Augmentative and Alternative Communication Systems

• The client and family have the greatest knowledge of the daily communication needs of the person with CCN. Family members are often the individuals’ primary communication partners and serve as advocates and facilitators.

• The speech-language pathologist has the greatest general understanding of communication in general and can assess language, and communication needs, abilities and skills; select AAC materials and technologies; and teach the individual, family, and staff to use AAC system components effectively.

• The teacher sets educational goals and oversees classroom implementation of each child’s AAC system and has knowledge of literacy, social interaction, and education.

• The physical therapist (PT) or occupational therapist (OT) carries out the motor evaluation, addresses seating and positioning, evaluates physical access to the AAC system, and has knowledge of how to support writing, drawing, and other activities of daily living.

• The teacher’s aide/job coach is also critical to the success of implementation. This individual supports the person in the school/work setting. Key team members are often referred to as “natural supports” because they have a continuing relationship with the individual (e.g., family, friends, coworkers, and employer). On occasion, physicians, psychologists, vision specialists, and other professionals also play an important role on AAC teams.

My new friend (Wendy) was good looking. She was just over five feet tall and had brown eyes that matched the color of her shoulder length hair. Her skin showed a summer tan and she had a dynamite smile. “Did he show ya all his electronic stuff?” one of my dorm mates asked her. “Go on, Bill, show her.” So I demonstrated the controls for my lights and clock radio. I showed off my door opener, which I could control via a radio transmitter attached to the Plexiglas tray on my wheelchair. She was impressed with the space-age technology. “Hey, show her your wheelchair and how it works. I’ll never understand how it works. It baffles me,” another dorm mate said. So, wondering if I should sell tickets, I wheeled about the room. I demonstrated how I went straight, reverse, and turned left and right. I was angry at my dormmates because I was a man, not a side show freak. My wheelchair was a tool for my mobility, not a novelty. Why couldn’t they see that? And why couldn’t they see that I was trying to get to know Wendy. Why didn’t they understand I had a right to my privacy just as they did? As I was wheeling around the room, I noticed that Wendy was typing something. I was disappointed in her. I thought she knew that I could hear and that she didn’t have to write things to me. Apparently I was wrong. When I was done showing my electric marvels to her and the guys, I rolled back to my typewriter to read, “I wish they would go, so we could talk by ourselves.” They finally left and we finally got to talk. Our friendship had started. (p. 137)

Traumatic Brain Injury.

TBI can result in the loss of speech and often causes physical, cognitive, and language impairments (Beukelman and Garrett, 1988; Light, Beesley, and Collier, 1988). Although the long-term recovery of speech is variable, immediately after the injury many individuals benefit from the use of AAC interventions to support functional communication. Associated changes in motor, perceptual, cognitive, and language abilities also may have detrimental effects on communication (Carlisle Ladtkow and Culp, 1992). Thus, it is critical that the limitations associated with a severe brain injury from a car accident, gun shot wound, explosion, and so forth be identified and considered when AAC goals and interventions are determined (Beukelman and Yorkston, 1989). In a follow-up study of nonspeaking individuals with TBI 1 year after discharge, DeRuyter and Kennedy (1991) found that only 56% of those for whom an AAC device was recommended were using the device, 24% had completely discarded the device, and 20% only used it in limited environments (DeRuyter and Kennedy, 1991). In many cases, speech had returned. However, there was no indication as to why 44% of the devices were abandoned or not being used to their full potential. Beukelman and Yorkston (1989) point out that there is a need for AAC devices to be integrated and adapted into the person’s living situation, outside the rehabilitation center. Frager, Hux, and Beukelman (2005) found that a group of communication partners and the continuing support of an AAC facilitator contribute to success. In a recent study, Frager, Hux, and Beukelman (2005) found that high-tech devices were favored over low-tech systems and that low-tech systems were apt to be used temporarily by people with TBI who regained speech.

Aphasia.

Persons who sustain CVAs often have language difficulties that we collectively call aphasia. One lasting problem these individuals have is vocabulary retrieval or word-finding difficulties. There are several AAC-related approaches with potential for aphasia rehabilitation (Jacobs et al, 2004; Kraat, 1990). For example, individuals who can recall first letters and recognize a desired word from a list may use word prediction devices/software. The individual begins typing a letter and then the device predicts several words from which to choose (see Figure 7-8). Colby et al (1981) developed a microcomputer-driven device that used a specially designed database containing words, their frequency of use, and features of each word, which was specifically designed for persons with aphasia. Features included words that “go with” the desired word. This can be a sound-alike relationship; a semantic (meaning) relationship; a categorization (e.g., a piece of furniture or a fruit); and initial, middle, and ending letters. Each was shown to be effective for persons with certain types of aphasia. A similar approach was developed by Hunnicutt (1989).

Many factors must be considered when AAC is applied in aphasia rehabilitation (Kraat, 1990). Some people with severe aphasia learn to augment their speech and communication efforts by relying on gestures and an alternative symbol system (Jacobs et al, 2004). However, although persons with aphasia may be able to use graphic symbols, many find it difficult to apply them socially or to generalize their use. One commercial device designed specifically for persons with aphasia is the Lingraphica, which organizes symbols by semantic categories (e.g., places, foods, clothing) and includes synthetic speech output and animation of verbs (Steele and Weinrich, 1986). Approaches such as CHAT and TALK that model conversational flow and provide clues to word vocabulary choices based on context can also assist aphasic individuals (Kraat, 1990).

A recent development is the use of visual scene displays (VSDs) (McKelvey et al, 2007). VSDs use personalized photos of scenes and arrange these on a dynamic display device. The technology enables individuals with severe aphasia to use familiar photographs to engage partners in interactions about multiple topics. In addition, the design of the technology makes it relatively easy for partners to provide conversational supports. Because of the dynamic nature of the display, the user is continually prompted, reducing the individual’s need to rely on recall memory. The potential for people with aphasia to benefit from the use of VSD technology is still under investigation.

The use of AAC in aphasia rehabilitation often involves the use of AAC strategies and partner training. Blackstone (1991) identified issues involved in aphasia rehabilitation that included a discussion of public policy issues (e.g., funding for assessment and therapy services) and a discussion of clinical studies involving AAC strategies and technologies. King and Hux (1995) describe the use of a talking word processor to increase writing accuracy for individuals with aphasia. The speech feedback provided additional monitoring of the written work, which helps the person to identify and correct errors. Garrett and Beukelman (1992) present a classification system for aphasic individuals that is useful in planning AAC interventions. This scheme guides intervention planning by describing five types of aphasic communicators: basic choice, controlled situation, augmented, comprehensive, and specific need. For each of these categories, the authors identify residual skills, intervention goals, and AAC skills and suggest AAC activities for both partners and the individual with aphasia. Finally, Fox and Fried-Oken (1996) propose questions related to effectiveness, efficiency, and generalization in AAC aphasiology research. One area of concern for future development in AAC is how best to represent meaning on AAC technologies for persons with aphasia (Beukelman and Ball, 2002).

Assessing Barriers to Participation.

In the Participation Model, opportunity and access barriers to successful AAC use are identified. Beukelman and Mirenda (2005) present detailed information regarding the implementation of the participation model, including sample assessment forms and case examples. For example, opportunity barriers are those that involve policies, practices, attitudes and knowledge, and skills of those who support the person with CCN and that interfere with successful AAC interventions. As an illustration, consider the situation where a school district purchases an SGD for a child, but the child is required to leave it at school at all times. This practice is a barrier to full societal participation and academic success. Keeping the device at school may be a policy of the district because uninformed administrators worry about the cost of the device and possible breakage or loss if it goes home. After all, schools allow students to take home band instruments, uniforms, books, and pencils. Another example of an opportunity barrier is the employer who is resistant to a worker using an AAC device. This may reflect the employer’s attitudes about disability or a lack of knowledge about AAC or a lack of skill in supporting individuals with CCN.

Another key element of the Participation Model is an “activity standards inventory” in which desired communication-related activities of the person with CCN (termed “target person”) are listed. The standard of desired performance is that a nondisabled peer carry out the same activity. The target person is then rated as to the level of participation (independent, independent with setup, verbal or physical assistance, or unable to participate), and the discrepancy between peer and target person (if any) are ascribed to “opportunity” or “access” barriers. These are evaluated in terms of potential needs: (1) to increase natural abilities, (2) to make environmental adaptations, and (3) to use AAC systems and/or devices. Finally, AAC potential is determined through an operational profile, a constraints profile, and a capability profile.

As described in Chapter 7, the human-technology interface evaluation is part of a capability profile. An important component of the capability assessment involves documenting the individual’s speech, language, motor, sensory, cognitive, and social communication skills. Dowden (1997) describes assessment approaches for individuals with CCN who have some functional speech. There are many language tests for both children and adults. Cognitive assessments help to determine how the individual understands the world and how communication can be best facilitated within this understanding (Beukelman and Mirenda, 2005). There are no formal tests that accurately predict the ability of an individual to meet the cognitive requirements of various AAC techniques and technologies, and expressive language by use of AAC is itself required to accurately assess cognitive ability. Thus, the individual’s cognitive ability is often estimated. Some cognitive skills that are important for AAC are shown in Box 11-2. Social communicative skills (e.g., degree of interaction, attention to task) are generally assessed by interviews with family, caregivers, teachers, and others and through observation during an assessment or during opportunities created specifically to encourage social interaction. One example of the information that may be required (and how it may be assessed) is delineated in the Medicare funding request for SGDs in the United States (shown in the first column of Table 11-2).

Assessing Representation.

One area of assessment unique to the area of AAC is determining what types of symbols an individual can use to communicate. A variety of symbol types are shown in Figure 11-5. Clinicians may select from several assessment protocols (Beukelman and Mirenda, 2005). In one protocol (functional object use), the evaluator shows the person a symbol and says “Show me what you do with this.” The response may be a gesture (e.g., hand to mouth for “eat it”) or pointing at a picture or symbol (e.g., “drink” if prompt is a soft drink can). In another approach (visual matching) the evaluator asks the individual to find a single stimulus item from a multiple symbol array or vice versa. The most flexible AAC systems are those that depend on spelling and require literacy skills, so word recognition and reading comprehension are often used to assess reading level, and spelling evaluations also may be conducted (Beukelman and Mirenda, 2005). There are multiple levels of spelling skills that may be useful in AAC:

Many types of symbol systems are used in augmentative communication (Lloyd, Fuller, and Arvidson, 1997; Vanderheiden and Lloyd, 1986). Several of these are illustrated in Figure 11-5. Perhaps the most concrete type of symbol is the use of real objects (full size or miniature). However, to a person with cognitive disabilities, a miniature object may not appear to represent the full-size version, and care must be taken to ensure that the association is made by the user (Vanderheiden and Lloyd, 1986). Real objects and photographs have the disadvantage that many communicative concepts (e.g., good, more, go, hurt) are difficult to portray. Pictographic symbols include provisions for more abstract communicative intents and allow much greater flexibility in developing vocabulary usage. A more flexible symbol type is the use of a symbol system possessing grammar and syntax (e.g., Blissymbols). The nature of this symbol system allows the inclusion of more linguistic functions, such as categorization by parts of language. Traditional orthography is the symbolic representation based on letters and words. Some individuals have reading skills that exceed their spelling skills, and they cannot rely on spelling for communication. If the person has a large word recognition vocabulary, the selection set should be based on words with possible “carrier phrases” that are filled in with limited spelling (e.g., “I would like a drink of ____”). Spelling is the most flexible symbol system because it can be used to create a large number of different utterances, but it can also be the slowest because of letter-by-letter entry rather than selection of whole words. Many computers, printers, and keyboards accommodate languages other than English.

Human Technology Interface.

The human technology or control interface for SGDs is the hardware by which the person with CCN accesses the low- or high-tech device (see Figure 2-6 and Chapter 7). The most commonly used control interfaces for augmentative communication devices are keyboards, single or dual switches, joysticks or multiple-switch arrays, and mouse or alternative pointing interfaces. SGDs; other AAC approaches use either direct selection or indirect selection (e.g., scanning, directed scanning, or coded access). These are discussed in Chapter 7. Most selection sets use visible symbols (e.g., letters, graphics, pictures) so individuals who have visual impairments and physical limitations requiring scanning may not be able to use visual arrays. For these individuals auditory scanning is used. Choices are presented in auditory form by a partner or an AAC and the user selects his or her choice from the auditory prompts. In some cases, both a prompting phrase and a selected auditory utterance are included and the user hears the prompting phrase through an earphone. In nonelectronic voice auditory scanning a list of vocabulary items is read aloud by the communication partner. The AAC user then chooses a vocabulary item by using a predetermined signal such as a vocalization to identify the desired vocabulary item. Kovach and Kenyon (1998) analyze a variety of approaches to auditory scanning, summarize current research in this area, and describe considerations to be included when developing an auditory scanning system for an AAC user.

Examples of Vocabulary Retrieval Techniques.

Many SGDs use the approaches to increase input rate that are discussed in Chapter 7 (abbreviation expansion, word prediction, word completion). In addition, there are several methods for storing and retrieving vocabulary that are designed specifically for SGDs.

Instant phrases are those used frequently for greetings, conversational repairs (e.g., “that’s not what I meant”) or similar actions; these are often included as single keystroke entries in an “activity row” or in a row of the scanning matrix, near the beginning of the scan. They can also serve as “floor holders” (e.g., “please wait while I type my question/answer.)

Coding of words, sentences, and phrases on the basis of their meanings and also known as semantic encoding or Minspeak (Baker, 1982). This approach uses pictorial representations that can have multiple meanings as codes, making recall easier. For example, when a picture of an apple is used for “food” and a sun rising for “morning.” then selection of “apple” and “sunrise” could be a code for “What’s for breakfast.” Icons can have multiple meanings. Thus the apple symbol can take on the meaning of “eat” or “red” or “fruit” rather than food. Several examples of Minspeak sequences are shown in Figure 11-8. Baker (1986) also developed an approach based on the use of syntactical labels coupled with icons. Figure 11-9 illustrates this concept. For example, the apple icon becomes “eat” when combined with the key labeled “verb” and becomes “food” when combined with the noun key. Unity is a family of Minspeak application programs included with Prentke Romich (Wooster, Ohio, www.prentrom.com) AAC devices. It includes 4, 8, 15, 32, 45, 84, and 128 location overlays that differ in the pointing resolution required by the user. Sequences of icons and their locations on the keyboard are kept as consistent as possible among the overlays to account for motor skill development while allowing growth in language usage. Versions of Unity vary from a few hundred words to more than 4000 words intended to address the core vocabulary that is responsible for the majority of conversational utterances. When large numbers of sentences, words, and phrases are stored, the icon sequences can become difficult to remember. Icon prediction initially lights an indicator associated with each symbol that forms the beginning of an icon sequence. When one of these icons is selected, only those icons that are part of a sequence light up or flash, beginning with the first selected icon. This continues until a complete icon sequence has been selected. This feature can aid recall and increase speed of selection because the device limits the number of icons that must be visually scanned for each selection.

Williams (1991), an accomplished user of numerical abbreviation expansion and word-based Minspeak, describes several advantages of this approach. In comparison to sentence-based Minspeak, he states that he (and most of the rest of us) does not think in sentences but in words or short phrases. This makes a word-based device easier to use. Second, he indicates that of the three encoding approaches in which he has achieved skill (each after hundreds of hours of practice), the word-based Minspeak “offers powerful advantages over the rest” (p. 133). His major reasons for this are the ease with which words are recalled during use and the large vocabularies that are possible with the use of icons rather than arbitrary codes. Williams also points out that it requires a large amount of practice and effort to become proficient with this type of device, which must be built into training programs. Williams also addresses the initial reluctance that many cognitively able but physically limited adults with CCN have to using pictorial representations as codes.

Examples of Vocabulary Programs for Language Development.

The Gateway (Dynavox Systems, Inc, Pittsburgh, Pa., www.dynavoxsystems.com) series is an approach to vocabulary organizations that is based on language development in typically developing children. The levels of Gateway are designated by the number of elements in the selection set, from 12 through 75. These are intended for six distinct target user groups beginning with the 12- to 24-month language development level, progressing to two formats for mild/moderate cognitive disability for children or adults, arrays for children and adolescents/adults with typical cognitive/language development and physical limitations, and a high-end array for augmented communicators who have well-developed syntactical skills. Pop-up menus with frequently used items (word, phrases, or sentences) are available on the larger arrays.

WordPower (Inman Innovations, available on several commercial AAC systems) combines a core vocabulary of 100 words that represent about 50% of spoken communication. It includes approximately 100 single hit words, hundreds of two and three hit words, a core dictionary for word prediction of 30,000 words, automatic grammatical endings (-ed, -ing, -s), and a QWERTY keyboard for spelling. For literate users, this approach is intuitive and leads to efficient communication. There are both direct and indirect (scanning) versions available. Picture WordPower uses labeled symbols as word cues. The same basic core vocabulary is available.

Augmentative and Alternative Communication System Outputs.

Dynamic communication displays create greater flexibility in selection sets by changing the selection set displayed when a choice is made, as shown in Figure 11-11. For example, a general selection set may consist of categories such as work, home, food, clothing, greetings, or similar classifications. If one of these is chosen, either by touching the display surface directly or by scanning, then a new selection set is displayed. For example, a variety of food-related items and activities (eat, drink, ice cream, pasta, etc.) would follow the choice of “foods” from the general selection set. The symbols on the display can be varied, and this changes the targets for the user. Because each new selection set is displayed, the user does not have to remember what is on each level. This approach, illustrated in Figure 11-12, also avoids having to squeeze several pictures into one square on a static display. It also relies on recognition memory rather than recall for identification of the selection set elements, which can make it easier to use. A dynamic passive display requires the user to select the next page to be displayed. The dynamic active display automatically branches to the selected new page once the item is selected.

Two types of dynamic displays were used in a matching task in a case study of a 16-year-old girl with a severe cognitive disability who had several years of experience in using fixed and dynamic displays (Reichle et al, 2000). There were no differences in accuracy of the matching tasks between the three types of display for a small number (15) of symbols. However, as the number of symbols increased (to 30 in dynamic and 60 in passive) the dynamic active display was significantly better than the other two. The response time was fastest for the passive display because all possible choices were displayed at once.

Blackstone (1994) describes a number of key features of dynamic displays. The nature of these devices allows the user to quickly change the screen and to configure the size, color, and arrangement of the symbols to match the topic. Dynamic displays reduce memory requirements because the user is prompted by the display after each choice. The constant vigilance to the screen requires a high level of visual attention and constant decision making. The user must also have mastered the concept of object permanence (Chapter 3). These may be challenging for some individuals who have cognitive limitations.

Visual scene displays (VSDs) take advantage of the graphical user interface (see Chapter 7) to create displays that capture events in a person’s life on the screen with “hot spots” that can be accessed to retrieve information (Blackstone, 2004). They also offer the AAC user and his or her partner a greater degree of contextual information to support interaction. The richness of the display and the information content also enables communication partners to participate more actively in a conversation. VSDs may represent either a generic or personalized context. The former includes drawings of places (e.g., house, schoolroom), whereas the latter is specific to user (e.g., a picture of his house, a family outing). Table 11-3 illustrates the difference between a traditional AAC display (referred to as a grid) and a VSD (Blackstone, 2004). The traditional grid supports communication of needs and wants and information exchange well. This type of display is usually restricted to symbols, text, or static drawings (although some animation is used with dynamic display items) and the vocabulary items are separated from any context to maximize their versatility. Figure 11-13 illustrates the differences between a typical grid display and a VSD (Blackstone, 2004). Personalization is also limited. The VSD is developed for conversational support as a shared activity. Because it uses a range of information media, including video and family pictures in addition to text, symbols and line drawings, it can be highly personalized, as shown in Figure 11-14 (Blackstone, 2004). In addition to communication of needs, wants and information exchange, VSDs also support social closeness. Because of the dynamic nature of the VSD approach, it can also serve as a learning environment. VSDs can stimulate conversation between interactants in which they play, share experiences, and tell stories. The dynamic nature of VSDs facilitates active participation of interactants during these shared activities. VSDs can also provide instruction, specific information, or prompts to help the user interact effectively. The populations expected to be served by VSDs are those with cognitive (e.g., Down syndrome) or language (e.g., aphasia, autism) limitations.

Young (2½ years old) typically developing children did significantly better at a birthday party communication task when using a schematic VSD layout (based on activities) than when using a grid layout (schematic or taxonomic) (Drager, 2003). One explanation is that the provision of a more meaningful context in the VSD reduced the language demand on the child. The VSD was organized around scenes of rooms: living room (arrival of children for party), kitchen (eating cake), family room (opening presents), and playroom (playing games); this reduces the demands on the child’s working memory because the location of the item required fewer demands on the VSD. The grid was organized around the activities, which required more language processing by the child (e.g., categorizing, remembering the symbols). For example, the topic of play could be illustrated by a digital photograph of the child’s room including the toy box in the VSD and as a symbol for play on the grid. Clicking on the hot spot associated with the toy box in the VSD or on the grid element for play resulted in branching in both formats to more detailed information.

Access to Mainstream Technologies.

Specialized assistive technologies such as EADL (Chapter 14) and power wheelchairs (Chapter 12) are also often of use to individuals with CCN. Many SGDs either provide the functions of EADL or interface with them through wireless connections. The use of an SGD interface to control a power wheelchair is a common application as well. The real power in connecting people with CCN to the rest of the information society lies in access to mainstream technologies. This includes entertainment (e.g., DVD, CD players) and other electronic devices such as electronic games. However, two technologies with enormous potential to create greater connectivity for people with CCN are the Internet and the cell phone.

Configurations of Commercial Speech-Generating Devices.

To describe current SGDs, we have created seven categories of the major commercially available devices, shown in Table 11-4. The categories reflect different groupings of the characteristics discussed earlier in the section on AAC characteristics as well as the funding codes and categories for Medicare reimbursement of SGDs in the United States (Blackstone, 2001). Table 11-4 also includes accessories and mounting systems for AAC devices. We have opted for a few large categories on the basis of the most essential features, resulting in variability within each category. The format in Table 11-4 appropriately groups devices serving distinct populations. Within each category there is still significant opportunity for decision making that is based on a thorough assessment of skills and needs. The commercial devices listed in each category are examples and, although a variety of manufacturers, models, products, and varying device features are included, Table 11-4 is not inclusive.

Simple scanners, the first category in Table 11-4, are generally operated by a single switch, although some can have dual-switch scanning and others allow four- or five-switch directed scanning. The devices in this category are distinguished by the use of a light to indicate the output selection, very limited vocabularies (32 items or less), no rate enhancement or vocabulary, and the general absence of voice output as a standard feature.

The devices categorized as simple speech output are further delineated by length of recorded digital speech. They were all developed to provide a limited-vocabulary, easy-to-use output for very young children or individuals with limited language abilities. In general, they require direct selection, but some also allow scanning. Rate enhancement in this category varies from none, to levels, to simple codes or key sequences. Vocabulary storage varies from a low of a few seconds to several minutes.

The devices in the direct selection, writing only category are distinguished by their small size and focus on features that support writing. Some may have a built-in printer. Several devices in this category provide direct file transfer to a desktop computer, and several also have rate enhancement (generally abbreviation expansion, instant phrases, or word completion).

The devices in the spelling only SGD category are primarily distinguished by their dependence on spelling for message formulation. They also generally are a small size and use direct selection through a keyboard or touch screen.

The last two categories in Table 11-4 represent the highest level of sophistication in currently available devices. They incorporate all the rate enhancement approaches discussed in Chapter 7. Those in the multiple selection method with rate enhancement category are based on SGD hardware that is specifically designed for AAC. The devices in the last category are software applications that are designed to run on general-purpose computers such as laptops, tablets, or PDAs. Vocabulary storage capacity varies from a few hundred utterances to thousands of utterances. Interaction with other devices (e.g., computers, Chapter 8; power wheelchairs, Chapter 12; or EADLs, Chapter 14) and peripherals such as printers is possible for most of the devices in this group using either serial or parallel ports (generally USB). Within these two categories are devices that can meet the needs of a variety of consumers, from very young children who cannot spell to quantum physicists who make full use of sophisticated rate enhancement techniques. In some cases the same device can serve a wide range of needs because the software and vocabulary stored can be customized. In other cases the devices are relatively inflexible.

Devices in the last two categories provide great flexibility in control interfaces and selection methods. Several of the direct selection types allow both standard size and expanded or contracted keyboards as control interfaces. Several devices in these two categories allow scanning with single-switch or four- or five-switch directed scanning. Some also provide both one- and two-switch Morse code, and some provide direct selection by head pointing. For direct selection by head pointing, some devices use light pointers or sensors attached to the head, whereas others use reflective systems requiring the attachment of only a reflective dot. Some light pointers can also be held in the hand.

The flexibility provided by devices in these categories is particularly useful in dealing with degenerative diseases such as ALS. Initially a person may use direct selection with the hand. As this ability is lost, direct selection by head control is feasible. However, because the device has not changed, the stored vocabulary, rate enhancement strategies, and operational characteristics of the device remain the same. If direct selection by head control becomes impossible, scanning or Morse code can be used. Once again the device is not changed, and the vocabulary, rate enhancement, and operational features remain the same. This is a great advantage over having to learn a new device at each stage of the disease.

What distinguishes the last two categories is the hardware platform. The human/technology interface may use either a static or dynamic display with a variety of ways in which each of these user interfaces is implemented on the various devices listed in Table 11-4. The devices in this category allow for a variety of physical and cognitive skills on the part of users.

In summary, the categories shown in Table 11-4 are intended to provide a rough framework with which to view SGDs. It is important for the ATP to remain current regarding technologies. One of the easiest ways to do this is to attend conferences that feature assistive technologies. If the ATP will be charged with making recommendations for AAC systems and approaches, it is also important to have his or her name placed on the mailing lists for the manufacturers of these devices. Most SGD manufacturers also maintain home pages on the Internet. There are several Web sites that provide links to SGD manufacturers.

Training System Use: Developing Communicative Competence.

Figure 11-16 shows a completed installation of an SGD. When the installation is completed, the individual and those working with him (e.g., care providers, family, teachers, employers, therapists, and speech-language pathologists) can begin the process of learning to use the device. Depending on the complexity of the device and the sophistication of the features included, this process can take from a few hours to several months.

The development of communicative competence is most effective when a comprehensive program is used. One such approach is the System for Augmenting Language (SAL) (Sevick, Romski, and Adamson, 2004). SAL involves a multimodal approach to training of individuals who rely on AAC, their partners, and continuing follow-up. Sevick, Romski, and Adamson (2004) illustrate the application of SAL through a case study of a preschool child who used both a VOCA and a manual display consisting of PCS symbols. For young children with cognitive and language disabilities, the development of both expressive and receptive vocabulary can be developed by using a VOCA in an exercise to teach requesting (Brady, 2000). Children were taught to request objects using PCS symbols on a VOCA. After learning these symbols, the children’s comprehension was evaluated. The use of the VOCA during the labeling instruction appeared to increase later comprehension of the symbols.

Scripts that are programmed into an AAC device can be used in a training paradigm. One formal approach is called “Script Builder” (Linda J. Burkhart, Eldersberg, Md., www.Lburkhart.com). The scripts are a way of training individuals to achieve greater social competence and more effective interactions. The scripts are coplanned and oriented toward the development of social closeness by encouraging social purposes and a sense of belonging. Typical topics of trivia, sports, gossip, hanging out, and “who’s cute” allow the individual to display aspects of his or her personality through humor, teasing, whining, and joke telling. Scripts change perceptions of individuals who use AAC because greater social competence is evident. Some scripts focus on information content, others on conversation scripts (new information plus social closeness). Example scripts are shown in Box 11-3. There are three roles in the training: the individual who is developing AAC skills, his or her partner, and a “prompter” who prompts only when necessary in an unobtrusive way. The partner’s role is communicating as naturally as possible, pausing when necessary, and not prompting at all. All social scripts start with a greeting and include a range of communicative functions such as positive and negative comments, teasing, and questioning. They provide for multiple turns and use topic maintainers like “tell me more.” They need to be designed to ensure that the individual doesn’t get “backed into a corner.” The vocabulary chosen is appropriate to the individual’s age and setting and personality.

Communicative competence depends on many factors (Light, 1989). The context in the HAAT model affects competence in several ways. The partner and his or her skill in listening, the environment of use, and cultural factors all contribute to or detract from communicative competence. The degree of competence is also variable, and complete mastery of an AAC device is not necessary to have functional communication interactions. Light (1989) describes four areas of competence required for successful use of AAC devices: (1) operational, (2) linguistic, (3) social, and (4) strategic.

Operational competence requires the physical skills described earlier and an understanding of the technical operation of the AAC device. Once again, the degree of operational competence can be quite variable, from very basic operation to advanced features. An AAC device is like a musical instrument that can be played by an accomplished AAC communicator. Operational competence includes the cognitive demands dictated by rate-enhancement techniques. Training operational competence requires a systematic introduction of technical features, coupled with ample opportunities for practice in their use, as shown in Figure 11-17. The individual’s facilitators must also be trained in certain operational features of the device (e.g., battery charging, connecting control interfaces), even though they will not develop the same level of competence as the individual.

The second phase, basic operation, includes how to connect the device to the control interface, how to charge batteries, how to attach it to a wheelchair, how to add vocabulary using rate-enhancement techniques (e.g., codes), and an introduction to troubleshooting in case the device fails to operate properly.

The last features to be introduced are those related to storage of new vocabulary, input acceleration techniques, and vocabulary manipulation features such as text editing and reformatting the output. Often the first two phases are accomplished in one session. However, in some cases, they may require multiple training sessions, and the process is often a lengthy one that may be integrated with the other aspects of training in communicative competence.

Linguistic competence requires that the symbol system and rules of organization be understood by the individual using the AAC system. As Light (1989) points out, the individual often must be competent in two languages: the spoken language of the community and the language of the AAC device. It is likely that the individual also lacks models of proficient use in the language of the device. Development of competence in this area may require many hours of practice. Often this practice is built around a functional reading task, such as that shown in Figure 11-18.

In contrast to the typical “drill and practice” approach to developing vocabulary and AAC use, Mirenda and Santogrossi (1985) used a prompt-free strategy to teach a young child to use a picture-based communication board. The approach involved a four-step process, which began with a picture of a soft drink being available to the child during her regular therapy session. A drink was visible to her, as was the picture of the drink. The child was not told that touching the picture would result in her getting a drink, nor was she prompted in any way to touch the picture. If she touched the drink directly, she was told that she could have some later. If she accidentally or deliberately touched the picture, she was immediately given the drink with the explanation, “Yes, if you touch the picture, you may have the drink.” Once the deliberate response had been established over several sessions, Mirenda and Santogrossi proceeded to shape the pointing behavior by progressively moving the picture farther away, until it was out of sight and the child had to actively find it. As the child became proficient in this task, the number of pictures was increased to four and the process repeated for the other choices. Eventually the child was able to generalize to a language board of 120 pictures. The advantage of this approach is that the child learns the meaning and significance of the symbolic representation by discovery rather than by drill, which leads to greater generalization and more functional use of the AAC system.

Many people who use AAC devices have little or no experience in social discourse. Even individuals who have used natural language for communication and who have sustained a disease or injury are faced with a very different mode of interaction when an AAC device is used. Rules of conversation are altered, and the perception of the individual by his or her communication partners is different. To be socially competent, the individual must have knowledge, judgment, and skills in both sociolinguistic (e.g., turn taking, initiating a conversation, conversational repair) and sociorelational areas (Light, 1989). The latter term describes the understanding of interaction between individuals. The effective communication device user is described (Light, 1988) as having a positive self-image, interest in his or her partner, skill at drawing others into the conversation, ability to put a partner at ease, and active participation in the conversation. These are sociorelational skills, and the degree to which they are understood and used is one measure of social competence. These skills are best taught in the contexts in which they are to be used. One example of such training is shown in Figure 11-19, in which the child is being taught strategies for interacting with an adult partner. Self-determination is difficult for people who rely on AAC (Collier, 2005). They must know what they want, know how to get it, and have a sense of self-worth. To achieve these goals, they need the “language of negotiation” and negotiation skills that require transactional language to supplement requesting, information exchange and conversational control vocabulary and skill. Without these skills, individuals who rely on AAC are dependent on others for the determination of their life goals and direction. They also need these skills to avoid abuse and harassment by care givers and others or to report incidents if they do occur.

Every person who uses an AAC system develops strategies to make that use more effective. Examples include letting the partner guess the next letter on a spelling board and using gestures (e.g., waving to indicate that a misunderstanding has occurred) in conjunction with an electronic device. Strategic competence describes the degree to which the person is able to develop adaptive strategies to make the most of the system. These may differ in different contexts. For example, a child’s speech may be better understood at home than at school. He or she will rely on the electronic SGD more in school but will also develop strategies to make maximal use of both systems.

Just as the individual using the AAC system must develop several types of competencies, there are many ways of carrying out the training. One approach, shown in Figure 11-19, is to simulate a situation, model the types of interaction likely to occur, and have the user “practice” the strategies and skills necessary to make it a success. This step can be followed by an actual situation in which the ATP accompanies the user as he or she encounters the situation. The ATP can then prompt the user at appropriate times, add encouragement, and help to clarify when necessary. This combination of clinic-based practice and community-based skill development is often very effective.

For training to be effective, staff must have sufficient skill and experience to assist the AAC user, which requires training for those who are supporting AAC use. Schepis and Reid (2003) identified seven basic steps in competence- and performance-based training for staff. These include specifying desired outcomes, roles for staff to support individuals in achieving these outcomes, providing both written and oral expectations and instructions to staff, demonstration of how to perform duties, and observation of staff performing the duties with corrective feedback as necessary.

Training of the individual who relies on AAC is only effective if communication partners are also trained. For children, the training of parents to recognize communication attempts and to understand the operational, linguistic, strategic, and social competencies is also important. Bruno and Dribbon (1998) evaluated a parent training program conducted as part of an AAC summer camp experience where parents attended the camp with their children. The camp featured structured therapy sessions, activities with nondisabled campers, and activities planned for families. The parent training had both device (conducted by manufacturers’ representatives) and interaction training aspects. Parents reported making positive changes in both operational and interactional skills during the camp (Bruno and Dribbon, 1998). These changes were reflected in gains made by the children in skills related to the use of pragmatic functions (e.g., giving and requesting information, requesting assistance, responding, and protesting) over the course of the camp. The camp training significantly increased the degree to which parents gave their children access to the AAC system. In some cases, skills in these areas remained constant at the 6-month follow-up, and in some there was a decrease. The areas of social exchange and giving of information continued to increase at the 6-month follow-up evaluation.

AAC training can be both complicated and lengthy (Beukelman and Mirenda, 2005). Light and Binger (1998) have developed a seven-step process for developing AAC communication competence: (1) specify the goal, do baseline observations; (2) select vocabulary; (3) teach the facilitators how to support development of the target skill; (4) teach the skill to the target individual; (5) check for generalization; (6) evaluate outcomes; and (7) complete maintenance checks. Light and Binger provide data collection and assessment forms and strategies for implementing this program. ACETS (Augmentative Communication Employment Training and Supports) (Institute on Disabilities, Temple University, Philadelphia, Pa., http://:disabilities.temple.edu) has been developed specifically to assist those who rely on AAC in seeking employment. A formal training manual is available for this program that is based on three principles: (1) immersion in the workplace culture, (2) acquiring a broad base of employment-related skills and experience, and (3) support of individualized goals.

Follow-up: Measuring Short- and Long-Term Outcomes.

The evaluation of communicative competence in the four domains (operational, linguistic, social, and strategic) will identify areas in which the AAC system is and is not adequately meeting the individual’s needs. Periodic re-evaluation of the individual’s skills and needs may also result in changes in the training or the AAC system(s). The re-evaluation may lead to new training goals in one or more of the four areas of communicative competence. In other cases the care givers, family, or other support staff may require additional training to facilitate the use of the AAC device.

The AAC device as it is configured may also be inadequate to meet the individual’s needs. It may be possible to adjust some of the features (e.g., scanning rate, stored vocabulary), or it may be necessary to consider a completely new device. The magnitude of the changes in the device dictates the amount of additional operational training required. In some cases the individual’s skills may decrease (e.g., degenerative disease) or increase (e.g., a young child who develops greater language skills). In either case a reevaluation and adjustments in the AAC system (device plus training and support) will be required.

Murphy et al (1996) identified obstacles to effective AAC system use in a study of 93 users of AAC systems and 186 partners (93 formal and 93 informal). The formal partners were speech-language pathologists (the majority), care providers in the day or living program, and teachers. Informal partners were family, friends, and others selected by the AAC users as those with whom they felt most comfortable using their AAC systems. In some cases one partner filled both the formal and informal roles. The majority of low- and high-tech AAC system use was in the day placement (90%), residential (70%), and leisure (60%) settings. Use was limited to organized therapy sessions in general.

AAC systems were only available to 48% of the users while shopping, 62% during outings such as day trips during their program, and 66% where they lived. SLPs were the most frequent (80%) formal partners, and residential or day care staff were the most common informal partners (62%). Friends and family were both reported as the primary informal partner in less than 10% of the cases. Only 57% of the low-tech and 59.4% of the high-tech AAC system users were able to independently access their systems (e.g., get a system out of a back pack on a wheelchair and set it up for use without a partner’s help). Knowledge of the system sufficient to interact with the AAC user was reported in less than half of the formal partners and one third of the informal partners. Eighty-eight percent of the users received training from their formal partners. However, for the majority of the users, the training consisted of 60 minutes or less (or 40 hours per year on the basis of sessions conducted). This number is low compared with other types of therapy and training such as that for second language instruction (estimated by Murphy et al to be more than 200 hours per year).

Murphy et al found that basic vocabulary required for daily interactions (see Table 11-5) was not included in the AAC systems. Few users had greetings, wrap ups, or conversational flow vocabulary (e.g., comments, repair vocabulary). Thus, for these areas, the users most commonly used other modes of communication (e.g., eye gaze, gestures, facial expressions) rather than their AAC devices.

The preponderance of formal partners also reinforces the need for inclusion of both useful vocabulary and multiple modes of communication. The development of strategic competence is vital to increase the likelihood that an AAC user will be able to independently carry out conversations in a variety of settings and with a variety of partners. Availability and accessibility of AAC systems can be addressed by appropriate mounting of systems on wheelchairs and training to ensure that care providers understand the need to have the system available to the user at all times. The results reported by Murphy et al also emphasize the importance of multiple modes of communication by AAC users.

Assessment of AAC outcomes can use the general measures (MPT, PIADS, Quebec User Evaluation of Satisfaction with Assistive Technology [QUEST]) described in Chapter 4. Because of the nature of communication, there are additional considerations. Because the major goal of AAC is the provision of expressive language capability, one of the most important considerations is the determination of communicative intent by the individual evaluating outcomes (typically SLP or special education teacher). Special education teachers tend to overassess intentionality (i.e., to assign intentionality more often than experienced researchers), whereas SLPs do so less often (Carter and Iacono, 2002). Individuals with different disorders are also assessed differently relative to intentionality, and observations are inconsistent across populations, sessions, and professional groups. These results are disturbing because intentionality is a key measure of communicative competence and effectiveness.

The MPT model (Chapter 4) has been adapted to AAC use as the AAC Acceptance Model (Lasker and Bedrosian, 2001). This model focuses on the prediction of acceptance of AAC by adults with acquired communication impairments. Although the technology may play a small role in acceptance or nonacceptance, other factors are more important. Among these other factors are the following: (1) the communication partners’ acceptance of the technology; (2) the rate (sudden or gradual) of onset of the communication impairment; (3) affective, behavioral, and cognitive components of a user’s attitude toward AAC technology; (4) perception of the user and other people toward the device, and (5) how other people view the person using the device. It is not clear whether this measure can be generalized to other populations (e.g., children with developmental disabilities).

There are a number of key reasons that provision of AAC systems may not achieve the goal of a “better life” for the AAC user (Beukelman and Mirenda, 2005): (1) payer resistance to or lack of acceptance of measures that reflect quality of life rather than more concrete functional outcomes, (2) increased costs of intervention necessary to achieve broader goals, (3) time limits set by payers on length of the intervention, (4) high demands on professionals to achieve and maintain skills, (5) family and user response to increases in their responsibilities in assuming a leadership role, (6) difficulty by families in envisioning the future for the AAC user, and (7) cultural differences between the user and professionals.

We can relate meaningful AAC outcomes to the three levels of the World Health Organization ICF classification system (see Chapter 2) (Beukelman and Mirenda, 2005). At the level of body structures and functions the degree to which AAC intervention compensates for lost or absent speech or language function can be determined. Evaluations related to activity focus on the quality and quantity of communication interactions and the degree to which these meet the goals and needs of the individual. Evaluations related to participation focus on the socially defined role and tasks within a sociocultural and physical environment. Some “Big Picture” AAC outcomes are shown in Box 11-4 (Beukelman and Mirenda, 2005). There are several types of measures for AAC system outcome. Operational measures evaluate the user’s ability to interact with the system itself (operational competence), whereas representational measures evaluate symbol and grammatical capability by the AAC user (Beukelman and Mirenda, 2005).

The most important result of the follow-up phase is to evaluate the outcomes of the AAC interventions to determine their effectiveness, including both the hard and soft technologies, and the appropriateness of the match between the originally specified needs and the resulting system. The principles of outcome measurement discussed in Chapter 4 apply to AAC system evaluation as well.

1. What are the two major communicative needs normally addressed by augmentative communication systems?

2. Distinguish between aided and unaided communication.

3. What are the major goals for augmentative communication systems designed for conversational use?

4. What AAC needs do parents have for their nonspeaking children? Do mothers and fathers have the same needs for their children?

5. Describe differences in the conversational rules that apply between two speaking persons and those between one speaking person and one augmentative communication user.

6. Describe the relationship between the Social Networks model and the Participation Model. How do each of these relate to the HAAT model described in this text?

7. How do attitudes of the communication partners differ for the five circles of the Social Networks model?

8. What factors influence the attitudes of children toward their peers who use AAC?

9. What features distinguish competent augmentative communicators from those who are not successful?

10. Select three discourse functions from those listed in Table 11-5. Now pick an augmentative communication device (e.g., electronic, direct selection, with voice output; or a language board with letters and words) and develop a vocabulary and set of strategies for the implementation of each of the discourse functions that you choose.

11. What are the three types of graphical communication? List three ways in which they differ.

12. In what ways does the formal writing of adolescent AAC users differ from that of nondisabled adolescent writers?

13. Distinguish between formal writing and note taking in terms of the characteristics AAC devices must have to meet each need. What is the most important feature in each case?

14. What two factors must be included for a math worksheet to be effective for both arithmetic and higher math (e.g., algebra)?

15. List three features that a drawing system should have to be of use in creative expression.

16. How do drawing systems differ in structure and function from systems for scientific plotting?

17. Describe auditory scanning. Give an example of both a low-tech or no-tech approach and an electronic AAC approach. What are the essential features for the AAC auditory scanning device?

18. List three encoding methods used in AAC devices, and give one advantage and one disadvantage of each.

19. What are the major types of abbreviation approaches used in AAC devices, and what are the major advantages and disadvantages of each?

20. Pick three discourse functions and develop a logical coding scheme for each using (1) numerical codes, (2) abbreviation expansion, and (3) Minspeak codes.

21. Compare word completion and prediction with abbreviation expansion and Minspeak encoding.

22. What are the major approaches used to increase conversational rate when the individual is using scanning?

23. What are dynamic displays, and what advantages do they provide?

24. What are visual scene displays and what unique features do they have?

25. What populations might benefit most from visual scene displays? Why?

26. Describe the major challenges and approaches for AAC intervention of individuals whose primary disorder is language or cognitively based? How does this compare with individuals whose primary disorder is motor or physical?

27. List and discuss three advantages that the Internet provides for communication by AAC users.

28. What are the major advantages of conversationally based communication devices such as CHAT and TOPIC?

29. What are the four types of competencies acquired in AAC training? Pick an AAC system for an individual and design the training. You must make assumptions regarding the person’s skills, her needs, and other people available to help facilitate the training.

30. For each of the categories of devices described in the section on current technologies, define a user profile (skills and needs) that would lead you to focus on that category in selecting a device for that person.