mands (to facilitate the use of verbal behaviors that produce an immediate effect on the environment, such as having a request reinforced); Tacts (to facilitate the use of verbal labels); Reception by feature, function and class (RFFC, to improve the use of verbal behaviors made in response to common verbal stimuli, such as written words); and Intraverbals (to improve the use of non-echoic verbal responses that occur in response to the verbal behavior of others, such as conversation). The program begins by teaching the child echoic and mand verbal behavior. As the child masters these more elementary forms of communication s/he is taught the more complex forms of verbal behavior. The program can, therefore, be described as a tiered approach to language acquisition. However, in addition to the language goals being presented in a tiered manner, within each language goal the child is taught in a tiered fashion. For example, echoic responses are initially taught by having the child imitate simple speech sounds (such as “ma” and “da”). Once these have been mastered the child is taught to imitate entire words, then entire phrases and so on, with the hope that the child’s programmed behavior will generalize to imitating novel words and phrases. The Verbal Behavior program has received empirical support, with graduates demonstrating increases in verbal production (e.g. Partington, Sundberg, Newhouse & Spengler-Schelley, 1994; Sundberg & Michael, 2001; Sundberg et al., 1995). It has also been adapted to facilitate the acquisition of nonverbal forms of communication (e.g., see Carbone, 2003). The practical applications of Skinner’s theory have, therefore, been quite limited in scope and number. This is despite Skinner’s original emphasis on the applied nature of Verbal behavior (1957) and despite scholars highlighting the significant applied and theoretical potential of the book (see Sundberg, 1991) (e.g., to the analysis of bizarre verbal behaviors, complex personal and family problems, and for changing the verbal behavior of juvenile delinquents). Interestingly, many of the potential applications suggested by Sundberg are relevant to the issues and behaviors that manifest within psychiatric hospital settings. The current application of Verbal behavior to aggression in psychiatric units, therefore, seems timely and appropriate. We begin this application by considering the scope and impact of aggression in psychiatric hospital settings. Aggression during psychiatric hospitalization Aggression during psychiatric hospitalization has long been recognized as a commonly occurring and significant problem (Fottrell, 1980; James, Fineberg, Shah, & Priest, 1990; Yesavage, 1983). That aggression is widespread in many psychiatric hospitals is unsurprising. Necessary criteria for admission in a mental health service as an involuntary patient is to prevent injury to both the public and the admitted person. Aggression has a significant impact on patients and staff, ward routine , and mental health services in general (Daffern & Howells, 2002). Aggression may affect patients’ treatment and access to rehabilitation programs. It can also influence the level of supervision and may result in the patient’s isolation from others or prolong hospitalization. Injury to patients engaging in aggression, and injury to co-patients and staff who are the victims of aggression, are also common consequences of inpatient aggression. Aggression can significantly compromise ward atmosphere, morale, and functioning (Monroe, Van Rybroek, & Maier, 1988). Organizational problems related to aggression include time lost from sick leave taken by staff in response to aggression, problems with staff recruitment and retention to hospitals where aggression is common, financial costs associated with compensation for injury, as well as official inquiries and litigation (Hillbrand, Foster, & Spitz, 1996). Of the considerable body of research examining aggressive behaviors in psychiatric hospitals, most studies have drawn upon structural rather than functional assessment approaches. Structural assessment approaches emphasize the correct classification of the form of a particular behavior, whereas functional assessment approaches emphasize the purpose of the behavior (Owens & Ashcroft, 1982). Furthermore, studies of aggressive behavior have typically focussed on the demographic and clinical characteristics of aggressive patients. Symptoms of psychosis, in particular delusions and hallucinations, have been a primary focus. Less emphasis has been placed on the 61 interpersonal context (i.e. the interaction between individuals on the unit) and environment in which aggression occurs (for exception see Gadon, Cooke & Johnstone, 2006,) though recently the interactional nature of aggression has been emphasized (Daffern, Duggan, Huband & Thomas, 2008; Whittington & Richter, 2005). Results of various studies suggest a complex array of interrelated factors contributes to aggression. Certain symptoms of mental disorder, specifically active symptoms of psychotic illness such as command auditory hallucinations and persecutory delusions , increase the likelihood of aggression during hospitalization for some patients. Disorder of thought, increased physiological arousal, disorganized behavior, and substance use may all contribute, although to a somewhat lesser extent (Daffern & Howells, 2002). Personality factors, including disorders such as psychopathy (Heilbrun et al., 1998) and interpersonal style (Daffern et al., 2008) also relate to aggression during hospitalization. The physical characteristics of the hospital, the rules and regulations by which it operates, and the behavior of ward staff and other patients are also important (Daffern & Howells, 2002). The functionality of behavior emitted by patients of psychiatric hospitals Published examples of behavior analytic assessment of aggression within psychiatric hospitals are scarce. This may be because traditionally, the behavior of mentally ill patients, particularly those in the acute phase of psychotic illness, was considered purposeless or driven exclusively by symptoms of psychiatric illness. Thus, they were viewed as unrelated to environmental contingencies. Furthermore, aggression generally is considered pathological because it breeches social mores and may have profoundly negative consequences for the perpetrator (Layng & Andronis, 1984). It is however, like other bizarre behavior, still generally adaptive when considered within the context of the aggressive patient’s limitations, tendencies, skills, or pre-existing vulnerabilities (Goldiamond, 1975a & b) and environmental contingencies (Daffern, Howells, & Ogloff, 2006). Layng and Andronis (1984) assumed that behavior, in their case delusional speech and hallucinatory behavior, is operant, in that its frequency is a function of contingent consequences. As an example, they report on the case of a psychiatric patient behavin g erratically and speaking in a delusional manner. The woman complained about how hard it was for her to go to the nurse’s station to talk to staff and that when she did her requests for help often incurred hostile responses. “Her delusion was an immediately less onerous, but ultimately very costly, alternative to the more difficult task of going to the nurse’s station to seek out unit staff. Both patterns, it was noted, appeared to produce the same maintaining consequences, i.e., conversations with staff” (p. 142). According to Layng and Andronis (1984), when the consequential alternatives that were available to the woman were considered, her so-called delusional behavior was comprehensible. In Skinner’s terminology , this patient’s delusional speech may be considered a dysfunctional mand in that the dysfunctional speech occurred in response to a state of deprivation/aversive stimulation (perhaps loneliness) and resulted in the desired outcome (i.e., talking with staff). Further demonstrations of the role of social and environmental contingencies on psychiatric symptoms are evident in Alford and Turner’s (1976) introduction of social reinforcement, and Belcher’s (1988) use of aversive contingencies to reduce hallucinatory behavior. In one of few function analytic studies of aggression within psychiatric hospitals, Shepherd and Lavender (1999) studied the antecedents and aggression management strategies of 130 incidents of aggression: dividing antecedents into external and internal factors. In this study the majority of incidents (60%) were preceded by external factors such as staff refusal of a patient’s request or staff demand for activity, hospital related matters such as ward restrictions and transfers between wards, or patient-on-patient conflict. Forty percent of all incidents were attributed to internal factors such as the patient’s mental state or substance use. The function of these incidents was not recorded by Shepherd and Lavender (1994). However, they noted that aggression frequently occurred consequent to social distance (indicating attention seeking was a primary function) or as a consequence to aversive stimulation, which may suggest aggression was reactive or anger-mediated rather than predatory. 62 Daffern et al. (2006) have developed a system for classifying and recording the functions of psychiatric inpatient aggression. This method, the Assessment and Classification of Function (ACF), was informed by the cognitive model of anger developed by Raymond Novaco (Novaco, 1976; 1994), the instigating mechanisms outlined by Albert Bandura in his social learning theory of aggression (Bandura, 1973), and the interactions antecedent to aggression identified in previous research on aggression in psychiatric inpatient wards (Shepherd & Lavender, 1999). The following functions are included in the ACF: 1. Demand avoidance: Demands by co-patients or staff to cease an activity or to complete a task (e.g. to adhere to hospital routine, to cease a behavior, or to attend to an activity such as taking a shower) often precede aggression. This function is evident when the patient does not wish to attend to the demand and acts aggressively to avoid or escape from it. 2. To force compliance: Aggression frequently occurs following the denial of a request (e.g., to make a telephone call, attend a program, leave the unit, obtain medication, or receive information about their treatment). This function is evident when the patient is refused a request and then behaves aggressively in a demand for compliance. 3. To express anger: Aggressive behavior usually follows an event that the patient perceives as provocative. Types of provocation include perceptions of disrespectful treatment; unfairness/injustice; frustration/interruption; annoying traits, and irritations. Provocation may also include physical assault or threat by others. This function is evident when the patient is obviously angry and their aggression appears to be an expression of their anger. 4. To reduce tension (catharsis): Aggression tends to reduce general physical arousal. Some people may deliberately behave aggressively to reduce tension. For others, aggression may reduce tension even if this was not their intention. This function is evident when the patient’s aggression reduces tension. 5. To obtain tangibles: Aggression may be used to obtain tangible items such as cigarettes or money. This function is evident when the patient seeks to or obtains tangible items as a result of their aggression. 6. Social distance reduction (attention seeking): Aggressive patients tend to seek high levels of supervision and/or compete more frequently for staff attention. Some patients behave aggressively to reduce social distance, a function that may colloquially be referred to as attention seeking. This function is evident when the patient’s drive for behaving aggressively is to obtain additional attention. 7. To enhance status or social approval: Humiliating affronts and threats to reputation often precipitate aggression, particularly in settings where dominance and privilege is afforded to those who use aggression. This function is evident when the patient uses aggression to enhance status or prevent deterioration in status. 8. Compliance with instruction : Aggression may occur following a command auditory hallucination or following instruction by another person. Compliance may alleviate distress or create alliances with others. This function is evident when the patient responds to a command/instruction to behave aggressively. 9. To observe suffering: Some patients may be motivated to act aggressively by the observation of suffering in their victim. This function is evident when the patient appears to find their aggressive behavior satisfying and where there is no obvious provocation. Most of the aforementioned purposes suggest verbal aggression is a consequence of inappropriate mands. As an example, for demand avoidance , the antecedent is a state of aversive stimulation (a demand made of the patient for activity such as taking medication), t he consequence is a reduction in the state of aversive stimulation (i.e. the demand goes away). At the same time, the aversive state may dissipate (To reduce tension) and other patients may praise the patient (To enhance status or social approval). Acts of aggression are therefore likely to be multifunctional. Verbal aggression may, therefore, arise because the patient has an inappropriate repertoire of mands ; their repertoire of mands is predominantly aggressive and antisocial rather than prosocial and calm, 63 but nevertheless functional for the individual . This problematic behavioral repertoire exists because the patient’s learning history has supported the acquisition and maintenance of these problematic mands which discouraged the development and expression of prosocial mands . Similarly, the ward environment may support aggressive mands, with such verbal behavior affording the speaker enhanced status and social approval. Another verbal operant, the tact, is evident in some of the functions that relate to aversive stimulation aroused by aspects of routine/hospitalization. For example, being forced to congregate with other patients during mealtimes or being forbidden to leave the unit may lead to a state of irritation. Regarding the other verbal operants, it is possible to argue that in an environment where verbal aggression is common, that such behavior is simply echoic. However, it is rare that aggression within psychiatric units is a simple imitation isolated from other aversive antecedent states. As such verbally aggressive behavior may be better construed as mands. Similarly, although some aggression may occur in response to a question or demand (e.g., a demand from a staff to return to their room) and could be considered as a dysfunctional intraverbal behavior, the aversive stimulation and anticipated and real consequences of aggression suggest it is better to regard these types of aggressive responses as mands. Glenn (1983) notes how rumination may be considered a form of intraverbal behavior. Originally, Skinner used the term intraverbal to refer to verbal behavior between two individuals. According to Glenn (1983), intraverbal behavior can also refer to inner speech (one’s own monologue), which may lead to the rehearsal of aggressive scripts, which often precede aggressive behavior (Bushman & Anderson, 2001). There is little similarity in the functions of aggressive behavior incorporated in the ACF with textual verbal behavior or transcription. Using the ACF, Daffern, Howells and Ogloff (2006) studied the functions of 502 acts of aggression perpetrated by patients of a secure psychiatric hospital in Melbourne, Australia. Their findings showed that most acts of aggression were functional, precipitated by identifiable events, and not simply the result of a spontaneous manifestation of underlying psychopathology. A number of dynamic interpersonal and contextual factors that contribute to aggression were identified. These included staff–patient interactions associated with treatment or maintenance of ward regime (Demand avoidance and To force compliance) that were considered frustrating and/or provocative (To express anger), or that threatened the status of the patient (To enhance status or social approval). Subsequent research has shown that the functions for aggression within hospitals may differ from the functions of aggression in the community and that the functions for aggression in personality disordered patients may be somewhat different from the functions of aggression for patients with mental illness (Daffern & Howells, in press). The extent to which this is true means that distinctly different functional relationships may guide the production of aggressive behavior (verbal or otherwise) in patients with personality disorder versus mental illness. Verbal aggression in psychiatric hospitals Like delusional speech and hallucinatory behavior, verbal aggression may result in considerable benefits (e.g., avoiding a demand) and/or costs (e.g., seclusion, restraint, restriction in privileges, physical injury, isolation and rejection) to the individual. Layng and Andronis (1984) suggest a cost/benefit type analysis can be extended to patterns of behavior not only considered irrational, but also to those behaviors whose costs are dramatic and immediate (like aggression), and that may obscure a clinician’s view of any possible benefits. Such an approach seems warranted for patients whose speech may be incomprehensible or for patients with limited vocabulary who may use topographically similar verbal behaviors (i.e., they may utter the same expletives) for dissimilar reasons. Personal histories of reinforcement and punishment in an environment likely influence the form of observed aggressive behavior.For instance, in hospitals where there is a history of particular aggressive acts (e.g. sexual assaults or assaults with particular weapons) then threats to sexually assault or attack staff with these weapons may predominate because they are effective, generating more fear and distress in victims than random utterances which may have no history within the 64 institution. Where verbal aggression occurs, the task for staff is to identify the determinants of the aggressive behavior and to attempt to modify these. According to Skinner (1993) “h ow a person speaks depends on the practices of the verbal community of which he is a member” (p. 99). Analysis of verbally aggressive behavior therefore demands investigation into (1) the characteristics of the environment that are conducive to aggression , and (2) the limitations existing within the individual that resulted in their needs being expressed in a problematic manner. For instance, an individual who needs analysis of low severity sexual aggression (e.g. threats to sexually aggress and lewd suggestions) may indicate that (1) sexual needs exist which are unable to be satisfied in more adaptive and prosocial ways, (2) that the individual has a sexual preference for aggression, or (3) that the aggression is more to do with expression of anger, and that this particular threat satisfies the drive for aggression without exposing the individual to aversive consequences (Daffern et al., 2008). At the contextual level, frequent sexual aggression, expressed verbally, may suggest consequences to such acts are negligible, or that other controlling factors are absent (e.g., male staff). Similarly, interpersonal consequences may be positive; lewd comments may be reinforced by co-patients, particularly when patients have an antagonistic relationship with staff. In one of few studies of the functions of verbally aggressive behavior in psychiatric hospitals, Daffern (2004) revealed the multifunctional nature of verbal aggression. The most common functions of verbal aggression in this study were to express anger (79.5%), reduce tension (61.1%), force compliance (44.4%), avoid a demand (36.8%), enhance status or social approval (33.3%), obtain tangibles (14.6%), narrow social distanc e (attract attention) (8.1%), cause suffering (2.4%), and to comply with instruction (1.7%). In Skinner’s terminology, then, the most common antecedent of verbal aggression is a state of aversion/deprivation (to express anger, reduce tension, force compliance, and avoid a demand). The functions of verbal aggression identified by Daffern were similar to those of physical aggression (Daffern et al., 2006). However, when compared to physical aggression, verbal aggression was more common when the purpose was to avoid a demand or to force compliance following the denial of a request. When the purpose of aggression was to reduce tension verbal aggression was also more likely. When patients were complying with instruction the aggression was more likely to be physical. Results also showed that patients were more likely to be verbally aggressive towards staff and physically aggressive towards other patients. It is likely that staff are protected from more severe physical aggression because of (1) anticipated consequences to assaulting staff (e.g., restraint, seclusion, reduced privilege s, prolonged hospitalization), which patients believe are less certain tha n when they assault patients, or (2) because patients accept that the demands and limits imposed by staff are routine aspects of care. The provocation although frustrating may not generate fear and excessive anger; physical aggression may not be justified. Verbal aggression as a warning One question often asked is whether verbal aggression, particularly threats, is a prelude to physical aggression, or whether the threat is adequate or satisfactory to obtain the desired outcome. Recent research (Ogloff & Daffern, 2006) on the prediction of imminent aggression suggests verbal aggression is a common antecedent to physical aggression. Two instruments used to assist psychiatric nurses to appraise risk for imminent physical aggression , the Broset Violence Checklist (BVC) (Almvik, Woods & Rasmussen, 2000) and the Dynamic Appraisal of Situational Aggression (DASA ) (Ogloff & Daffern, 2006); both incorporate a measure of verbal aggression. It is less clear whether verbal aggression directed towards a particular individual is a prelude to an attack against that person, or whether it is indicative of a generalized increase in risk to others. Threats, specifically threats to kill, are a particular form of aggressive behavior, which in many jurisdictions such as Victoria, Australia, constitutes criminal behavior carrying a prison sentence. Some (e.g. Calhoun, 1998) have suggested threateners may be differentiated into those who ‘howl’ and those who ‘hunt.’ In this typology, some individuals who threaten presumably intend violence (the hunters) and do not find the harm caused by the actual threat to be satisfactory, while others merely want to threaten to draw attention to themselves (howlers) and find this verbal aggression satisfactory. Those who ‘howl’ may well simply be verbally aggressive to avoid an aversive state (such as anger), they are just ‘sounding off,” whereas those who go on to physically 65 aggress (‘hunters’) might be using their verbal aggression to communicate some sort of request (mand). When this request is not understood or not met, they may then become physically aggressive. Glenn (1983) makes the point that inappropriate tacting in clients can often lead to feelings, needs etc. not being communicated effectively. The same point applies to aggression; an inability to label the behaviors of oneself and others (inappropriate tacting) makes it very difficult for someone to communicate in a clear way. So if patients can’t communicate their problems and desires clearly they are likely to feel frustrated and may then escalate from verbal to physical aggression. According to Warren, Mullen, Thomas, Ogloff, and Burgess (2007) the evidence that may be drawn upon to assist discrimination of those who eventually act from those whose who do not, is limited. In a data linkage study Warren et al. (2007) examined subsequent criminal convictions in 613 individuals convicted of threats to kill. Within 10 years, 44% of their cohort had been convicted of further violent offending. The original threat victim was subsequently a victim in 85 (13.9%) instances. Five of the original victims were eventually killed by the threatener, and for three others the threatener was later convicted of attempting to murder them. Subjects also reoffended against the threat victim by assaults (n=50), rapes (n=3), stalking (n=11), and further death threats (n=10). Conclusion Verbal aggression by psychiatric patients is multiply determined and multifunctional. Most commonly, verbal aggression , particularly towards staff, occurs consequent to demands for activity, or when requests are denied. As patients are typically angry when verbally aggressive, it is likely that these acts are not predatory. Psychiatric illness impairs the ability of many patients to manage the anger that is aroused by the demands and expectations of involuntary treatment. Forums allowing prosocial expression of discontent, access to staff who can support and encourage appropriate conflict resolution, and adoption of limit setting styles by staff that do not activate anger (Lancee, Gallop, & McCaye-Toner, 1995) are critical. Effective treatments of characteristics within patients that contribute to their inability to manage the demands of the psychiatric hospital without recourse to aggression are also required. This includes attempts to modify established contingencies assoc iated with interpersonal aggression. Psychiatric hospitals should carefully monitor aggressive behavior and compare their experiences with other hospitals. Only when patterns of aggression are compared can the determinants of aggression be elucidated and effective controlling variables introduced. Finally, although several verbal operants identified by Skinner (1957) seem relevant to verbal aggression in psychiatric hospitals there are several operants that are not immediately relevant. Further, the most relevant operants, mands and tacts, may be inadequately sensitive to the functions of aggression, when compared with more novel classifications systems such as that developed by Daffern and colleagues (2006). However Skinner’s work on verbal behavior (1957) may have implications for other components of patient behavior and response to treatment within psychiatric hospitals, for example, the treatment of aggressive behavior. Verbal ability is an important component of treatment readiness. Without the ability to make requests (mands) or engage in coherent conversation individuals are unlikely to develop prosocial assertiveness skills. Instead, they may rely on violent and coercive acts to meet their needs. Similarly, a key part of therapy is disclosing feelings and emotion. Without the ability to label these emotions, thoughts and feelings (tacts), therapy will necessarily become very difficult. It may be that verbal assessment guided by Skinner’s terminology could function to effectively identify behavioral readiness for treatment (i.e. patients with a sufficient grasp of mands, tacts and intraverbals). This would work in much the same way that children and adults with communication difficulties are assessed using this methodology. References Alford, G. S., & Turner, S. M. (1976). Stimulus interference and conditional inhibition of auditory hallucinations. Journal of Behavior Therapy and Experimental Psychiatry, 7, 155-160. 66 Almvik, R., Woods, P., & Rasmussen, K. (2000). The Broset Violence Checklist: Sensitivity, specificity, and interrater reliability. Journal of Interpersonal Violence, 15, 1284-1296. Belcher, T. L. (1988). Behavioral reduction of overt hallucinatory behavior in a chronic schizophrenic. Journal of Behavior Therapy and Experimental Psychiatry, 19, 69-71. Bandura, A. (1973). Aggression: A social learning analysis. Englewood Cliffs, NJ: Prentice Hall. Burns, C. E. S., Heiby, E. M., & Tharp, R. B. (1983). A verbal behavior analysis of auditory hallucinations. The Behavior Analyst, 6, 133-143. Bushman, B. J., & Anderson, C. A. (2001). Is It Time to Pull the Plug on the Hostile Versus Instrumental Aggression Dichotomy? Psychological Review, 108, 273-279. Calhoun, F. S. (1998) Hunters and howlers: Threats and violence against federal judicial officials . Arlington, VA: United States Marshals Service. Carbone, V. (2003). Promoting speech production skills in children with autism [Workshop]. Valley Cottage, NY: Carbone Clinic . Daffern, M. (2004). Unpublished PhD thesis . University of South Australia. Daffern, M., Duggan, C., Huband, N., & Thomas, S. D. M. (2008). The impact of interpersonal style on aggression and treatment non-completion in patients with personality disorder admitted to a medium secure psychiatric unit. Psychology, Crime and Law, 14, 481-492. Daffern, M., & Howells, K. (2002). Psychiatric inpatient aggression: A review of structural and functional assessment approaches. Aggression and Violent Behavior: A Review Journal. 7, 477-497. Daffern, M., & Howells, K. (in press a). The function of aggression in personality disordered patients. Journal of Interpersonal Violence . Daffern, M., Howells, K., & Ogloff, J. R. P (2006). What’s the point? Towards a methodology for assessing the function of psychiatric inpatient aggression. Behavior Research and Therapy, 45(1), 101-111. Fottrell, E. (1980). A study of violent behavior among patients in psychiatric hospitals. British Journal of Psychiatry, 136, 216-221. Frost, L., & Bondy, A. (2006). A common language using B. F. Skinner’s verbal behavior for assessment and treatment of communication disabilities in SLP-ABA. The Journal of SpeechLanguage Pathology and Applied Behavior Analysis, 1(2), 103-109. Gadon, L., Cooke, D. J., & Johnstone, L. (2006) Situational variables and institutional violence: A systematic review of the literature. Clinical Psychology Review, 26, 515-534. Glenn, S. S. (1983). Maladaptive functional relations in client verbal behavior. The Behavior Analyst, 6, 47-56. Goldiamond, I. (1975a). Alternative sets as a framework for behavioral formulations and research. Behaviorism, 3 , 49-86. Goldiamond, I. (1975b). Insider-outsider problems: A constructional approach. Rehabilitation Psychology, 22, 103-116. Heilbrun, K., Hart, S. D., Hare, R. D., Gustafson, D., Nunez, C., & White, A. J. (1998). Inpatient and post discharge aggression in mentally disordered offenders: The role of psychopathy. Journal of Interpersonal Violence, 13, 514-527. Hillbrand, M., Foster, H. G., & Spitz, R. T. (1996). Characteristics and costs of staff injuries in a forensic hospital. Psychiatric Services, 47, 1123-1125. Indermaur, D. (1999). Situational prevention of violent crime: Theory and practice in Australia. Studies on Crime and Crime Prevention , 8, 71-87. 67 James, D. V., Fineberg, N. A., Shah, A. K., & Priest, R. G. (1990). An increase in violence on an acute psychiatric ward: A study of associated factors. British Journal of Psychiatry, 156, 846852. Lancee, W. J., Gallop, R., & McCaye -Toner, B. (1995). The relationship between nurses’ limit setting styles and anger in psychiatric inpatients. Psychiatric Services, 6, 609–613. Layng, T. V. J., & Andronis, P. T. (1984). Toward a functional analysis of delusional speech and hallucinatory behavior. The Behavior Analyst, 7, 139-156. Monroe, C., Van Rybroek, G. J., & Maier, G. J. (1988). Decompressing aggressive inpatients: Breaking the aggression cycle to enhance positive outcome. Behavioral Sciences and the Law, 6, 543-557. Novaco, R. W. (1976). The functions and regulation of the arousal of anger. American Journal of Psychiatry, 133 , 1124-1128. Novaco, R. W. (1994). Anger as a risk factor for violence among the mentally disordered. In J. Monahan & H. J. Steadman (Eds.), Violence and mental disorder (pp. 137-159). Chicago: University of Chicago Press. Ogloff, J. R. P., & Daffern, M. (2006). The Dynamic Appraisal of Situational Aggression: An instrument to assess risk for imminent aggression in psychiatric inpatients. Behavioral Sciences and the Law, 24, 799-813. Owens, R. G., & Ashcroft, J. B. (1982). Functional analysis in applied psychology. British Journal of Clinical Psychology, 21, 181-189. Partington, J., Sundberg, M., Newhouse, L., & Spengler-Schelley, M. (1994). Overcoming an autistic child’s failure to acquire a tact repertoire. Journal of Applied Behavior Analysis, 27, 733-734. Paul R., & Sutherland, D. (2005). Enhancing early language in children with autism spectrum disorders. In F. R. Volkmar, R. Paul, A. Klin, & D. Cohen (Eds.), Handbook of autism and pervasive developmental disorders, Vol. 2: Assessment, interventions, and policy (3rd Ed., pp.946-976). Hoboken, NJ: Wiley. Salzinger, K. (1958). A method of analysis of the process of verbal communication between a group of emotionally disturbed adolescents and their friends and relatives. Journal of Social Psychology, 47, 39-53. Shepherd, M., & Lavender, T. (1999). Putting aggression into context: An investigation into contextual factors influencing the rate of aggressive incidents in a psychiatric hospital. Journal of Mental Health, 8, 159-170. Skinner, B. F. (1957). Verbal behavior. Englewood Cliffs, NJ: Prentice-Hall. Skinner, B. F. (1981). How to discover what you have to say- A talk to students. Behavior Analyst, 4, 1-7. Skinner, B. F. (1993). About Behaviorism. London: Penguin Books. Sloane, H. N., Endo, G. T., & Della -Piana, G. M. (1980). Creative behavior. Behavior Analyst, 3(1), 11-21. Sundberg, M. L. (1991). 301 Research topics from Skinner’s book Verbal Behavior. The Analysis of Verbal Behavior, 9, 81-96. Sundberg, M., & Michael, J. (2001). The benefits of Skinner’s analysis of verbal behavior for children with autism. Behavior Modification, 25, 698-724. Sundberg, M., Michael, J., Partington, J., & Sundberg, C. (1995). The role of automatic reinforcement in early language acquisition. Analysis of Verbal Behavior , 13 , 21-37. Warren, L. J., Mullen, P. E., Thomas, S. D. M., Ogloff, J. R. P., & Burgess, P. M. (2007). Threats to kill: A follow -up study. Psychological Medicine, 38, 599–605. 68 Whittington, R., & Richter, D. (2005). Interactional aspects of violent behavior on acute ps ychiatric wards. Psychology, Crime and Law, 11, 377-388. Yesavage, J. A. (1983). Inpatient violence and the psychiatric patient. Acta Psychiatrica Scandinavica, 67, 353-357. Zoellner, R. (1969). Talk-write: A behavioral pedagogy for composition. College English, 30, 267320. Author Information Michael Daffern Centre for Forensic Behavioral Science c/o Thomas Embling Hospital, Locked Bag 10, Fairfield, Victoria 3078 Australia Tel +61 394959160 Fax +61 394959195 Email [email protected] Affiliations Centre for Forensic Behavioral Science School of Psychology, Psychiatry and Psychological Medicine Monash University Clayton Melbourne, 3800, Australia, Victorian Institute of Forens ic Mental Health (Forensicare), Locked Bag 10, Fairfield, Victoria, 3078 Australia Peaks Academic and Research Unit Rampton Hospital, Retford, Nottinghamshire, DN220PD United Kingdom Division of Psychiatry, Forensic Mental Health Section Nottingham University, Nottingham, NG3 6AA United Kingdom Matthew Tonkin Peaks Academic and Research Unit Rampton Hospital Retford, Nottinghamshire, DN220PD United Kingdom, 69 Evaluation of Two Communicative Response Modalities for a Child with Autism and Self-Injury Stacy E. Danov, Ellie Hartman, Jennifer J. McComas, and Frank J. Symons Abstract There is little empirically replicated guidance from the research literature on selecting a communicative response modality when implementing functional communication training (FCT). In this study, two forms of communicative responding (verbal speech and picture cards) were evaluated during functional communication training (FCT) treatment of self-injury for a three-year-old boy with autism. The functional analysis indicated the self-injury was maintained by positive reinforcement in the form of access to preferred items. Findings indicated (a) SIB was eliminated during FCT sessions, and (b) independent picture cards (but not verbal speech) were used in all evaluation sessions. Results are discussed in relation to the clinical issue of choosing among different possible communication response modalities to effectively compete with severe problem behavior. Keywords: Communication response, Self-Injurious Behavior, Functional communication training, mand selection Introduction Severe problem behavior among children with pervasive developmental disorders including autism is relatively prevalent (Horner, Carr, Strain, Todd & Reed, 2002) and costly both to the individual and society (Schroeder, Rojahn, & Oldenquist, 1989). Severe forms of behavior problems such as selfinjurious behavior (SIB) or aggression can hinder communicative development and limit verbal and nonverbal communication capacity (National Research Council, 2001). Behavioral interventions based on the function of the problem behavior have been successfully applied to children with pervasive developmental disorders for a range of problem behaviors including SIB (Carr & Durand, 1985) and aggression (Richman, Wacker, & Winborn, 2001). Identifying a functional relation through analyses designed to expose reinforcement contingencies before treatment increases the likelihood of developing a targeted function-matched intervention to decrease problem behavior and increase adaptive behavior. It is important to identify a behavioral function prior to treatment selection for severe behavioral problems because interventions based on behavioral function are more likely to be effective than arbitrarily chosen interventions (Carr & Durand, 1985; Repp, Felce, & Barton, 1988; Wacker et al., 1998; Wacker et al., 2005). Among behavioral interventions, functional communication train ing (FCT) consists of teaching communicative responses such as words, gestures, or signs that are used to effectively compete with problem behavior by producing the same functional reinforcer (Carr, 1988; Wacker et al., 1996). Carr and Durand (1985) demonstrated the power of such an approach with children (N = 4) diagnosed with autism by developing a functional assessment tool to identify environmental conditions in which problem behaviors such as aggression, self-injury, and tantrums occurred. The results of the functional assessment were used to define and teach an appropriate replacement behavior with corresponding reductions reported in problem behavior. Wacker et al. have demonstrated consistently (e.g., 1998, 2005) that FCT is highly effective in reducing problem behavior displayed by individuals with developmental disabilities including autism. Less clear, however, is empirical guidance on selecting the type or form of communicative response to be used during FCT when more than one form (e.g., verbal, gestural) concurrently exists in 70 the child’s repertoire. Recently, Ringdahl et al. (2008) compared two mand topographies (high and low proficiency) during FCT and concluded that FCT was more effective when the high proficiency mand was incorporated into FCT. There are only a limited number of studies, however, explicitly examining the effects of alternative modes of communication on verbal communication in autism and related pervasive developmental disorders. Bondy and Frost (1998) reported that alternative modes of communication do not prohibit the acquisition of verbal behavior, but may actually promote it. They demonstrated that a boy who began training with picture cards began speaking after using the system for 11 months. Eighteen months later his speech replaced the cards as his mode of communication. More recently, Charlop-Christy, Carpenter, Le, LeBlanc, and Kellet (2002) reported an increase in spontaneous and imitative speech following the implementation of picture cards. Further, Ganz and Simpson (2004) found that picture card use was mastered rapidly and word utterances increased in number of words and complexity. The use of an alternative mode of communication in the form of a mand or request for a functional reinforcer, can also lead to a decrease in problem behavior. Frea, Arnold, and Vittimberga (2001) reported that the problem behavior of a four-year-old boy with autism decreased when he started using the Picture Exchange Communication System. Winborn et al. (2002) showed for two subjects that both existing and novel requests were effective replacements for problem behavior using a concurrentschedules design, without a reversal or extinction phase. Horner and Day (1991) and Richman, Wacker, and Winborn (2001) demonstrated that when replacing a problem behavior with a request during FCT, response efficiency is important. Problem behavior and requests can be viewed as concurrently available response options, with the goal being to promote the use of the request rather than problem behavior to access reinforcement. In this single -case study, two communicative response forms (speech, picture cards) were compared during ongoing FCT in which a child with autism was being taught to appropriately request toys and related materials to compete with SIB. Reinforcers and response forms were selected following a functional analysis. Comparisons of the occurrence of SIB and requests were made across both communicative forms using a within-subject ABAB design. Method Participant John was a 3 year 2 day old boy diagnosed with autism. John was previously diagnosed with autism by a licensed psychologist based in the DSM-IV-TR. He received home-based services consisting of Applied Behavior Analysis (ABA) therapy to develop communicative, social, educational, and behavioral skills. Reportedly, these services included using a discrete trial method based on individual treatment goals focusing on spoken communication. He participated in approximately 20-30 hours per week consisting of both individual and family skill sessions. The program selection and target goals were determined by parents and trained behavioral professionals. He also attended a pre-school program three days a week. The pre-school or school based program consisted of functional skills and speech and language goals. The frequency (daily/weekly) and intensity of John’s self-injury lead to tissue damage and bruising. Forms of self-injury specifically included hitting the front and back of his head to the floor, biting and scratching his forearm, and hitting his forearm to an object. John had limited communication skills. Expressively, he could verbalize single words when prompted, but his articulation was poor and his speech was often difficult to understand. Receptively, he could follow short, one-step directions. At the time of the investigation, he was being introduced to a picture exchange system as part of his school communication program, which was introduced by the teacher and speech practitioner and verbal communication was introduced as part of his home-based behavior therapy program by the behavior therapist. 71 Procedures Phase 1: Functional Analysis General procedure. First, a functional analysis interview (adapted from O’Neil et al., 1997) was conducted with John’s mother and home-based therapist by a graduate research assistant to collect information about the environmental and social events influencing problem behavior. Next, direct observation of the target problem behavior was conducted to clarify and validate the interview findin gs and gather further information regarding the target problem behavior and the social context in which it occurred. Finally, an experimental (i.e., functional) analysis was conducted at home in John’s therapy room. His mother was coached by a trained graduate research assistant to conduct the functional analysis sessions. Materials used during the functional analysis included puzzles, books, animal toys, a ball, and cards with graphic symbols (cards were used in tasks involving matching 2-D pictures to 3-D items). The functional analysis used a brief multi-element design to evaluate the influence of social reinforcement contingencies on John’s SIB (Northup et al., 1991). Based on the outcome of the descriptive assessment, two behavioral mechanisms (positive and negative social reinforcement) were tested through three conditions including contingent attention, contingent access to tangibles, and contingent escape from task demand. A control condition in the form of free play was also conducted. During the attention condition, John’s mother sat on the sofa and read a book while John played alone. Approximately 10 s of attention in the form of touching his arm and saying, “Keep your hands down,” was delivered contingent on each occurrence of SIB. During the tangible condition, John had continuous access to his mother’s attention, but access to prefer play items/toys was restricted. Contingent on each occurrence of SIB, John’s mother provided him with 10 s access to the preferred toys. During the escape condition, John’s mother directed him to complete tasks consistent with his ABA therapy program, such as matching items, receptive labeling, puzzles and imitation. Contingent on each occurrence of SIB, the task was removed for 10 s. After 10 s, the task demand was re-presented. During the free play condition John had access to preferred toys and his mother’s attention, and no task demands were delivered. All sessions were 5 min in length and were videotaped. Dependent measure. Self-injury directed to John’s head was selected as the primary dependent variable based on physician and family concerns. Any instance of hitting the front and back of his head to the floor or wall was recorded (event-count) during the 5-minute session. Inter-Observer agreement. Twenty-seven percent of the sessions were coded by an independent second observer. Inter-observer agreement (IOA) was calculated by comparing the frequency of the recorded behavior by one observer with that of the second independent observer for the 5-minute session. Percent total agreement was determined by taking the smaller rating and dividing it by the larger rating and the results are multiplying by 100 (Primavera, Allison, & Alfonso, 1997). The mean IOA for SIB was 100%. Phase 2: Preference Assessment Procedure. Following the functional analysis, a variation of a multiple stimulus preference assessment with replacement was conducted by a graduate research assistant (Windser et al., 1994) to identify highly preferred play items for use during functional communication training and to verify the items were items he elected to play with. Items he liked to play with were drawn from the same pool of items used in the functional analysis. Items were placed in groups on the table, bookshelves, and on the floor in the room. John was allowed to wander about the room and pick up item. After he selected a toy he was allowed to play with it for as long as he wanted to. At the point in which John was done playing with the toy and he dropped the toy from his hands, the toy was removed, placed back into the items on 72 the table, bookshelf or floor and the procedure began again. Three sessions were conducted, each lasted approximately 10 min. Dependent measure. The dependent measure for the multiple stimulus preference assessment with replacement was the duration (seconds) of engagement for each toy for John. Phase 3: Functional Communication Treatment (FCT) General procedure. Based on the results of the functional analysis (see below) that suggested John’s SIB was reinforced by access to preferred toys, FCT was implemented to teach a communication response as an alternative to SIB (Carr & Durand, 1985). A graduate research assistant taught John to request, both verbally and with picture cards, for preferred toys and items. An ABAB reversal design was used to show the effects of FCT between the two communicative response forms. Session length ranged from 10 to 20 min. At the end of each session, John received a 5-min break with no access to the items used during the request training. The treatment sessions were conducted in John’s therapy room at a small picnic table. Items identified in the preference assessment as being highly preferred (i.e., the most number of seconds engaged with) were used in the treatment sessions to increase the likelihood that John would request the items. Verbal request sessions. The request procedures from Hartman and Klatt (2005) were used. The experimenter sat across the table from John and presented a single item in a counterbalanced order in front of John. The experiment asked, “What do you want?” A 3 s prompt delay procedure was used to transfer stimulus control from the experimenter’s prompt to the presence of the item. In the first trial the experimenter provided an immediate prompt, “(name of item).” After the first trial the verbal prompt was delayed 3 s followed by a prompt “(name of item).” If John gave a correct response (with or without a prompt) John received both verbal praise (e.g., “Good job, you want the (name of the item”) and access to the item for 10 s. If John made an error during a trial or did not respond before the prompt for two consecutive trials, that item was terminated for that trial. In the next trial the experimenter again waited 3 s before giving a prompt. If John responded before or after the question, the response was recorded as an independent correct request. Five consecutive trials for each item were presented in each session. If John turned away from the toy or asked for a different item after the experimenter asked, “What do you want?” the experimenter stopped the trial. If two consecutive trials were stopped, the trials for that item were ended for that session. Dependent measure. The dependent variable in verbal requesting training was the frequency of independent verbal requests. An independent verbal request was defined as John verbally requesting the item presented by the experimenter (without prompts). Picture card request sessions. The research assistant sat across the table from John and presented him a picture card board. First, only one picture was presented on the picture board, so that the verbal and picture card procedure would be the same. John requested independently almost immediately, so another card was added to the board. He demonstrated that he could discriminate between the cards, so more cards were added until the picture board was full. The board consisted of pictures of items identified as highly preferred from the preference assessment and other items requested by his mom (e.g., milk). Graphic symbols/icons from Microsoft Clipart of the preferred items were placed on 2 inch by 2 inch note cards that were attached with Velcro onto an 8 by 11 inch pieces cardboard. The cardboard held a total of 12 picture cards. The experimenter asked John, “What do you want?” and placed her hand out for John to place a card into. A 3 s prompt delay procedure was used. In the first trial the experimenter provided an immediate prompt and used hand over hand prompting to have John choose a picture card and place it in the experimenter’s hand. After the first trial the physical prompt was delayed 3 s and a verbal prompt was given, (“What do you want?”). If John gave the correct response (with or without a prompt) he received both verbal praise (“Oh you want the (name of the item), good job!”) and access to the item for 10 s. John 73 was able to request any item he asked for, even if it was the same item for the entire time of the session. If SIB occurred during FCT, it was not reinforced. If John verbally requested an item, he was prompted to use the card and then reinforce his behavior, but John did not have any verbal requests during the picture card request sessions. Dependent measure. The dependent variable for the FCT picture card intervention was the frequency of independent correct picture card requests. An independent correct request was defined as John pulling a picture off of the picture communication board and handing it to someone else without prompts (event-count). Extinction of pictu re card. The procedures for this condition were the same as the FCT picture card intervention with a key exception. If John requested with the picture card, he was praised (e.g., “That was a nice way to ask.”) but was not given access to the item (i.e., their use did not produce access to a requested item). John could only receive access to the preferred item if he verbally requested (e.g. said, “Milk”). Inter-Observer agreement for FCT. Twenty-two percent of the sessions were coded by a second independent observer. IOA was calculated by comparing the frequency of the recorded behavior by one observer with that of the second independent observer for the 5-minute session. Percent total agreement was determined by taking the smaller rating and dividing it by the larger rating and the results are multiplying by 100. The mean IOA for SIB was 100%. The mean IOA for independent requests was 91.5% (Range = 83% -100%). Treatment integrity. Treatment integrity was calculated for 28% of the treatment sessions randomly selected across all conditions. Treatment integrity was calculated for as a percentage of requests that were followed by access to the tangibles requested. Treatment integrity was 100%. Results Phase 1: Brief Functional Analysis The results of the functional analysis can be seen in Figure 1. The data indicate a differential pattern of SIB responding with elevation during conditions associated with positive reinforcement in the form of contingent access to tangibles (i.e., toys). SIB occurred 10 times during both tangible conditions. SIB either did not occur or occurred at very low frequencies in the other conditions (attention, demand, and control). Phase 2: Preference Assessment Results of the preference assessment showed that the duration of engagement with the toys ranged widely and included in descending order: toy sheep (543 s), toy horse (262 s), toy car (159 s), beads (63 s), blocks (60 s), toy chicken (38 s), and toy bus (5 s). Following maternal request, milk was also included as a preferred item accessible during FCT. Phase 3: Functional Communication Treatment Results of treatment are shown in Figure 2. When picture cards were available (condition 2, sessions 3-6; condition 4, sessions 9-11; condition 6, sessions 14-15; condition 8, session 18), John independently requested using pictures (approximately 1.5 times per minute). During verbal conditions, a single independent verbal response was made in session 7 and not again in verbal conditions. Following the initial ABABAB evaluation, pictures were put on extinction in sessions 16 and 17. In the picture extinction condition use of pictures did not produce access to a requested item. No corresponding 74 ‘crossover’ to verbal requesting was observed. During FCT, SIB was observed only during the first verbal FCT condition. Figure 1. Frequency of self-injurious behavior (SIB) during functional analysis conditions. Discussion Severe problem behavior among children with developmental disorders including autism can interfere with communicative development. Identifying the function of the problem behavior can lead to appropriate function-matched targeted interventions that are often communication based. When the child has two or more possible communicative response modalities already in his or her repertoire, no clear guidelines exist for practitioners or parents to choose among them. In this single -case demonstration, two existing communication modalities were directly compared following a functional analysis of self-injury. Overall, the results showed clearly the effectiveness of a picture versus a vocal (i.e., verbal) response modality for this child to request preferred items that competed with and functionally replaced self-injury. Although SIB did not occur after the initial verbal condition, verbal sessions were not associated with any independent requests. 75 £ = independent requests ¢ = SIB (self-injurious behavior) Note: All independent requests were picture requests; none were verbal. During sessions 16 and 17, picture cards were placed on extinction but responding continued to occur exclusively in the form of picture requesting. Figure 2. Frequency per minute of independent requests and SIB during FCT intervention (verbal and picture requesting training) and extinction trials of picture card. Because the participant was a clinical referral not selected randomly the results are necessarily limited and are not generalizable to other children his age with autism and SIB or communication difficulties. Because SIB was only observed during the first session, it is impossible to infer that one modality was superior to his SIB. If access to picture cards was removed for longer periods of time, SIB may have reoccurred. It would be predicted that problem behavior would be more likely to occur if a child is required to use a mode of communication that he/she cannot use independently, consistent with Ringdahl et al.’s (2008) recent demonstration. Anecdotally, John appeared more engaged and attentive during picture sessions when compared to verbal sessions. The overall time scale (and therefore the intensity) of the intervention was limited. Past research has reported that over longer intervention periods the acquisition of verbal language may appear during picture card training (Bondy & Frost, 1998; Charlop-Christy et al., 2002). In addition to response efficiency and reinforcement, another possible reason for the clear differentiation between the picture and verbal communication response options could be related to the structural differences between the instruction sessions themselves that made them nonequivalent. Because the picture board contained many pictures and John could discriminate between the pictures, he was able to select from the full array of cards associated with any toy he may have wanted at the moment. Although, in principle, he could verbally ask for any toy during the verbal request training session, he never made an independent verbal request. Because of this, we ‘rotated’ through possible preferred toys during verbal instruction sessions , but it remains possible that we were prompting him to request toys that he did not want at the moment. For this reason, we created an extinction condition. Sessions during the 76 extinction condition allowed John to choose the toy he wanted from the full pic ture array in which he could see the pictures but was required to use a verbal request for a toy. However, no independent verbal requests occurred during the extinction condition sessions either. In related areas of clinical research, studies examining picture communication found the picture exchange communication system (PECS) to be effective for teaching functional communication to children with limited speech (Bondy and Frost, 2001; Charlop-Christy, Carpenter, Le, LeBlanc, and Kellet, 2002; Ganz and Simpson, 2004), while studies examining manual signs or total communication found faster receptive and/or expressive vocabulary acquisition than speech alone (Brady & Smouse, 1978; Barrera, Lobato-Barrera, & Sulzer-Azaroff, 1980). Durand (1999) found assistive devices to be effective for communication in recruiting natural communities of reinforcement. Although the present study was not designed to demonstrate the acquisition of different communication modalities per se, the results suggest that picture based requests were more likely to occur and produce a functional reinforcer than verbally based requests at this time for this young boy with autism. Similar to the results reported here, Frea, Arnold and Vittimberga (2001) found that problem behavior was decreased when picture were introduced as a mode of communication for a four year old boy with autism. Winborn et al. (2002) showed that training both novel requests and existing requests can be effective for replacing problem behavior. Additionally, Horner and Day (1991) and Richmand, Wacker, and Winborn (2001) demonstrated that response efficiency is important when replacing a problem behavior with a request during FCT. In this study, it may be that picture cards were a more efficient alternative thereby reducing problem behavior. The applied behavioral literature on FCT and behavioral problems shows clearly the importance of teaching a replacement behavior (Carr & Durrand, 1985; Wacker et al., 1998). One of the most effective routes for determining what function the replacement behavior should serve is based on conducting a functional analysis prior to beginning intervention (Durand, 1999; Horner et al., 2002). But, determining what form the replacement behavior might take in relation to gains in adaptive behavior and reductions in problem behavior remains a relatively unexplored clinical area in need of further research. References Barrera, R., Lobato-Barrera, D., & Sulzer-Azaroff, B. (1980). A simultaneous treatment comparison of three expressive language training programs with a mute autistic child. Journal of Autism and Developmental Disorders, 10, 21-37. Bondy, A., & Frost, L. (1998). The Picture Exchange Communication System. Seminars in Speech and Language, 19, 373-389. Bondy, A., & Frost, L. (2001). The Picture Exchange Communication System. Behavior Modification, 25, 725-744. Brady, D. O., & Smouse., A.D. (1978). A simultaneous comparison of three methods for language training with an autistic child: an experimental single case analysis. Journal ofAutism and Childhood Schizophrenia, 8, 271-279. Carr, E. G., & Durand, V. M. (1985). Reducing behavior problems through functional communication training. Journal of Applied Behavior Analysis, 18, 111-126. Charlop-Christy, M. H., Carpenter, M., Le, L., LeBlanc, & Kellet, K. (2002). Using the Picture Exchange Communication System (PECS) with children with autism: Assessment of PECS acquisition, speech, social-communicative behavior, and problem behavior. Journal of Applied Behavior Analysis, 35, 213-231. 77 Durand, V. M. (1999). Functional communication training using assistive devices: recruiting natural communities of reinforcement. Journal of Applied Behavior Analysis, 32, 247-267. Frea, W. D., Arnold, C. L., & Vittimberga, G. L. (2001) A demonstration of the effects of augmentative communication on the extreme aggressive behavior of a child with autism within an integrated preschool setting. Journal of Positive Behavioral Interventions, 3, 194-198. Ganz J. B., & Simpson, R. L. (2004). Effects on communicative requesting and speech development of the picture exchange communication system in children with characteristics of autism. Journal of Autism and Developmental Disorders, 34, 395-409. Hartman, E. C. & Klatt, K. P. (2005). The effects of deprivation, precession exposure, and preferences on teaching manding to children with autism. The Analysis of Verbal Behavior, 21, 135-144. Horner, R. H., Carr, E. G., Strain, P. S., Todd, A. W., & Reed, H. K. (2002). Problem behavior interventions for young children with autism: A research synthesis. Journal of Autism and Developmental Disorders, 32, 423-446. Horner, R. H., & Day, H. M. (1991). The effects of response efficiency on functionally competing behaviors. Journal of Applied Behavior Analysis, 24, 719- 732. equivalent Iwata, B. A., Dorsey, M. F., Slifer, K. J., Bauman, K. E., & Richman, C. S. (1994). Toward a functional analysis of self-injury. Journal of Applied Behavior Analysis, 27, 197-209. (Reprinted from Analysis and Intervention in Developmental Disabilities, 2, 3-20). National Research Council. (2001). Educating Children with Autism. Washington, DC: National Academy Press. Northup, J., Wacker, D., Sasso, G., Steege, M,, Cigrand, K., Cook, J., & DeRaad ,A. (1991). A brief functional analysis of aggressive and alternative behavior in an outclinic setting. Journal of Applied Behavior Analysis, 24, 509-522. O’Neil, R.E., Horner, R.H., Albin, R.W., Sprague, J.R., Storey, K., & Newton, J.S. (1997). Functional assessment and program development for problems behavior: A practical handbook. Pacific Grove, CA: Brooks/Cole Publishing Company. Primavera, L. H., Allison, D. B., & Alfonso, V. C. (1997). Measurement of dependent variables. In R. D. Franklin, D. B. Allison, & B. S. Gorman (Eds.), Design and analys is of single-case research (pp. 41- 91). Mahwah, NJ: Erlbaum. Repp, A., Felce, D., & Barton, L. (1988). Basing the treatment of stereotypic and self-injurious behaviors on hypotheses of their causes. Journal of Applied Behavior Analysis, 21, 281-289. Richman, D., Wacker, D., & Windborn, L. (2001). Response efficiency during functional communication training: effects of effort on response allocation. Journal of Applied Behavior Analysis, 34, 73-76. Ringdahl, J.E., Falcomata, T.S., Christensen, T.J., Bass-Ringdahl, S.M., Lentz, A., Dutt, A., & SchuhClaus, J. (2008). Evaluation of pre-treatment assessment to select mand topographies for functional communication training. Research in Developmental Disabilities. Schroeder, S. R., Rojahn, J., & Oldenquist, A. (1989). Treatment of destructive behaviors among people with mental retardation and developmental disabilities: overview of the problem. NIH Consensus Development Conference. Bethesda, Maryland. Wacker, D. P., Berg, W. K., Harding, J. W., Barretto, A., Rankin, B., & Ganzer, J. (2005). 78 Treatment effectiveness, stimulus generalization, and acceptability to parents of functional communication training. Educational Psychology, 25, 233-256 Wacker, D. P., Berg, W. K., Harding, J. W., Derby, K .M., Asmus, J. M., & Healy, A. (1998). Evaluation and long-term treatment of aberrant behavior displayed by young children with disabilities. Journal of Developmental & Behavioral Pediatrics, 19, 260-266. Winborn, L., Wacker, D. P., Richmand, D. M., Amus, J., & Geier, D. (2002). Assessment of mand selection for functional communication training packages. Journal of Applied Behavior Analysis, 35, 295-298. Windsor, J., Piche´, L. M., & Locke, P. A. (1994). Preference testing: A comparison of two presentation methods. Research in Developmental Disabilities, 15, 439–455. Acknowledgements This research was supported, in part, by NICHD Grant No. 44763 to the University of Minnesota. The authors gratefully acknowledge the cooperation of the family and their support in completing this project. Author Contact Information Stacy E. Danov Department of Educational Psychology Education Science Building 56 River Road University of Minnesota Minneapolis, MN 55455 USA e-mail: [email protected] Phone: (612) 624-5241 Jennifer McComas Department of Educational Psychology Education Science Building 56 River Road University of Minnesota Minneapolis, MN 55455 USA e-mail: [email protected] Phone : ( 612) 624-5854 Ellie Hartman U. of Wisconsin-Stout Vocational Rehab. Institute Pathways to Independence Projects Wisconsin Department of Health and Family Services 1 W Wilson St Madison, WI 53703-3445 USA e-mail: [email protected] Phone: (608) 266-2756 Frank J. Symons Department of Educational Psychology Education Science Building 56 River Road University of Minnesota Minneapolis, MN 55455 USA e-mail: [email protected] Phone: (612) 626-8697 79 Using DRO, Behavioral Momentum, and Self-Regulation to Reduce Scripting by an Adolescent with Autism Vanessa Ann Silla-Zaleski and Mary J. Vesloski Abstract This case study report describes the use of behavioral momentum, differential reinforcement of other behavior (DRO), and self-regulation to decrease vocal scripting behavior produced by a 12-year-old male with autism, obsessive compulsive disorder, and attention deficit hyperactivity disorder. The behavior was documented using partial interval time sampling throughout the day on 67 school days over a 4-month period. DRO was delivered on a graduated fixed interval schedule. During the self management phase of the program, a MotivAider® was used by the participant to monitor the beginning and end of each interval. A dependent paired samples t-test was used to compare initial and final rates of scripting. Results showed a decrease from an average of 44% to an average of 29%. The implications and limitations of this study are discussed. Keywords : autism, self stimulatory behavior, vocal scripting, differential reinforcement, self regulation, behavioral momentum, MotivAider, fixed interval, DRO Review of the Literature Self -Stimulatory Behavior Considerable impairments in both social and communicative behavior affect most individuals with autism (Rutter & Schopler, 1978). For example, research indicates that during vocal exchanges and in social settings, these individuals often produce forms of verbal behavior that significantly differ from the speech produced by individuals without autism (Fine, Bartolucci, & Szatmari, 1994). This verbal behavior is often categorized as self-stimulatory or stereotyped behavior, and it consists of recurring physical movements or vocalizations that serve no obvious function in the external environment (Harris & Wolchik, 1979). When high rates of delayed echolalia (Fine et al.), also known as scripting, occur in conjunction with low rates of appropriate conversational skills, the social relational opportunities for persons with autism may be greatly reduced (Ross, 2002). Additionally, self-stimulation may interfere with learning or performance (Koegel and Covert, 1972). Since self-stimulation can be potentially harmful to the student as a learner and to the individual as a peer, a number of studies have investigated procedures that could decrease this behavior. Laws, Brown, Epstein, and Hocking (1971) decreased self-stimulatory behavior by directing a teacher to remove attention when his students self-stimulated and to reinforce suitable behavior when they attended. Azrin, Kaplan, and Foxx (1973) reduced self-stimulation in nine individuals with mental retardation by instructing them on the suitable use of vocational and recreational materials. The present study adds to this literature by demonstrating a decrease in the production of self-stimulatory behavior of vocal scripting by an adolescent male with autism, obsessive compulsive disorder, and attention deficit hyperactivity disorder through the use of differential reinforcement, self-regulation and behavioral momentum. Differential Reinforcement of Other Behavior (DRO) All forms of differential reinforcement entail “reinforcing one response class and withholding reinforcement for another response class” (Cooper, Heron, Heward, 2007, p. 470). One of the most frequently used types of differential reinforcement is the differential reinforcement of other behavior (DRO) (Wolery, Bailey, & Sugai, 1988). DRO schedules reverse a contingency by delivering reinforcers based on the absence of a target behavior (Baer, Peterson, & Sherman, 1967). This type of reinforcement 80 delivery is sometimes referred to as omission training since delivery of the reinforcer is contingent upon the omission of the target behavior (Weiher & Harman, 1975). DRO is a useful behavior-reduction procedure for several reasons. First, it highlights the use of positive reinforcement while avoiding the use of aversive stimuli. As a result, many adverse side effects may be avoided. Secondly, target behaviors tend to reduce rather quickly under conditions of DRO, especially if a specific replacement behavior is reinforced (Wolery, Bailey & Sugai, 1998). Lastly, DRO has been shown to be useful in decreasing or eliminating a variety of behavioral excesses, physical aggression and tantrums (Allen, Gottselig, & Boylan, 1982). Behavioral Momentum Generally speaking behavioral momentum involves the use of a series of high-probability requests to increase compliance with lower-probability, instructor-issued requests. Over time, in order to increase generalization and maintenance, fading procedures are implemented such that the interval of time between high probability requests to low probability requests is increased and the ratio of high probability requests to low probability requests is decreased (Ray, Skinner & Watson, 1999). Previous research has demonstrated the benefits of behavioral momentum for increasing compliance with low-probability requests across commands, demands, or requests (Ray, Skinner & Watson, 1999). It has been used successfully with children with autism and other developmental delays to increase compliance in various settings. Mace (1988) used behavioral momentum to increase compliance with low-probability commands when addressed to adults. Since this experiment, other researchers have demonstrated that antecedent high-probability commands can be used to increase the likelihood of gaining student compliance with lower probability commands (Ducharme & Worling, 1994; Rortvedt & Miltenberger, 1994). Self -Regulation Self-regulation develops within the first few months of life, when an infant begins to take interest in the environment while also regulating his or her arousal and responses to sensory input (DeGangi, 1991). As infants develop, they normally demonstrate a growth in more refined self-regulatory competencies (Gomez & Baird, 2005). However, self-regulation difficulties have been cited as among the first indicators of autism (Gomez & Baird, 2005). Gomez and Baird’s (2005) research found that children with autism were reported by their parents to have exhibited significantly more self-regulatory deficits at 1 year of age than the comparison group. Self-regulation is an important concept to teach individuals with autism because it can lead to empowerment and an enhanced quality of life (Suk-Hyang, Simpson, & Shogren, 2007). Positive outcomes of teaching self-management procedures to individuals with autism have been reported by several researchers. Newman, Buffington, and Hemmes (1996) demonstrated the benefits of using a selfmanagement strategy to help three adolescents with autism follow an activity schedule and transition within their included classrooms. Morrison, Kamps, Garcia, and Parker (2001) used self-monitoring techniques to teach 44 individuals with autism to monitor their social interaction skills when engaged in game-play with typical peers. Self-management methods have also been used successfully with preschoolers with autism to increase sharing behavior (Reinecke, Newman, & Meinberg, 1999). Methodology Participant The participant in this case study, herein known as “Zack”, was a 12-year-old male with autism spectrum disorder, obsessive compulsive disorder, and attention deficit hyperactivity disorder. These 81 diagnoses were established prior to the beginning of this study by an independent evaluator based on DSM-IV criteria (American Psychiatric Association, 1994). Zack lived in northern New York and attended a regular elementary school in a typical 5th grade classroom. He was pulled out of his classroom twice a day for support services in math and reading comprehension. He also received three, 30-minute speech therapy sessions and two, 30-minute occupational therapy sessions per week. Zack had an individualized education program (IEP) which included adaptations and modifications for him to be included in the regular education environment with typical peers. He was instructed according to the class-wide curriculum with the support of a one-on-one aide throughout the day, five days per week. Zack was verbal and could answer questions addressed to him. He was considered by the staff to be an average functioning individual. However, prior to the start of this program, Zack produced scripting behavior on frequently throughout the day. It consisted of vocalizing words and/or sentences previous ly heard in videos, TV shows, commercials or video games. This behavior was disruptive in the classroom, especially when vocalizations were loud. In fact, it placed his continued enrollment in the general education classroom at risk. It was also disruptive in one-on-one settings (e.g., speech therapy, occupational therapy) and in remedial services. One month prior to the initiation of the intervention described in this paper, an attempt was made to decrease Zack’s scripting in the general education classroom by using a token system with response cost. However this procedure did not reduce the scripting behavior because reinforcement could not be provided at the required level of intensity in that environment. Also at about one month prior to the initiation of this intervention, Zack began engaging in low frequency aggressive behaviors such as yelling and hitting. This, in combination with continued scripting, resulted in a modification of his classroom setting. While he continued to be included in the general education classroom for one third of the school day, he began to receive instruction in a separate classroom with a resource teacher and an aide for two-thirds of the day. A functional assessment of Zack’s aggressive behavior was not conducted due to its low frequency of occurrence prior to the classroom modification and due to its absence once the modification was implemented. Target Dependent Variable As indicated above, scripting was the behavior targeted for reduction in this case study. Scripting was defined, topographically , as vocalizing words and/or sentences previous ly heard in videos, TV shows, commercials or video games without any apparent social function. During the week before implementation of the program described below , a Board Certified Assistant Behavior Analyst (BCABA) and members of the school staff collected five days of baseline data. These results of this assessment indicated that Zack engaged in scripting behavior , on average, every five minutes, and sometimes as frequently as every 30 seconds. Program Design and Implementation Functional Assessment: As a basis for designing the intervention program, it was necessary to conduct a functional assessment of Zack’s scripting behavior. This assessment was organized by the BCABA. Members of the school staff were provided with an operational definition of scripting and with forms on which to document antecedent-behavior-consequence (ABC) data. Patterns resulting from the data collection suggested that scripting served two separate functions: (1) socially mediated negative reinforcement (i.e., avoiding academic tasks) and (2) automatic positive reinforcement (i.e., selfstimulation). These two functions further suggested that Zack’s academic demands may have been too challenging or him and may have required too much focus on non-preferred subjects or activities. 82 Selection of Antecedent Strategies: Since the functions of scripting appeared to be avoidance and self-stimulation due to overly-challenging academic tasks and non-preferred environmental activities, it was hypothesized that scripting could be reduced in part by two related antecedent strategies: (1) modifying the academic demands through behavioral momentum and (2) enriching Zack’s educational environment with preferred objects or activities. Intervention Agents: The behavioral program described in this paper was implemented by all of Zack’s instructors. His one-on-one aide was present with him throughout the day. This individual collected the data and provided access to reinforcement. Prior to program implementation, the entire staff received training from the BCABA in the concepts of reinforcement, behavioral momentum, and selfregulation. Once the program began to be implemented, the BCABA also conducted biweekly treatment fidelity checks to ensure that program implementation and data collection were being done properly. Implementation of Antecedent Strategies: Antecedent strategies were implemented while Zack was in the resource room with his resource room teacher and his one-to-one aid. Academic demands were modified by interspersing easy and difficult tasks to create behavioral momentum. Further, the environment was enriched by the availability (upon Zack’s request) of preferred objects. Some of the objects were provided by Zack’s parents (e.g., race car, Pokemon magazines, Doritos cool ranch chips, preferred candy, etc.) and others were provided by the school (e.g., a computer software history package for which Zack had shown preference). Differential Reinforcement of Other Behavior (DRO): DRO was implemented across the entire school environment, including lunch and recess. As indicated above, Zack engaged in scripting behavior an average of once every 5 minutes during the five days of baseline data collection. Therefore, a 5-minute fixed interval (FI-5) reinforcement schedule was used initially to compete with the inappropriate behavior. This schedule was extended when the data indicated a 10% decrease in the number of intervals during which the target behavior was observed. The end of a time interval (and beginning of a new one) was signaled by a vibrator (MotivAider®) worn initially by Zack’s personal aid and later by Zack himself (see Self-Management, below). If Zack did not engage in the scripting behavior during the specified time interval, he was allowed access to reinforcement for a 2-minute period. If he engaged in the behavior during the specified interval, then the clock was reset. Reinforcers were identified and updated periodically by asking Zack to name his preferences and by noting his requests when he named them spontaneously. All reasonable requests were honored. Examples included history learning activities for the computer, magazines, and snacks. Self-Management: A self management program was implemented when the targeted behavior had decreased by 30% and when Zack had learned to refrain from scripting for an 8-minute time interval. At this point, the reinforcement schedule was increased to FR-10, and Zack began wearing the MotivAider®. It became his responsibility to inform the teaching assistant of when an interval had ended and when it was time for reinforcement. Additionally, a token program was implemented in which Zack earned one token for each 10-minute interval in which the behavior did not occur, and when he received three tokens, he could “cash them in” for reinforcement. Non-Target Dependent Variables (Laughing, Hand-Flapping) Zack periodically engaged in inappropriate laughing. When asked about it he said he was scripting silently in his head and that “[it’s] not scripting if you don’t say it out loud”. To address this, a behavioral momentum procedure was used in which effortful academic tasks were interspersed within sequences of effortless tasks such as crossword puzzles and word searches. Reinforcement then became contingent upon task completion in addition to the absence of scripting behavior. 83 Data Collection Procedures Zack was observed throughout the day for 67 days across a 4-month period. On each day, his one-on-one aid used a partial interval strategy to collect data on the total number of intervals during which Zack demonstrated any amount of scripting (scripting intervals) and the total number of intervals during which no evidence of scripting was observed (non-scripting intervals). The percentage of scripting intervals was then determined per day by dividing the total number of scripting intervals with the total number of scripting plus non-scripting intervals. On days 19 through 67, data were also collected on the percentage of intervals during which laughing and hand-flapping were observed. Results Figure 1 shows Zack’s rate of scripting from day 1 through day 67 across the 4-month period. As indicated above, each data point represents the percent of intervals per day during which scripting behavior occurred at any point within the interval. During the first 22 days, observation intervals were based on an FI-5 schedule. As the program progressed, the intervals expanded to FI-6, FI-8, and finally to FI-10 during the self-management phase. The trend line shows that scripting continued to decrease until the last day. Figure 1. Percent of Scripting Behavior per Day (Partial Interval Time Sample) (Sessions 1-67) Figure 2 shows the percent of three self-stimulatory behaviors during days 19 through 67 of the intervention period. The three behaviors included scripting, laughing, and hand-flapping. To be clear, scripting was the behavior targeted by the intervention in this study. Laughing was addressed through behavioral momentum, and a hand-flapping was not addressed at all. The rates of laughing and handflapping were tracked to determine whether non-target self-stimulatory behaviors would increase and become replacement behaviors when the targeted behavior (scripting) decreased. Figure 2 shows that the laughing and hand-flapping did not increase. Specifically, laughing decreased and hand-flapping remained the same. Figure 2. Percent of Three Self-Stimulatory Behavior per Day (Sessions 19-67) 84 A dependent paired samples t-test was run to compare the rate of scripting during the first 10 days of intervention with the last 10 days. Results are summarized in Table 1. The alpha level for the t-test was set at .05 and the significance level (p value) obtained was <.001, indicating a significant difference between initial and final rates of scripting. The mean rate of scripting for first 10 days was 44.15% and the mean for the last 10 days was 29.30%. Table 1. Results from the Paired Samples T-Test Initial Mean Final Mean 95% Confidence Interval of the Difference (Lower) 95% Confidence Interval of the Difference (Upper) t Value Significance (2-tailed) 44.1475 29.2951 9.0453 20.6596 5.116 .001 Note : Means for the Initial and final rates of scripting differ at p<.001 in the Tukey honestly significant difference comparison. Discussion In this case study DRO, self-regulation, and behavioral momentum were used to decrease the amount of scripting behavior produced by a single adolescent male with autism, obsessive compulsive disorder, and attention deficit hyperactivity disorder. A dependent paired samples t-test indicated a significant difference between the initial and final rates of scripting. Initially, Zack engaged in scripting during an average of 44% of partial intervals. At the end, he engaged in scripting during an average of only 29% of the intervals. This was a substantial decrease. This study extends the existing literature on the use of DRO, behavioral momentum and selfregulation by demonstrating their combined implementation in a school setting during academic activities and social opportunities. These procedures appeared to be effective in reducing this participant’s scripting behavior. Moreover, reductions in scripting were usually maintained, even when the length of fixed intervals increased, resulting in less frequent opportunities for reinforcement. Although positive results were observed in the data, there are some important limitations to this study. First, the intervention was based on a functional assessment and not a functional analysis. However the results of the functional assessment combined with he results of the DRO intervention support the hypothesis that escape and automatic positive reinforcement (self-stimulatory behavior) were the maintaining reinforcers for the participants’ scripting behavior. A second limitation was the case study design, which included only one participant and only one phase (i.e., intervention). As such, there is no way to determine whether changes in the dependent variable (scripting) were due to the combined effect of the independent variables (DRO, self-regulation, behavioral momentum) or whether they were due to maturation over a 4-month period. If changes in behavior had been demonstrated systematically through another type of single subject design (e.g., multiple baseline across participants; multiple baseline across targets), there would have been clearer evidence that the reducing in scripting was due to the interventions. As it stands, this case study design does not allow us to rule out the effects of maturation. Also, the case study was conducted in a school environment which was heavily staffed by an instructor and one-on-one aide at all times. It may become more difficult to implement similar procedures in a generalized setting with fewer resources, especially given the resources required to maintain data collection. This limitation was accounted for in this study through the use of teaching 85 strategies for self-regulation, thereby teaching the participant the data collection process. Continued research in more applied settings should address issues such as the effectiveness of different schedules of DRO thinning as well as the application of this program among a broader range of individuals and a greater number of participants. Future research should assess the impact of the combined interventions (DRO, behavioral momentum, self-regulation) using a single subject design with greater control than a case study. Additionally, future research should examine the use of the combined interventions for decreasing other types of self-stimulatory target behaviors. For example, it would be interesting to see if the use of these interventions can reduce the frequency of nonverbal self-stimulatory behaviors such as rocking and hand flapping. References Allen, L. D., Gottselig, M., & Boylan, S. (1982). A practical mechanism for using freetime as Reinforcer in the classroom. Education and Treatment of Children, 5, 247-253. Azrin, N. H., Kaplan, S. J., & Foxx, R. M. (1973). Autism reversal: Eliminating stereotyped selfstimulation of retarded individuals. American Journal of Mental Deficiency, 78, 241-248. Baer, D. M., Peterson, R. F., & Sherman, J. A. (1967). The development of imitation by Reinforcing behavioral similarity to a model. Journal of the Experimental Analysis of Behavior, 10, 405-416. Cooper, J. O., Heron, T. E., Heward,W. L. (2007). Applied Behavior Analysis. Second Edition. Upper Saddle River, NJ: Pearson. DeGangi, G. (1991). Assessment of sensory, emotional, and attentional problems in regulatory disordered infants: Part 1. Infants and Young Children, 3 (3), 1-8. Ducharme, J. M., & Worling, D. E. (1994). Behavioral momentum and stimulus fading in the Acquisition and maintenance of child compliance in the home. Journal of Applied Behavior Analysis, 27, 639-647. Fine, J., Bartolucci, G., & Szatmari, P., (1994). Cohesive discourse in pervasive developmental disorders. Journal of Autism and Developmental Disorders, 24, 315-329. Gomez, C. R. & Baird, S. (2005). Identifying early indicators for autism in self-regulation difficulties. Focus on Autism and Other Developmental Disabilities, 20 (2), 106-116. Harris, S. L. & Wolchik, S. A. (1979). Suppression of self-stimulation: Three alternative strategies. Journal of Applied Behavior Analysis, 12, 185-198. Koegel, R. L. & Covert, A. (1972). The relationship of self-stimulation to learning in autistic children. Journal of Applied Behavior Analysis, 5, 381-387. Laws, D. R., Brown, R. A., Epstein, J., & Hocking, N. (1971). Reduction of inappropriate social Behavior in disturbed children by an untrained paraprofessional therapist. Behavior Therapy, 2, 519-533. Morrison, L., Kamps, D., Garcia, J., & Parker, D. (2001). Peer mediation and monitoring strategies to improve initiations and social skills for students with autism. Journal of PositiveBehavior Interventions, 3, 237-250. Newman, B., Buffington, D. M., & Hemmes, N. S. (1996). Self-reinforcement used to increase The appropriate conversation of autistic teenagers. Education and Training in Mental Retardation and Developmental Disabilities, 31, 304-309. 86 Ray, K. P., Skinner, C. H., & Watson, T. S. (1999). Transferring stimulus control via momentum To increase compliance in a student with autism: A demonstration of collaborative consultation. The School Psychology Review, 28 (4), 622-628. Reinecke, D. R., Newman, B., & Meinberg, D. L. (1999). Self-management of sharing in three preschoolers with autism. Education and Training in Mental Re tardation and Developmental Disabilities, 34, 312-317. Rortvedt, A. K., & Miltenberger, R. G. (1994). Analysis of high-probability instructional sequence and time-out in the treatment of noncompliance. Journal of Applied Behavior Analysis, 27, 327-330. Ross, D. E. (2002). Replacing faulty conversational exchanges for children with autism by establishing a functionally equivalent alternative response. Education and Training in Mental Retardation and Developmental Disabilities, 37 (4), 343-362. Rutter, M., & Schopler, E. (Eds.). (1978). Autism: A reappraisal of concepts and treatment. New York: Plenum. Suk-Hyang, L., Simpson, R. L., & Shogren, K. A. (2007). Effects and implications of self-management for students with autism: A meta-analysis. Focus on Autism and Other Developmental Disabilities, 22 (1), 2-13. Weiher, R. G. & Harman, R. E. (1975). The use of omission training to reduce self-injurious behavior in a retarded child. Behavior Therapy, 6, 261-268. Wolery, M., Bailey, D., & Sugai, G. M. (1998). Effective teaching principles and procedures of applied behavior analysis with exceptional students. Boston: Allyn & Bacon, Inc. Author Contact Information Vanessa Ann Silla -Zaleski, Ed.D., BCBA The University of Scranton Education Department 139 McGurrin Hall Jefferson Avenue Scranton, PA 18510 Phone: 570-941-5810 Fax: 570-941-5515 E-Mail: [email protected] Mary J. Vesloski, MA, BCBA Autism Behavioral Services, Inc. 620 Wyoming Avenue West Pittston, PA 18643 Phone: 570-655-1667/570-947-5363 Fax: 570-602-4100 E-Mail: [email protected] 87