Janet Norris is a professor of speech-language pathology in the Department of Communication Sciences and Disorders at Louisiana State University. She has received ASHA's Clinical Achievement Award twice and has published over 30 articles in the area of language disorders. She is the co-author of two books on language intervention, as well as several book chapters. She is an ASHA Fellow.

Lisa Scott Trautman is an assistant professor of speech-langauge pathology in the Department of Communication Disorders and Sciences at Wichita State University. Her dissertation research investigated the effects of contextualization on fluency in language-disordered, fluency-disordered, and normally-fluent children. She also has presented and published a series of studies on the relationship between stuttering and language as a dynamic interactive process.

Mike Susca is a doctoral student at the University of Nebraska-Lincoln. Prior to enrolling in doctoral studies, he was a practicing speech-language pathologist for the past 21 years in San Diego, California. He has considerable clinical experience with clients of all ages in multiple settings and multiple communication disorders. He is a fluency specialist and will focus his Ph.D. program in the area of fluency disorders.

For decades, clinicians have recognized that stuttering is related to a number of factors (i.e., physiological, genetic, psychosocial, environmental, and linguistic) which contribute to the onset, development and maintenance of the disorder. Recognizing the need to address these multiple factors in treating stuttering, several integrated treatment programs have been described in the literature (Guitar, 1998). For example, Cooper and Cooper (1985) developed an integrated approach that addresses the affective (feelings), behavioral (speech changes) and cognitive (thoughts) components of the disorder (i.e., the ABC's of stuttering). Guitar (1998) describes a number of integrated approaches, including his own, which focus on creating speech changes along with a reduction in or elimination of negative feelings, emotions and avoidance behaviors. Because of the multidimensional nature of stuttering in older school-age children and adults, it is easy to understand why many clinicians treat stuttering from an integrated perspective rather than relying on fluency shaping or stuttering modification procedures exclusively. Starkweather and Givens-Ackerman (1997) point out that an integrated approach to treating stuttering is not only popular among practicing clinicians, but has been adopted by the American Speech-Language-Hearing Association's guidelines for practice in stuttering treatment (Starkweather et al., 1995).

Generally, an integrated treatment philosophy involves teaching individuals with intermediate and advanced stuttering a combination of fluency skills and stuttering modification procedures as well as how to modify negative feelings and attitudes about stuttering. Usually, the first stage of therapy might address increased awareness as well as exploration and desensitization to stuttering. The next phase involves teaching the child or adult to stutter easily and speak more fluently in various ways through the use of progressively longer and more complex linguistic units (i.e., single words, sentences, phrases, paragraph reading and conversation). Following this training, a specific portion of therapy might be spent on reducing negative emotions and attitudes as well as reducing avoidance behaviors. The final phase of an integrated program usually involves assisting the client to maintain the skills that have been learned, generalize speech changes to realistic speaking situations, and increase self-monitoring of performance.

We believe the overall structure of most current integrated treatment programs are effective for both children and adults who stutter. Clinicians who use this approach are not concerned that a client's speechafter therapy may contain some stuttering, contending that normally-fluent speakers occasionally produce dysfluencies (Starkweather & Givens-Ackerman). However, improvements in stuttering along with the client's improved self-perceptions, nonavoidance, and reduction of fears are typical target goals of most integrated approaches.

Although integrated approaches address a number of factors related to stuttering, we believe the structure of most integrated treatment programs, like the one described above, limits the true interaction among the factors that maintain stuttering. It appears that many treatment integrated programs address each factor (i.e., changes in speech, reduction of negative emotions, cognitive restructuring, etc.) as a series of isolated, unidimensional activities. For example, considerable time in treatment might be spent teaching clients to reduce speech rate and/or use voluntary stuttering as a means of improving fluency. Cognitive, emotional, and social factors also might be addressed during this phase of treatment but tangentially. The main focus remains on creating speech changes (i.e., modifications of motor skills) and once improved speech performance is achieved, another factor will be addressed.

Unfortunately, we believe that following this approach will make it difficult to change the dynamic interactions among all factors maintaining the stuttering. For instance, changes in motor skills might facilitate greater fluency, but those skills will be difficult to maintain unless simultaneous changes take place in the client's cognitive, emotional, and linguistic capacities. Additionally, the influence of behaviorism has led clinicians to rely on gradually increasing the length and complexity of the client's response (i.e., using repeating word lists, short sentences or isolated phrases) which might create artificial communicative interactions during treatment. We would propose that providing realistic, meaningful communicative interactions between client and clinician as a basis for managing the interaction among factors associated with stuttering could be a more efficient way to address a number of factors simultaneously in treatment.

The purpose of this paper is to propose a few enhancements in the methods used in implementing an integrative treatment approach to stuttering. First, we will provide a brief overview of a new perspective on stuttering offered by Smith and Kelly (1997) which serves as a theoretical basis for our recommended enhancements in an integrated approach to therapy. Second, we will discuss how the Situational-Discourse-Semantic (SDS) model of language by Norris and Hoffman (1993) supports our recommended enhancements to an integrated approach.

A New Theoretical Perspective on Stuttering

Recently, Anne Smith and Ellen Kelly of Purdue University have developed a new, multi-dimensional perspective on stuttering. Briefly, Smith (in press) and Smith & Kelly (1997) argue that too much attention has focused on the "surface structure" of stuttering or the observed stuttered events (e.g., number of stuttered moments, the onset and offset of stuttering, tremors that occur during a moment of stuttering, etc.) rather than the complex interaction of factors underlying stuttering behavior. Smith (in press) points out how scientists once attempted to classify volcanoes by the shape of the landform and the type/shape of the eruptive materials. A dramatic shift in understanding volcano activity came in the 1960's when scientists discovered volcanoes resulted from dynamic plate movement in the Earth's surface.

Smith's volcano analogy helps us understand how the field of stuttering has attempted to understand stuttering by only examining the observed stuttering behavior. According to Smith, researchers and clinicians in stuttering have paid too much attention to the superficial, disfluent moment in researching and treating the disorder focus on isolated stuttered moments fails to take into account the underlying dynamic interaction of all of the factors "below the surface of the stutter". A dynamic, multifactorial approach to stuttering suggests that variations in interactions among factors contribute to the variability of stuttering that occurs across several different discourse contexts (reading, expository discourse, narrative discourse) and with different types of communicative partners (parents, peers, co-workers). Thus, a stuttered moment is viewed as a variable phenomenon that results from a complex and dynamic interaction among multiple factors associated with various speech contexts and variety of listeners.

Proposed Enhancements To An Integrated Approach

We believe Smith and Kelly's (1997) multidimensional perspective on stuttering and the whole language principles underlying Norris and Hoffman's (1993) SDS model, can serve as a foundation for enhancements to the implementation of an integrated approach for treating stuttering. We propose two key enhancements. They include: a) utilizing speech tasks during therapy that involve thematic, topic-centered contexts of interest to the client; and b) having client-clinician interactions follow a continuum of high- to low-structured interactions within a session and throughout the program. Structured interactions through scaffolding can assist clients in adjusting message formulation, cognitive changes, emotional responses, listener needs and sensori-motor restructuring for a given social speaking situation. In order to understand how these enhancements could be incorporated into a treatment program, we will discuss each of them in greater detail.

Using Thematic, Topic-Centered Speech Contexts

In our opinion, the use of thematic, topic-centered speech contexts in stuttering therapy is an efficient way of creating realistic client-clinician interactions when treating language disorders in children and improving the literacy skills of adults (Calvin & Root, 1987; Lerche, 1985; Norris, 1997, Radencich, 1994). Use of topic-centered speech contexts are in direct contrast to the use of single word lists, carrier phrases, short sentences or any type of isolated stimulus materials that have minimal social purpose and consequences. Using words lists/unrelated phrases can result in verbal exchanges which are contrived, non-meaningful speaker-listener interactions. By contrast, using thematic, topic-centered speech contexts throughout the program will make interactions more socially appropriate and meaningful. Moreover, we believe that finding a topic relevant to clients' interests will increase their motivation for therapy and make the treatment program more enjoyable. For example, adults may choose to discuss a favorite sports, hobbies, or aspects of their profession. One of our adult clients chose to explore information about stuttering theories. A child might want to talk about their pet dog and learn more about the breed of his/her dog and about dogs in general. A school-age child we treated chose rocks as a topic.

Any topic can form the foundation for meaningful dialogue during the session as long as the client and clinician jointly construct knowledge about the topic during each session and expand the topic across the treatment program. All of the goals and objectives of the therapy program are learned, practiced, and generalized. Another advantage of using thematic topics is that other communicative partners (spouses, friends, parents) can be involved in the ongoing discussion of the topic which may act as a bridge to generalize strategies and techniques beyond the clinic. Finally, we believe that selecting thematic topics as the basis for communicative interactions allows the client to devote greater attention to specific cognitive, linguistic, motor, and emotional responses because of his/her familiarity with the topic.

Structuring Communicative Interactions Through Scaffolding

Once a topic has been selected, the clinician can manage the within-clinic interaction by systematically scaffolding responses from the client. Scaffolds are cues, prompts, or any type of physical, visual, auditory support which assists the client in becoming an active rather than passive participant in an interaction. Scaffolds such as real objects, replicas, pictures, reading materials or semantic maps (i.e., contextualized materials) can be used to minimize processing demands and create structured communicative interactions. The amount of text and picture support can be varied to increase or decrease cognitive-linguistic processing demands. For example, a clinician might want to engage in choral reading with the client in order to establish a speech rate and phrasing pattern that is conducive to fluency. A delayed auditory feedback unit (a motor speech scaffold) might be used to scaffold a client's increased awareness of movements associated with talking in a more fluent manner.

The SDS model can serve as a guide for how the clinician can systematically manipulate the degree of contextualization and scaffolding as they interact with the semantic complexity and discourse level of the client's response. This can minimize the client's linguistic processing and motor speech demands since the materials present help determine word selection and utterance formulation. The materials also can help determine the complexity of the client's speech acts by having the client provide labeling or simple descriptions versus interpretation or inferences (i.e., increased semantic complexity) about the materials.

Other scaffolding techniques might include having the clinician paraphrase or summarize information that is about to be read by the client so the linguistic processing demands are minimized. The clinician can also structure the amount of information to be read by chunking the information into meaningful units (Norris, 1997). Finally, verbal cues such as binary choice, cloze procedure, expansion, and summarization can be used to limit or expand the information provided to the listener during an interaction.

As the client learns how to coordinate the cognitive, linguistic, emotional, motor, and social aspects of structured interactions through various forms of scaffolding, the demands on the client can be increased. This would involve a gradual shift from contextualized to decontextualized (removal of materials and minimal scaffolding) interactions such as spontaneous discussions and story telling. The clinician might also expect the client to increase the number of speaking turns, produce more complex discourse, speak to other listeners and/or reduce the use of fluency enhancing or stuttering modification techniques. Additionally, improvements in the client's cognitive, linguistic, and emotional capacities could be by including another listener as part of the treatment program.

Conclusion

The popularity of integrated approaches for the treatment of stuttering suggests that clinicians see the need to address the multiple factors which influence stuttering. However, if each factor is treated separately and linearly, then the approach will limit the interaction that takes places among factors. Change in one factor creates changes in all other factors. Dynamic, multifactorial models provide a framework for meaningful, purposeful speech acts as the basis for client-clinician interactions. With the inclusion of thematic, topic-centered speech contexts, strategies are learned and processes are acquired as they exist in realistic speaking situations. We also suggest that the systematic use and gradual removal of various types of scaffolds can assist a client in formulating a message about a topic of interest.

We have attempted to show how the enhancements described in this paper offer clinicians alternative methods for how to address the multidimensional factors "below the surface" of stuttering. We propose that simultaneously attending to these multiple factors enables learning to occur in balance. Thus, as fluency skills are improved, those skills become integrated into the speaker's cognitive-linguistic-emotional-motor-social speech system rather than an artifact that must be generalized--a step in therapy that many older children and adults find difficult to achieve.