Aided Language Modeling for Preschoolers with Autism: A Literature Review
By Karyn E. Whitney
Florida International University
This issue of NASET’s Autism Spectrum Disorder Series was written by Karyn E. Whitney. Young children with autism exhibit significant communication deficits that may impact their ability to engage in social interactions and develop and maintain relationships (American Psychological Association, 2022). Enhanced social communication skills not only improve children’s social relationships but also predict better self-regulation and academic achievement in kindergarten (Ramsook et al., 2020). While research has established positive effects of augmentative and alternative communication (AAC) for young children with autism, AAC interventions comprise a variety of strategies that differ widely in cost and complexity (Steinbrenner et al., 2020). The studies discussed here support using an adaptable and relatively simple strategy called aided language modeling, which involves pairing spoken words with graphic symbols to improve the receptive and expressive language of individuals with complex communication needs. Though most existing research involves tightly controlled conditions, two of the studies discussed below support the use of aided language modeling by teachers in real-world classroom settings.
Abstract
Young children with autism exhibit significant communication deficits that may impact their ability to engage in social interactions and develop and maintain relationships (American Psychological Association, 2022). Enhanced social communication skills not only improve children’s social relationships but also predict better self-regulation and academic achievement in kindergarten (Ramsook et al., 2020). While research has established positive effects of augmentative and alternative communication (AAC) for young children with autism, AAC interventions comprise a variety of strategies that differ widely in cost and complexity (Steinbrenner et al., 2020). The studies discussed here support using an adaptable and relatively simple strategy called aided language modeling, which involves pairing spoken words with graphic symbols to improve the receptive and expressive language of individuals with complex communication needs. Though most existing research involves tightly controlled conditions, two of the studies discussed below support the use of aided language modeling by teachers in real-world classroom settings.
Keywords: augmentative and alternative communication, aided language modeling, autism, early childhood
Aided Language Modeling for Preschoolers with Autism: A Literature Review
The Individuals with Disabilities Education Act (IDEA) (2004) mandates free appropriate public education in the least restrictive environment for every child with a disability. When a child’s disability adversely affects his or her educational performance, the law requires creation, implementation, and monitoring of an Individualized Education Plan (IEP). Developed by a team that includes the child’s parent and teacher, an IEP describes the special education and related services a child with a disability needs to meet appropriate educational goals. Because communication skills play an essential role in learning, IDEA requires the IEP team to consider the child’s communication needs when developing the IEP. The law also requires that teachers select instructional methods supported by scientific evidence.
The Centers for Disease Control estimates 1 in 36 children meets the criteria for a diagnosis of autism spectrum disorder (Maenner et al., 2023). According to the American Psychological Association (2022), autism spectrum disorder is characterized by markedly impaired communication and social interaction. Carahan et al. (2016) assert that these communication deficits present the single greatest obstacle to the success of individuals with autism. In 2020, the National Clearinghouse on Autism Evidence and Practice (NCAEP) identified 25 practices that promote communication development in children with autism between the ages of zero and five. NCAEP reported that empirical support for augmentative and alternative communication (AAC) interventions increased significantly from 2011 to 2017, and the research base continues to grow. AAC interventions comprise a wide variety of strategies including but not limited to aided language modeling (Steinbrenner et al., 2020).
Predicated on the idea that symbolic communication development depends on linguistic input, aided language modeling involves pairing spoken words with AAC tools such as communication boards and speech generating devices to improve the receptive and expressive language of individuals with complex communication needs (CCN) (Biggs et al., 2018). Through a systematic review of studies conducted prior to 2016, Biggs et al. concluded that available evidence supported the use of aided language modeling to facilitate language mapping in children with CCN, provide instructional models for AAC use, and elicit specific communicative behaviors from AAC users.
Unfortunately, Paisley et al. (2023) found that teachers’ knowledge of evidence-based practices (EBPs) varies widely. Of 303 teachers surveyed, only about one quarter correctly identified an ASD-specific EBP and reported using an ASD-specific EBP in their classrooms. While research shows that early intervention can increase social communication in preschoolers with autism (Fuller & Kaiser, 2020), classroom teachers may lack the knowledge and resources to implement effective interventions. This literature review examines recent studies that explore potential benefits of aided language modeling for preschool students with autism while also examining potential challenges to successful implementation in the classroom setting.
Literature Review
Rowland (2013) identifies four primary reasons that people communicate: “to refuse things that we don’t want; to obtain things that we do want; to engage in social interaction; and to provide or seek information” (p. 7). According to Edgar et al. (2024), most AAC interventions focus on teaching children with CCN to request preferred activities. While learning to make requests empowers individuals to express their wants and needs, children communicate for other equally important purposes. Edgar et al. designed an intervention package to increase not only initiating requests but also answering questions and making comments in four preschool children with autism.
Aided Language Modeling and Speech Generating Devices (SGDs)
After assessing the activity preferences of study participants, Edgar et al. (2024) provided systematic instruction in identifying target symbols on speech generating devices (SGDs). Baseline, intervention, generalization, and maintenance phases began only after participants identified target symbols with 100% accuracy. During the intervention phase, a researcher provided aided language modeling coupled with a system of least-to-most prompting, constant time delay, error correction, and reinforcement. Unique among studies reviewed here, this study included both generalization and maintenance phases during which nondisabled peers also provided aided language modeling. While data suggests a relationship between the intervention package and increases in all three target behaviors for two study participants, researchers did not establish such a relationship for the other two participants. They offer the conclusion that aided language modeling paired with systematic instruction “may lead to gains in socio-communicative behaviors in some minimally speaking, preschool-aged autistic children” (Edgar et al., 2024, p. 4466).
Aided Language Modeling and Visual Scene Displays (VSDs)
Edgar et al. (2024) used iPad tablets with Proloquo2Go or TouchChat to create grids containing words and phrases associated with participants’ preferred activities (e.g., go, let’s play a game, etc.). In contrast, Muttiah et al. (2022) combined aided language modeling with visual scene displays (VSDs). VSDs represent real-life scenes or events with images or videos, and users communicate by selecting elements (i.e., hotspots) to convey specific messages. Though VSDs provide access to limited vocabulary, they provide more context than traditional grid-based AAC systems by using scenes that represent familiar environments and activities (Tuthill, 2014).
While applications like GoVisual from Attainment Company allow users to create VSDs with voice-output capabilities, Muttiah et al. (2022) designed nonelectronic VSDs for an intervention involving preschool children with CCN including one four-year-old with autism. First, researchers produced laminated photographs to represent common play activities and songs. Then, they created 2 to 6 hotspots for each VSD by attaching copies of specific elements to the photographs with transparent Velcro. During the intervention phase, a researcher used the VSDs to provide aided language modeling coupled with wait time and gestural cues. Comparison of pre- and post-intervention data indicated a significant increase in the number of communication turns and unique semantic concepts expressed by the child with autism. Due to the exploratory nature of this study, researchers did not measure generalization or maintenance.
Aided Language Modeling and Enhanced Milieu Teaching (EMT)
Logan et al. (2024) found that research regarding aided language modeling often occurs in “highly structured, decontextualized environments” (p. 125). For example, Edgar et al. (2024) conducted their research in a secluded corner of a classroom (p. 4470), while Muttiah et al. (2022) conducted their research in a hallway (p. 95). While controlled studies like these provide valuable data, they fail to replicate real-world conditions. In a recent study, Logan et al. (2024) combined aided language modeling with a naturalistic teaching strategy called enhanced milieu teaching (EMT). EMT draws on behavioral, developmental, and social interactionism theories, promoting communication development in the context of everyday interactions (Kaiser & Hampton, 2017).
Logan et al. (2024) recruited three participants with autism between the ages of 2 and 7 to participate in a home-based intervention. Following preference assessments, researchers developed six grid-based communication boards for each participant and made them available during preferred activities such as waterplay and snack time. During the intervention phase, a researcher engineered the environment to promote communication through such strategies as placing preferred objects in sight but out of reach, creating opportunities for turn-taking, and offering choices. The researcher also used specific interaction techniques characteristic of EMT such as language matching and expectant looking to promote communication between researcher and child. Additionally, the researcher provided aided language modeling through a system of providing wait time, requesting a response, requesting a response and providing verbal and gestural models, and physically prompting symbol use.
All three participants demonstrated increased production of symbolic communication over baseline levels, and data provides some evidence that aided language modeling coupled with EMT may support the acquisition of new communication functions in young children with autism. Furthermore, two of the three study participants generalized symbolic communication to interactions with caregivers. Although researchers conducted this study in participants’ homes, many of the strategies could be adapted to school-based environments (Logan et al., 2024).
Aided Language Modeling and the Visual Immersion System™ (VIS)
Unlike the investigators mentioned above, Schlosser et al. (2020) enlisted both children with autism and highly qualified educators as participants in a mixed-methods study of the Visual Immersion System™ (VIS). In Enhancing Communication for Individuals with Autism: A Guide to the Visual Immersion System, Shane et al. (2014) outline the principles of this proprietary clinical strategy, which include using low-, mid-, and high-tech visual supports to promote comprehension, organization, and expression in individuals with autism.
In a proof-of-concept study, Schlosser et al. (2020) taught educators to implement VIS in a self-contained classroom for children with autism ages 6-8. Training topics included “problem behaviors, prioritizing goal areas, technology applications and approaches, classroom design, curricular issues and applications, and materials creation” (pp. 451-452). While their report mentions aided language modeling repeatedly, Schlosser et al. do not explicitly describe aided language modeling procedures employed in the study.
Reported data indicates that the coaching intervention resulted in student learning gains as measured by the Communication Matrix (Rowland, 2013) and by progress toward the attainment of individualized goals. The teacher, speech-language pathologists, and instructional technologist involved in the study perceived the intervention as effective and socially valid, while educational aides and parents, who did not receive coaching from researchers, expressed a desire for training in the use of visual supports (Schlosser et al., 2020). At the very least, this study suggests that, given appropriate training and support, school personnel can successfully implement aided language modeling in classroom settings.
Aided Language Modeling and Core Vocabulary
Like Schlosser et al. (2020), Dorney and Erickson (2019) recruited both children and educators to participate in a mixed-methods study. They designed an instructional method in which educators attribute meaning to conventional and unconventional behaviors, provide aided language modeling using a grid-based communication board, and target core vocabulary. During seven professional development sessions over the course of one school year, investigators taught educators these procedures and the underlying principles. They also conducted eight to ten observations in each of three self-contained preschool classrooms for children with autism to gather data regarding implementation fidelity and intervention effects.
Dorney and Erickson (2019) found that, although teachers frequently provided aided language modeling, they rarely attributed meaning to students’ non-symbolic behaviors or demonstrated symbolic equivalents. Nevertheless, students’ median scores on the Communication Matrix (Rowland, 2013) increased, indicating progress toward conventional communication. Specifically, participants “learned to use abstract graphic symbols representing core vocabulary to request… [And] a number of students demonstrated growth in use of non-symbolic communication for social interaction and information sharing” (p. 42).
Conclusion
The researchers who conducted these studies used various terms for aided language modeling including aided AAC modeling (Edgar et al., 2024), aided modeling (Muttiah et al., 2022), aided AAC (Logan et al., 2024), augmented input (Schlosser et al., 2020), and aided language input (Dorney & Erickson, 2019). Regardless of the terminology or specific AAC tools they chose, they all designed interventions that involved pairing spoken words with graphics to support children’s receptive and expressive language development. When combined with low-tech tools like communication boards, aided language modeling seems relatively easy and inexpensive to implement. Even so, these studies highlight the complex challenges presented by the CCN of preschoolers with autism.
Steinbrenner et al. (2020) found that various researchers have designed intervention packages that pair aided language modeling with other evidence-based practices “such as prompting, reinforcement, visual supports, and peer-mediated interventions” (p. 66), and each of the studies discussed in this review involved a packaged intervention. Therefore, any observed effects cannot be attributed solely to aided language modeling (Biggs et al., 2018). However, like studies examined by Biggs et al. (2018), these studies suggest that aided language modeling may benefit young children with autism.
Unfortunately, only two of these studies took place in classroom settings. Martin et al. (2020) found that interventions for individuals with autism “have disproportionately been studied in clinical settings” (p. 84), and “the majority of school-based intervention studies feature tightly controlled study conditions that may limit external validity” (p. 75). While carefully controlled studies yield invaluable information, teachers may struggle to implement interventions developed under ideal conditions successfully. Further research will help determine the best strategies for implementing aided language modeling under conditions typical found in preschool classrooms.
References
American Psychiatric Association. (2022). Diagnostic and statistical manual of mental disorders (5th ed., text rev.). doi.org/10.1176/appi.books.9780890425787
Biggs, E. E., Carter, E. W., & Gilson, C. B. (2018). Systematic review of interventions involving aided AAC modeling for children with complex communication needs. American journal on intellectual and developmental disabilities, 123(5), 443-473. doi.org/10.1352/1944-7558-123.5.443
Carnahan, C., Chung, C., Schaefer-Harbison, A., & Carnett, A. (2016). Teaching students with autism to communicate. In E. A. Boutot (Ed.), Autism spectrum disorders: Foundations, characteristics, and effective strategies (2nd ed., pp. 96-122). Pearson.
Dorney, K. E., & Erickson, K. (2019). Transactions within a classroom-based AAC intervention targeting preschool students with autism spectrum disorders: A mixed-methods investigation. Exceptionality Education International, 29(2), 42-58. doi.org/10.5206/eei.v29i2.9401
Edgar, T. C., Schlosser, R. W., & Koul, R. (2024). Socio-communicative behaviors involving minimally speaking autistic preschoolers and their typically developing peers: Effects of an augmentative and alternative communication intervention package. Journal of Speech, Language, and Hearing Research, 67(11), 4466-4486. doi.org/10.1044/2024_JSLHR-24-00210
Fuller, E. A., & Kaiser, A. P. (2020). The effects of early intervention on social communication outcomes for children with autism spectrum disorder: A meta-analysis. Journal of Autism and Developmental Disorders, 50(5), 1683-1700. doi.org/10.1007/s10803-019-03927-z
Individuals With Disabilities Education Act, 20 U.S.C. § 1400 et seq. (2004).
Kaiser, A. P., & Hampton, L. H. (2017). Enhanced milieu teaching. In R. J. McCauley, M. E. Fey, & R. B. Gillam (Eds.), Treatment of language disorders in children (2nd ed., pp. 87-120). Paul H. Brookes Publishing.
Logan, K., Iacono, T., & Trembath, D. (2024). Aided enhanced milieu teaching to develop symbolic and social communication skills in children with autism spectrum disorder. Augmentative and Alternative Communication, 40(2), 125-139. doi.org/10.1080/07434618.2023.2263558
Maenner, M. J., Warren, Z., Williams, A. R., Amoakohene, E., Bakian, A. V., Bilder, D. A., Durkin, M. S., Fitzgerald, R. T., Furnier, S. M., Hughes, M. M., Ladd-Acosta, C. M., McArthur, D., Pas, E. T., Salinas, A., Vehorn, A., Williams, S., Esler, A., Grzybowski, A., Hall-Lande, J., … Shaw, K. A. (2023). Prevalence and characteristics of autism spectrum disorder among children aged 8 years—autism and developmental disabilities monitoring network, 11 sites, United States, 2020. Morbidity and Mortality Weekly Report. Surveillance Summaries, 72(2), 1-14. dx.doi.org/10.15585/mmwr.ss7202a1
Martin, R. J., Anderson, C. M., Gould, K., Morganelli, M., & Kleinert, W. L. (2021). A descriptive secondary analysis of evidence-based interventions for students with autism spectrum disorder. Contemporary School Psychology, 25(1), 75-85. doi.org/10.1007/s40688-020-00282-0
Muttiah, N., Drager, K. D., Beale, B., Bongo, H., & Riley, L. (2022). The effects of an intervention using low-tech visual scene displays and aided modeling with young children with complex communication needs. Topics in Early Childhood Special Education, 42(1), 91-104. doi.org/10.1177/0271121419844825
Paisley, C. A., Eldred, S. W., Cawley, A., & Tomeny, T. S. (2023). Teacher-reported classroom strategies and techniques for students with autism spectrum disorder. Focus on Autism and Other Developmental Disabilities, 38(3), 158-167. doi.org/10.1177/10883576221125485
Ramsook, K. A., Welsh, J. A., & Bierman, K. L. (2020). What you say, and how you say it: Preschoolers’ growth in vocabulary and communication skills differentially predict kindergarten academic achievement and self-regulation. Social development, 29(3), 783–800. doi.org/10.1111/sode.12425
Rowland, C. (2013). Communication Matrix for Parents and Professionals. Oregon Health & Science University. www.communicationmatrix.org/uploads/pdfs/handbook.pdf
Schlosser, R. W., Shane, H. C., Allen, A. A., Benz, A., Cullen, J., O’Neill, L., Chiesa, L., Miori-Dinneen, L. Koul, R., & Pasupathy, R. (2020). Coaching a school team to implement the Visual Immersion System™ in a classroom for children with autism spectrum disorder: A mixed-methods proof-of-concept study. Advances in Neurodevelopmental Disorders, 4(4), 447-470. doi.org/10.1007/s41252-020-00176-5)
Shane, H. C., Laubscher, E., Schlosser, R. W., Fadie, H. L., Sorce, J. F., Abramson, J. S., Flynn, S., & Corley, K. (2014). Enhancing communication for individuals with autism: A guide to the Visual Immersion System. Paul H. Brookes Publishing.
Steinbrenner, J. R., Hume, K., Odom, S. L., Morin, K. L., Nowell, S. W., Tomaszewski, B., Szendrey, S., McIntyre, N. S., Yücesoy-Özkan, ?., & Savage, M. N. (2020). Evidence-based practices for children, youth, and young adults with autism. Frank Porter Graham Child Development Institute at The University of North Carolina at Chapel Hill. ncaep.fpg.unc.edu/sites/ncaep.fpg.unc.edu/files/imce/documents/EBP%20Report%202020.pdf
Tuthill, J. (2014). Get real with visual scene displays: Use of real-life scenes adds appeal to AAC apps that clients use to communicate about everyday needs and situations. The ASHA Leader, 19(6), 34-35. https://doi.org/10.1044/leader.APP.19062014.34
Note: Correspondence concerning this article should be addressed to Karyn E. Whitney, Florida International University. Email: kwhit145@fiu.edu
To download a PDF file version of this issue of NASET’s Autism Spectrum Disorder Series: Click Here
To return to the main page for NASET’s Autism Spectrum Disorder Series – Click Here