Search
Search
Close this search box.

Using Speech-Generating Devices to Teach Language Skills

In November, OAR’s Board of Directors authorized funding for eight applied autism research studies in 2024. These new grants, totaling $297,569, bring OAR’s total research funding to more than $5 million since 2002. This article is the second of eight previews to be featured in The OARacle this year.

Despite the growing body of autism research, studies often exclude students with profound autism and who are minimally verbal. Of the nearly 876,000 autistic students between the ages of 3 and 21 who were served under the Individuals with Disabilities Education Act (IDEA), nearly 30%, or almost 263,000, were minimally verbal.

Minimally verbal individuals often use augmentative and alternative communication (AAC) systems, including speech-generating devices. Preliminary research suggests that speech-generating devices can be used to teach a variety of language skills. However, research examining the use of speech-generating devices by children in the classroom tends to focus on functional communication skills, such as requesting, with far less attention to using the devices during academic lessons.

Cognitive and language skills are critical areas for intervention, given the association with reduced symptom severity. To ensure that autistic students who use those devices can receive that instruction, more research is needed on teaching language skills during academic instruction.

Studies done at the Munroe-Meyer Institute at the University of Nebraska Medical Center, including studies done by co-principal investigators Alice Shillingsburg, Ph.D., BCBA-D, and Sarah Frampton, Ph.D., BCBA-D, have indicated that direct instruction may be effective for promoting language development for autistic children. Additionally, a small study the OAR-funded researchers did with three autistic minimally verbal students using speech-generating devices supported the use of a direct instruction curriculum to teach language skills.

In this one-year OAR-funded study, Extending Direct Instruction to Students with ASD Using Speech Generating Devices, Dr. Alice Shillingsburg and Dr. Frampton will evaluate the feasibility and efficacy of direct instruction with autistic children who are minimally verbal and use speech-generating devices.

Their goals are to demonstrate:

  • Preliminary efficacy of the 10-week program using the Language for Learning curriculum to improve language in minimally verbal autistic children who use speech-generating devices. Language for Learning is a direct instruction curriculum commonly used in elementary school classrooms that will be modified for use of speech-generating devices
  • High caregiver and participant acceptability and satisfaction with the curriculum.

Dr. Shillingsburg is the director of the integrated Center for Autism Spectrum Disorders at the University of Nebraska Medical Center’s Munroe-Meyer Institute. She has expertise in developing and implementing interventions to address communication deficits associated with autism. She has provided direct intervention services and overseen the delivery of intervention services to hundreds of autistic children and adolescents with significant communication deficits for almost 20 years.

Dr. Frampton is an assistant professor with an emphasis in research at the University of Nebraska Omaha, with a particular interest in creating and implementing effective and efficient procedures to teach language skills to autistic individuals. She has worked clinically with autistic individuals for over 16 years and successfully implemented a variety of language interventions.

Methodology

The researchers plan to recruit 20 autistic children between the ages of 4 and 8 who use speech-generating devices. The children will receive three hours a week of instruction as usual or using Language for Learning. The program will run for 10 weeks at the Munroe-Meyer Institute at the University of Nebraska Medical Center.

Before beginning instruction, researchers will do a baseline assessment of participants’ skills and knowledge and conduct a post-instruction assessment within two weeks after the last week of instruction.

At the beginning of instruction, the instructor will administer a placement test to determine where to start in the curriculum. During the instruction, the instructor will adhere to the modified Language for Learning manual. For example, instead of just vocally modeling responses, the instructor will provide a vocal model and demonstrate target responses on the participant’s device. Instead of correcting errors by speaking, the instructor will provide gentle physical guidance to enter correct responses on the device.

For lessons that require a response using the device and a motor response, such as describing what they are touching and touching their head, the instructor will add supplementary prompts by also engaging in the motor response.

Each set of lessons includes a test, which the instructor will conduct according to the manual. If the participant correctly answers 80% or more of the questions, the next lesson will begin. If not, previous lessons will be re-taught.

Evaluation

The researchers will use the curriculum lesson tests, an improvement scale, vocabulary tests, a parent report of their child’s language abilities, and participation and happiness ratings by the children to provide data on the curriculum’s efficacy.

To ensure objectivity, an independent evaluator — a speech-language pathologist who does not know which participants are in the treatment as usual group and which in the Language for Learning group — will watch videos of pre- and post-trial lesson test performances and provide ratings of improvement.

Practical Relevance

The researchers outlined three primary results that would be relevant to the autism community:

  1. Support for a commercially available educational curriculum. Adoption of an evidence-based practice that is already established as highly effective, such as the Language for Learning curriculum, is a practical option for using speech-generating devices during educational instruction.
  2. Examples of how to measure and monitor assent to instructional sessions by minimally verbal autistic children. In the researchers’ previous pilot study, they monitored assent to instruction by offering participants the opportunity to leave the instructional space or pause instruction at any time. This information is included as an outcome measure because they strongly feel this is an important topic to report on with this vulnerable population.
  3. Resources to accelerate implementation of the Language for Learning curriculum with autistic children who use speech-generating devices. As part of this study, the researchers will expand a table of symbols/words required to participate in the lessons that they developed for the pilot study. In addition, they will expand recommendations for how to modify standard curriculum instructional practices for speech-generating devices, which they also developed in the pilot study.

Interventions that address cognitive and language skills are particularly important since these areas are associated with lower long-term symptom severity. Because direct instruction curricula have been shown to be effective with autistic people who are vocal communicators, it is critical to evaluate direct instruction with autistic individuals who use speech-generating devices.


Sherri Alms is the freelance editor of The OARacle, a role she took on in 2007. She has been a freelance writer and editor for more than 20 years.

The post Using Speech-Generating Devices to Teach Language Skills first appeared on Organization for Autism Research.