Return to REFERENCES

Teaching Language to Children with Autism

(Mareile Koenig, Ph.D., CCC-SLP, CBA/Pa)

Associate Professor of Communicative Disorders at West Chester

University in West Chester, Pennsylvania

CT FEAT Newsletter, In This Issue:  Volume 2, Number 3

Editor’s Note: During the past year or so, there has been tremendous interest in the work of Mark Sundberg, Ph.D., and James Partington, Ph.D., behavior analysts who have developed innovative assessment and intervention strategies for teaching language skills to children with autism and related disorders. Some of their material is quite technical, and they frequently conduct workshops for training people to use the specialized instruments and techniques they have developed. (See Conference Listing on page 6 for two such workshops). In this book review, Mareile Koenig, Ph.D., provides a very extensive description of their work.

Since 1996, a number of publications have brought behavioral strategies for educating children with autism into clearer view of a larger audience (e.g., Maurice, Green & Luce, 1996; Sundberg & Partington, 1998; Leaf & McEachin, 1999). Each was written for parents and professionals, and each has undoubtedly improved the lives of many children in many ways. That in itself deserves a round of applause. Two of these manuals present frameworks for addressing target skills across several developmental domains (Maurice et al., 1996; Leaf & McEachin, 1999), while one is devoted exclusively to language (Sundberg and Partington, 1998). This review focuses on the latter, which is a "must read" for anyone serious about communicative enhancement.

Before I review the book’s contents, here’s some background: The current edition of Teaching Language To Children with Autism or Other Developmental Disabilities is actually the 7th version of a manuscript more than 25 years in the making, and it is based on research and experience accumulated during that time. It has also been influenced directly by some of the best minds in the field of Applied Behavior Analysis, including Dr. Jack Michael, Dr. Ernest Vargas, and Dr. Jerry Shook. For more information, visit Behavior Analysts, Inc..

Now for a review of the book’s content: Teaching Language to Children with Autism or Other Developmental Disabilities is based on the conceptual framework of Skinner’s (1957) Verbal Behavior. Its 315 pages include four divisions and an appendix. The first division (chapters 1-4) describes an assessment protocol for children at the earliest stages of language development. The protocol utilizes a combination of interview and direct observation to identify a child’s typical performance levels in several natural settings. Observations are focused on the following types of behaviors: 1) cooperating with adults, 2) requesting (manding), 3) motor imitation, 4) vocal play, 5) vocal imitation (echoic), 6) matching-to-sample, 7) receptive language, 8) labeling (tacting), 9) receptive identification by function, feature and class (RFFC), 10) conversations (intraverbal), 11) letters and numbers, and 12) social interaction.

A child’s performance is rated in each area relative to five behaviorally defined proficiency levels. A form is provided to document the ratings and to profile the child’s repertoire. Intervention goals are selected in consideration of the profile. Five model profiles are provided to illustrate the translation of performance patterns to IEP Objectives. Also included are guidelines for the selection of an appropriate communication modality (e.g., speech, sign language, picture system). While the long-range focus remains on speech for all children, augmentative systems are recommended as a supportive interim step for some.

The second division (chapters 5-9) describes specific procedures and a curriculum for establishing early language skills in nonverbal children. Specific procedures are described for establishing rapport, requiring a response, establishing positive alternatives to negative behavior, and teaching a child to request ("mand training"). Mand training is the first form of direct language instruction in this protocol, and it is reinforced as the program progresses. Specific procedures for mand training vary depending on a child’s assessment profile.

Detailed instructional sequences are provided for four types of learners: 1) children who are non-echoic and non-imitative; 2) children who demonstrate imitative skills and are candidates for sign language; 3) children who demonstrate echoic skills and are candidates for vocal mand training; and 4) children with physical impairments who are candidates for learning a pointing response. The manual also provides guidelines for the selection of first words as mands, for establishing a mand through multiple prompts, for fading each prompt, for thinning the reinforcement, for expanding the mand repertoire, and for supporting the use of mands in the natural environment. This systematic approach to early mand training distinguishes the Sundberg and Partington approach from some behavioral approaches (e.g., Maurice, Green & Luce, 1996; Leaf & McEachin, 1999) but not others (e.g., Bondy, 1996).

Top

A number of related language skills are targeted following initial mand training, including imitation, echoic responses, receptive discriminations, and matching-to-sample. Strategies are given to support the instruction of each related skill to learners with different profiles. Trials targeting these skills are interspersed systematically with mand trials, resulting in a "mixed verbal behavior" (aka "mixed VB") format. This is different from behavioral approaches that rely heavily on massed discrete trials during the early intervention phase.

Tact training and receptive identification of common objects are introduced concurrently following the successful acquisition of 5 to 10 mands and some success on related skills. Work in all areas continues and is interspersed with trials targeting tacts, thus continuing and expanding the mixed VB format. Again, different strategies for teaching each new skill are provided to accommodate learners using different communication modalities (e.g., speech, sign, picture exchange). Criteria are provided for the selection of specific words to expand the receptive and expressive repertoire. A sample vocabulary sequence for tact and receptive language training is provided to illustrate the selection criteria.

The identification of receptive (labels) by function, feature, and class (RFFC) is initiated following the acquisition of about 50 words in previous language training. Targets include only those items that the child can already tact and receptively discriminate. Specific examples are provided. Once the child can perform this skill with a large number of targets, the skill is brought into a mixed VB format.

The last set of targets included in the beginning language curriculum is intraverbal skills (e.g., word associations, word sequences, fill-in-the-blank items, etc.) These are introduced at about the same time as the RFFC targets. As with the other skill sets, the authors provide detailed behavioral descriptions of strategies for selecting and establishing initial intraverbal skills. Again, strategies vary depending on a child’s communication modality (speech, sign, pictures). Detailed shaping and fading procedures are also tailored to modality.

Division three (chapters 10-12) presents advanced language training as a continuation of the procedures and curriculum described in division two. The materials and procedures are considerably more complex than those involved in the first two levels. Here is a small sample of some of the skills targeted at this level: use of phrases, simple sentences, and more complex syntactic forms; use of language for describing experiences gained through different sensory modalities (e.g., touch, vision, hearing) and verbalization of private events (feelings); responses to a variety of Wh-questions; expansion of previously stated information; responses to a sequence of questions about a single topic; verbalization of event sequences; definition of words; narration of stories; description of current events; discussion of politics, and much more. The goal at this level is to develop fluency in the production and use of language adequate for a variety of social and academic functions.

Division four (Chapters 13-14) covers issues relevant to the implementation of a language program in a child’s home and in his/her school environment. Included here is a detailed comparison of discrete trial teaching (DTT) to natural environmental training (NET). When considered from within the conceptual framework of Skinner’s model, the authors indicate that each strategy supports a different aspect of verbal behavior. NET primarily benefits mand training while DTT primarily benefits tact, receptive, echoic, and imitating training. The importance of NET and DTT may also vary at different points in developmental time. NET may exert a greater influence during initial acquisition and peer social interaction while DTT may exert a greater influence on academic work.

Also covered in division four are characteristics of an effective educational program. Specific issues addressed under this umbrella are home-based ABA programs, intensive school-based programs, traditional special education classrooms, inclusion, procedures and checklist for evaluating potential programs. Included within the checklist is a consideration of staff philosophy, acknowledgement/use of basic behavior modification, augmentative communication, language analysis and teaching, the classroom environment, staff-student ratio, length of the instructional day, number of school days in the academic year, classroom’s daily schedule, physical layout of the classroom, parent education and involvement, skill acquisition data, and disruptive student behavior.

At the back of the book, there is an appendix providing important background information about Skinner’s (1957) model of Verbal Behavior, as well as information from more recent publications related to this model.

Top

My personal reactions to this book are varied, with a net rating that is very positive. In their description of the program, I very much appreciated the authors’ care in tying aspects of the structure and process to the theoretical framework. This enabled me to develop a deeper understanding of the protocol. While operational definitions are expected in the world of ABA, I nevertheless appreciated the clear behavioral descriptions given for procedures that are often simply assumed (sometimes erroneously) to exist in the clinical repertoires of teachers and clinicians.

For example, there is an excellent description of the procedures for (and necessity of) establishing rapport with a child prior to the initiation of any assessment procedure. I also appreciated the discussions about alternative communication systems, DTT versus NET, and about guidelines for evaluating the appropriateness of an educational program. I assigned parts of this book as required reading for a graduate course that I taught last summer, because I think it’s the most detailed illustration of a behavioral language intervention protocol available and because evidence indicates that behavioral approaches are the most effective primary intervention for children on the autism spectrum.

Would I recommend this manual to the parent of a newly diagnosed child with autism? Probably not just yet, even though the child might very well be able to benefit from the program it describes. Although there is a great deal of redundancy across the pages, the jargon (while justified) is not necessarily reader friendly. Also, some of the information was difficult to assimilate from this text alone. The authors made frequent reference to their companion publications (the ABLLS Assessment, Curriculum Guide, and Skills Tracking System for Children with Autism and Other Developmental Disabilities and The ABLLS Scoring Instructions and IEP Development Guide) for more detailed information and more specific guidance regarding the assessment tool and the curriculum. These additional sources do help to clarify the picture.

Finally, while one of the hats that I wear is behavioral, the other hat is that of a speech-language pathologist. In that role, I have additional reactions. First, some of the statements in Teaching Language to Children with Autism or Other Developmental Disabilities appeared simplistic. For example, in chapter 2 the authors state "the primary purpose of a language assessment should be to identify specific verbal deficits, and to serve as a guide for the development of an appropriate language intervention program for the individual being assessed."

In my opinion, the primary purpose of a language assessment is to answer the referral question, and the choice of an assessment protocol should be based on whether that protocol results in data that answers the question. When the information that is being requested is information about where and how to begin therapy, the Sundberg and Partington protocol will be a great asset. When information about developmental levels is sought, other approaches may be superior. This example, by itself, involves a simplification that is tangential to the overall focus of the book, but I offer it as an example of simplification patterns that occurred frequently in some parts of the book.

My second reaction as an SLP is that I do not understand why, during graduate school, I was not exposed to the Sundberg and Partington literature even once. Similarly, I do not understand the omission of relevant references from the SLP literature in Sundberg and Partington’s reference list. For example, Wetherby and Prizant (1996) developed a behavioral alternative to standardized assessments for the very same reasons that Sundberg and Partington did. Why wouldn’t that be mentioned? Wouldn’t it corroborate Sundberg and Partington’s reasoning? Where is the gremlin that makes professionals in related fields blind to each other? For the sake of children who would benefit from interdisciplinary collaboration, let’s communicate! Thanks to Connecticut FEAT for opening the doors.

Top

Return to REFERENCES