Training natural communication partners how to model AAC

Model, model, model! We all know how important and effective AAC modeling can be (see here and here, for example)—however, modeling is only as good as the partners who are implementing it. If you’re working with kids who use AAC, chances are there are communication partners who need guidance in how to model, and that’s no simple task. If you’re thinking “I agree, but HOW do I teach the partners?”, this review is for you!

The authors of this study gathered 29 studies in which more than 250 communication partners (including peers, teachers, paraprofessionals, parents, and other adults) implemented modeling strategies across various settings. Although they looked at a handful of research questions, the most clinically relevant questions were: How were the communication partners trained and what did they have to say about the training they received?

The most common training strategies were:

  • orally sharing information

  • modeling the strategies, and

  • allowing the partners to practice in controlled settings (role plays), or with a child, while providing feedback


Overall, partners rated instruction as worth the time, easy to understand, practical, and transferable to other children. Some additionally offered the suggestion to provide more direction on how to model during a child’s regularly occurring activities (something to consider when you are providing training).

Seems pretty straightforward, right? We train the partners using those strategies and then off they go? Not so fast. The authors found that most communication partners also benefited from simultaneous support while learning to model. So after you train the partners, it’s important that you stick around to offer coaching and consultation as necessary.

If this seems daunting (how can I possibly fit this into my already jam-packed day!?), it’s important to remember that teaching communication partners can drastically improve the reach of our interventions—the amount of time we spend with our students is so limited compared to their interactions with natural communication partners.

If partner instruction is something you’d like to improve, be sure to check out the full article (specifically Table 2) for a list of the included studies and the training strategies used in each.


Biggs, E. E., Carter, E. W., & Gilson, C. B. (2019). A scoping review of the involvement of children's communication partners in aided augmentative and alternative communication modeling interventions. American Journal of Speech–Language Pathology. doi: 10.1044/2018_AJSLP-18-0024

Measuring the earliest forms of communication

As you may have realized (with frustration!) by now, we have limited options for evaluating the expressive communication skills of children who are minimally verbal. Enter: the Communication Complexity Scale (CCS), designed to measure just that. Prior papers have described the development of the CCS and determined its validity and reliability, but in this study, we get to see it in action with a peer-mediated intervention.

First, a little bit about the tool. It’s a coding scale—not a standardized assessment—that can be used during observations. Because prelinguistic communication skills often take time to develop with this population, this tool helps us think about all the incremental steps along the way and accounts for the variety of communicative modes the children might use. It’s a 12-point scale following this pattern:


The researchers found that the CCS could measure improvement in overall communication complexity and behavior regulation for preschoolers with autism after a peer-mediated intervention (the same one we reviewed here!).

So far in the research, the CCS has only been used during structured tasks meant to elicit communicative responses (see the supplemental material), such as holding a clear bag with toys where the child can see it, but can’t access it independently. We know it's crucial to observe our students in natural communication opportunities, though, so we'd have to be a little flexible in using the CCS during unstructured observations. The scale could definitely be useful when describing communication behaviors during evaluations or when monitoring progress. Wouldn’t it be much more helpful to say “The child consistently stopped moving (i.e. changed her behavior) in response to the wind-up toy stopping” instead of “The child was not observed to demonstrate joint attention”? Using the CCS, we have new ways of describing those “small” behaviors that really aren’t small at all!

NOTE: This study crosses over our Early Intervention vs. Preschool cut-offs, with kids from 2 to 5 years old. So for those of you who also read the Early Intervention section, we’ll publish this there next month! Just giving you the heads-up so you don’t feel like it’s Groundhog Day :)

Find links to the scale and score sheets, here.

Thiemann-Bourque, K. S., Brady, N., & Hoffman, L. (2018). Application of the communication complexity scale in peer and adult assessment contexts for preschoolers with autism spectrum disorders. American Journal of Speech-Language Pathology. doi:10.1044/2018_AJSLP-18-0054

Teaching peers to understand how their classmates communicate

If you’ve been following TISLP for a while, you’ve probably seen reviews about incorporating preschool peers into therapy with children who use AAC. Do you want to try involving peers, but work with older kids with complex needs? Here’s a way to start: In this study, typically-developing teens learned to interpret their peers’ presymbolic communication behaviors—after a training lasting only 15 minutes!

Wait, what do we mean by presymbolic behaviors? Think facial expressions, eye gaze, vocalizations, and body movements. These behaviors can be challenging (even for us as SLPs!) to decipher. It makes sense that middle-schoolers might struggle with them, too. If typically-developing peers misinterpret their fellow student’s behaviors, it would be pretty difficult for them to engage in meaningful interactions—and isn’t that the end goal? 


The researchers worked with families and teachers to define the presymbolic behaviors unique to three different students. They captured each student’s two most frequent behaviors (and other non-communicative behaviors) in very short video clips. Peers watched a series of the videos, tried interpreting the behaviors, and got corrective feedback when they misunderstood. Peers who completed the training were significantly better at understanding behaviors, and classroom staff noticed that the quality of communication between the students improved, too

Want to hit the ground running? The clinical implications section outlines some considerations for how you could choose (and define) presymbolic behaviors to target. Keep in mind that the study was just trying out this approach to peer training. So while we can’t yet say for sure that these skills would transfer to authentic interactions, this is a starting point in helping typically-developing peers better understand their classmates with multiple disabilities.

Holyfield, C., Light, J., Drager, K., McNaughton, D., & Gormley, J. (2018). Effect of an AAC partner training using video on peers’ interpretation of the behaviors of presymbolic middle-schoolers with multiple disabilities*. Augmentative and Alternative Communication. Advance online publication. doi:10.1080/07434618.2018.1508306

Preschool peer-mediated video modeling

Much of the early childhood school day involves play or peer interactions. We’ve talked before about peer-mediated interventions for preschoolers with autism (e.g., here, here, and here). It’s no surprise to SLPs that there’s also a growing evidence base for using video modeling with this population. Since preschoolers with autism may need structured opportunities to learn and practice social skills, you could use video modeling or you could include a typically developing peer in intervention, or… you could use both (aka: joint video modeling)!

Including a peer is what makes joint video modeling different. Since we’re talking about preschoolers here, the authors needed to do some structured training with those peers. First, the preschool director and head teacher selected “play partners”—peers with age appropriate social skills and good interactions with classmates with autism. The authors showed the play partner a picture of the other child, explained the difficulties the child had during play, and read a book about understanding differences. Then, through role play and video modeling, the play partners learned how to initiate during play. 


Now on to the joint video modeling intervention: The four-year-old dyads jointly (sitting next to each other listening via split headphones) watched a 30-second video model twice on an iPad. Cool note about the video models: Although they showed adults’ hands manipulating toys, they were based on typically developing preschoolers’ play with those same toys. So if you need some inspiration, you may need to look no further than peers in the classroom! 

After intervention, the children with autism used more scripted verbalizations (from the videos) and unscripted verbalizations (even better!) during pretend play—and generalized the skills with other peers even after the videos were taken away. We still need more information on how the skills were maintained, and to help explain the individual differences within the participants’ performance, but joint video modeling is definitely showing promise.


Dueñas, A. D., Plavnick, J. B., & Bak, M. Y. S. (2018). Effects of Joint Video Modeling on Unscripted Play Behavior of Children with Autism Spectrum Disorder. Journal of Autism and Developmental Disorders. doi:10.1007/s10803-018-3719-2

Including preschool peers in AAC intervention

When talking about preschoolers who require AAC, we all know the importance of communication partners in supporting interactions. As SLPs, we model, we wait, we do all the right things to promote communication. But is there more we can do?  How can we make an even bigger impact? How about training peers to be responsive communication partners?

The authors of this study included preschool peers in their SGD intervention for a large group of nonverbal preschoolers with ASD. Half of the peers were taught how to be responsive communication partners, and half of the peers were not. The trained peers were taught a modified version of Stay-Play-Talk, an evidence-based preschool program (Goldstein et al., 1997), which included the following steps:

  • Stay with your friend (sit close)
  • Play with your friend (share and take turns)
  • Talk with your friend (look at your buddy, listen, push picture, say words out loud)
  • Get attention and Hold and wait

After the peers were trained, they interacted and played with their classmates with ASD with an iPad programmed with relevant vocabulary to match the play context. There was also always a trained staff member guiding the interaction. The untrained peers played with their classmates with ASD as usual. Again, a staff member was present, doing what they would normally do during the activity. 

Following training, the children with ASD who played with the trained peers communicated more during the interactions and showed more balanced responses and initiations (they contributed more equally to the interaction) than the children who played with the untrained peers.  They also generalized their communication improvements to new settings with trained peers and to familiar settings with untrained peers.  

The really cool part about the peer training?  It only took a handful of short sessions over 2–3 days (80 minutes total)! So the next time you’re working with preschoolers with ASD who use AAC, let’s not forget about the invaluable communication partners sitting right beside you in the classroom.  Peers without disabilities can be taught to interact effectively with children using SGDs, and in the end, this could have implications that even go beyond communication (think: social participation and friendship development).

Note: We’ve seen research on Stay-Play-Talk before! See here.


Thiemann-Bourque, K., Feldmiller, S., Hoffman, L., & Johner, S. (2018). Incorporating a peer-mediated approach into speech-generating device intervention: Effects on communication of preschoolers with autism spectrum disorder. Journal of Speech, Language, and Hearing Research. Advance online publication. doi:10.1044/2018_JSLHR-L-17-0424

Peer mentors and PEERS

These authors added a new spin to PEERS, a well-researched social skills and friendship curriculum, by including peer mentors. First, note that PEERS is designed for adolescents with autism who do not have intellectual disabilities. The curriculum manual contains 90-minute weekly interventions for 14 weeks, but there is also a version structured for school delivery. Targeted social skills include choosing appropriate friends, how to handle teasing, bullying, gossip, and more.

Participants in this study (high school students with verbal IQs of 70 or above) were placed in one of three groups:

  1. Followed the traditional PEERS curriculum (only included adolescents with autism)
  2. Followed PEERS and incorporated peer mentors (PEERS with Peers)
  3. Control group (Followed the traditional PEERS curriculum, with a delayed start)

Participants in groups 1 and 2 increased social skills knowledge and decreased loneliness. Students in group 1 increased get-togethers with peers compared with those in the control group. Students with autism in the PEERS with Peers group had significant improvements in social skills and problem behaviors, as reported by their parents, showing a “modest advantage” over the traditional model. Many of these skills were maintained four months later.


The PEERS curriculum has a growing evidence base, but what’s really exciting about this study is that the researchers purposely had clinicians lead the intervention, paying attention to the research to practice gap in this area. Saving some of the best news for last, training to implement PEERS was manageable—staff participated in a 3-day initial training and 1-day refresher, and peer mentors received only 1 hour of training, along with their parents. All in all, this curriculum is worth checking out if you serve adolescents with autism who do not have ID.


Matthews, N. L., Orr, B. C., Warriner, K., DeCarlo, M., Sorensen, M., Laflin, J., & Smith, C. J. (2018). Exploring the Effectiveness of a Peer-Mediated Model of the PEERS Curriculum: A Pilot Randomized Control Trial. Journal of Autism and Developmental Disorders. Advance online publication. doi:10.1007/s10803-018-3504-2

Peers can help passive communicators learn conversation skills

Raise your hand if you know a student like this: ASD diagnosis, doesn’t initiate with peers, can’t keep a conversation going, always seen sitting alone or with other students with disabilities. Do you have an effective way to help?

The authors of this study crafted a peer-mediated intervention for a small group of high-school aged “passive communicators” with ASD. They picked four “focus students” who struggled with initiating and maintaining conversations, along with 16 peer mentors. Note that all the kids involved actually wanted to participate. Four mentors were assigned to each focus student, based on who shared the same lunch period.

The article goes into lots of detail about how the whole intervention was designed and structured (enough that you could pull off something similar!), but here are some key points:

  • Mentors joined the focus student for lunch, two at a time
  • Mentors learned specific strategies to promote initiation, conversation maintenance, and follow up questions
  • Focus students received some direct instruction as well, and brainstormed potential conversation topics daily with a teacher
  • Both focus students and peers used cue cards as supports
  • Mentors got regular feedback/check-in sessions with staff throughout the process
  • After a pre-training baseline period, intervention lasted 12–16 weeks

So how did the focus students do, by the end of the intervention? They participated in more (and longer) conversations, initiated more often, and contributed more follow-up questions and comments (which weren’t directly taught!). In some cases, these gains generalized to new peers (both trained and untrained). Focus students and peers rated the process highly. And, if that wasn’t enough for you, teachers (who didn’t know about the intervention) noticed improvements in conversational skills. Even considering the small sample size, this is a really impressive result.

Okay, real talk. This model of intervention is a lot of work. You have to recruit and train peers, juggle schedules and logistics, and find a way to monitor the intervention and give/receive feedback. But as the authors point out, you get the advantage that “training can occur whenever peers and focus students are available, and intervention can be applied whenever and wherever conversation is appropriate.” The students also get the benefits of learning in a natural, inclusive (dare I say, least restrictive?) setting, and a chance to form meaningful relationships with non-disabled peers. This is big stuff. With some planning, creative thinking, and great team collaboration, a determined SLP could get it done. In many high schools, there are classes or clubs relating to community service, disability issues, inclusion, etc. These are great places to recruit and train peers mentors. And hey, if there isn’t one at your school, think about starting one! Remember, just sitting with peers isn’t enough, since “[peer-mediated intervention] must include explicit intervention strategies designed to elicit targeted outcomes.”

Bambara, L. M., Cole, C. L., Chovanes, J., Telesford, A., Thomas, A., Tsai, S.-C.,  Bilgili, I. (2018). Improving the assertive conversational skills of adolescents with autism spectrum disorder in a natural context. Research in Autism Spectrum Disorders, 48, 1–16. doi: 10.1016/j.rasd.2018.01.002.

And more...

  • Brock et al. found that a Pivotal Response Training program facilitated by school staff and implemented by peers increased peer interaction and appropriate play in school-age children with autism.
  • Kelley et al. add to the evidence base for the Story Friends curriculum, which targets vocabulary and reading comprehension for 4- to 6-year-olds with limited oral language, and which can be used as a Tier 2 intervention in an RtI framework. The researchers found that by analyzing scores from Unit 1 (roughly one month of instruction), they could predict which children would respond to the program as a whole and who would require more specialized, Tier 3 interventions.
  • Kuster et al. found that the Dyslexie font did not improve children’s reading speed or accuracy (whether or not they had dyslexia), and that the readers didn’t prefer it to traditional fonts.
  • Na et al. introduce an interview-based assessment tool to help SLPs gather culturally-relevant data about the emotional competence of young AAC users, while learning about the family’s priorities and values around expressing emotions. A small pilot study showed that American and Korean-American respondents gave qualitatively different responses to the questions, as predicted.

Brock, M. E., Dueker, S. A. & Barczak, M. A. Brief report: Improving social outcomes for students with autism at recess through peer-mediated pivotal response training. Journal of Autism and Developmental Disorders. Advance online publication. doi: 10.1007/s10803-017-3435-3.

Kelley, E., Leary, E., & Goldstein, H. (2017). Predicting response to treatment in a tier 2 supplemental vocabulary intervention. Journal of Speech, Language, and Hearing Research. Advance online publication. doi:10.1044/2017_JSLHR-L-16-0399.

Kuster, S.M., van Weerdenburg, M., Gompel, M. et al. (2017). Dyslexie font does not benefit reading in children with or without dyslexia. Annals of Dyslexia. Advance online publication. doi: 10.1007/s11881-017-0154-6.

Na, J. Y., Wilkinson, K., & Liang, J. (2017). Early Development of Emotional Competence (EDEC) Assessment Tool for Children With Complex Communication Needs: Development and Evidence. American Journal of Speech-Language Pathology. Advance online publication. doi: 10.1044/2017_AJSLP-16-0058.

Teaching peers to communicate with children with autism using AAC

When supporting children with ASD who use SGDs (speech generating devices), have you considered incorporating their peers into social interactions? This study reveals that training typically-developing peers can lead to reciprocal communication in the classroom.

Thiemann-Bourque et al 2017.png

We’ve talked before about teaching peers of preschoolers with ASD to use PECS. In this study, three typically-developing peers are taught to communicate with three minimally verbal children with ASD in the preschool classroom. A GoTalk 4+, a four-button SGD, was available during the communication exchanges. The preschoolers with ASD had not used the SGD prior to participating in the study. 

Using a “Stay-Play-Talk” social intervention, peers learned to 1) stay at the table with the child with ASD, 2) play by sharing toys or taking turns, and 3) talk to their friend, using the SGD as an option for communicating. The researchers also describe teaching the peers “Ways to be a Good Buddy” and outline the training steps used during three 30-minute training sessions.

In this study, each peer was paired with one child with autism during a six-minute classroom activity (centers, snack time, or requesting a toy) twice per week. A total of 15 to 18 activities were observed over 10 weeks.  A researcher led a social interaction between the peer and child with ASD and showed them a Stay-Play-Talk visual support, prior to stepping back from the pair in order to observe. During the observation, the implementer used least-to-most prompting to encourage interaction whenever 30 seconds without communication occurred.

As a result of the intervention, the peers successfully learned to use the SGD with their friend with ASD. More importantly, the children with ASD demonstrated more initiations and spontaneous communication. The results were observed across contexts, although snack and requesting toys were more successful than centers.

Because all children had access to an SGD during the intervention, more balanced and natural communication occurred. This is in contrast to the authors’ findings when teaching PECS, in which the peer takes on the “responder” role by taking a picture symbol from their friend. More research is needed to understand how social contexts influence peer interaction and to examine across a variety of communication skills.

Thiemann-Bourque, K., McGuff, S., & Goldstein, H. (2017). Training peer partners to use a speech-generating device with classmates with Autism Spectrum Disorder: Exploring communication outcomes across preschool contexts. Journal of Speech, Language, and Hearing Research. doi:10.1044/2017_JSLHR-L-17-0049