JCPSLP Vol 21 No 3 2019

may hit a single switch on which is placed a graphic representation of the message “my turn,” which results in the child being given a turn in a game. In this example, the child may not have the representational understanding of the picture, but the aim may be to teach the equivalence between a consistent behaviour (hitting a switch) and a desired outcome, thereby enhancing understanding that this is the way to gain access to a game (i.e., to request a turn). Teaching basic pragmatic functions through the use of conventional forms provides the individual with the experience of power to effect a response from the environment, thereby countering experiences of learned helplessness and frustration. An early AAC text by Reichle et al. (1991) provides chapters addressing the use of symbols to teach a generalised request (e.g., graphic symbol or sign for “want”) or generalised rejection (e.g., graphic symbol or sign for “no”). Through these interventions, individuals who may be presymbolic learn basic pragmatic functions that are readily communicated and responded to, and thereby provide a means to shape behaviours towards intentionality. The next step is to provide the opportunity to expand them to symbols for specific requests and rejections, which may or may not eventuate. Implications for clinical practice Rejection of candidacy requirements in favour of the participation model, and viewing all extant behaviours as part of multimodal AAC carries the potential for symbolic systems to be introduced to individuals who may not have symbolic skills. Arguably, and as previously mentioned, this practice is appropriate if the AAC system is used to teach use of a conventional form to demonstrate basic pragmatic functions. A potential problem arises if the expectation is that the person will be able to use the symbols, regardless of the modaliy, as abstract representations, which can lead to failure of the AAC intervention (Rowland & Schweigert, 2000). In a study involving children with severe and multiple disabilities, colleagues and I argued against this practice because of the potential ‘to increase the child’s communication failures, reduce his/her attempts to communicate, and reduce the faith in the effectiveness of augmentative and alternative communication systems among care givers’ (Iacono et al.,1998, p. 112). It would seem then, that the onus is on clinicians to select goals based on their understanding of a person’s underlying competence, and to clearly articulate these to families and other support people to ensure a common understanding of the role of an AAC system and the various modalities encompassed in the intervention: for example, to expand pragmatic functions or to extend language skills. Failing to do so can create confusion about the intention of intervention and resentment towards clinicians, as was found by Edwards, Brebner, McCormack, and Macdougall (2016) in interviews of parents about what they wanted from therapists working with their children on the autism spectrum. Conclusion An early focus on multimodality in the AAC literature reflected concerns about practices that limited options for people with complex communication needs. As understanding of AAC as being multimodal by its very nature, and the benefits of providing people with varied and multiple forms of AAC grew, the profession of SLP expanded its scope of practice in relation to AAC such that

parents to adapt their responses to their children’s communication signals … can be considered AAC”. Issue can be taken with this statement in light of an accepted definition of AAC as combining communication strategies, techniques and interventions, in which formal modalities are provided through unaided and/or aided systems (Speech Pathology Australia, 2012). Hence, Cress and Marvin appear to be confusing an intervention strategy (one component) with the whole (system). On the other hand, Cress and Marvin (2003) reflected a concern in the AAC field of waiting until certain prerequisites, mostly cognitive (especially imitation, object permanence and means-end behaviour), were demonstrated or individuals failed to develop speech after prolonged speech training before being given access to AAC (Wilkinson & Hennig, 2007). This practice reflected a candidacy model, whereby individuals, regardless of age, needed to demonstrate that they were ready for AAC: that is, they had demonstrated candidacy requirements. The implementation of candidacy models left people with complex communication needs without access to a functional means of communication, and perhaps of the means by which they could demonstrate underlying communication capacity. Such capacity can be masked by multiple impairments (Iacono, Carter, & Hook, 1998; Rowland & Schweigert, 2000), learned helplessness arising when people in the social environment cannot read subtle or unconventional signals (Reichle, York, & Sigafooos, 1991), or years of failure to learn speech despite intensive intervention (Wilkinson & Hennig, 2007). Multimodal and AAC interventions for non-symbolic communicators The shift from a candidacy to a participation model, whereby assessment to determine appropriate AAC systems is conducted with individuals across ability levels (Mirenda, Iacono, & Williams, 1990), resulted in attention to interventions for children who were pre-intentional (Rowland & Schweigert, 2000) or adults who were unintentional 1 (McLean, Brady, McLean, & Behrens, 1999). These individuals have benefited from access to symbolic forms either in sign or on aids, but some have not progressed to symbolic communication. Rowland and Schweigert (2000), for example, implemented an AAC intervention in which 41 children with various severe and multiple disabilities were taught to request using tangible symbols. The children had varied outcomes: some quickly acquired the tangible symbols and then moved to more abstract forms (including speech for some), others acquired a few symbols and some, none at all. Rowland and Schweigert argued that children who did not progress to symbolic communication during the three years of the study should receive intervention with a focus on ‘strengthening presymbolic means of communication and providing receptive exposure to symbols’ (p. 74). A number of researchers have focused intervention on enhancing presymbolic communication, with the work of Yoder, Warren, Kim, and Gazdag (1994) providing an example. The intervention in this study was not inclusive of AAC, but rather demonstrated the need for consistent responses to behaviours that meet criteria for intentional communication or can be interpreted as communicative, even if intention to communicate is not evident (Iacono et al., 1998). AAC interventions can mediate between difficult to read signals and consistent responses from people in a person’s environment. As an example, a child

128

JCPSLP Volume 21, Number 3 2019

Journal of Clinical Practice in Speech-Language Pathology