Teaching Sentence Writing to Students with Autism and Complex Communication Needs using Matrix Training, Sentence Frames, and Speech-Generating Devices

Approximately 30% of individuals with autism have complex communication needs (CCN). These individuals are unable to use vocal speech as their primary form of language and typically require support across several areas of communication such as comprehension, pragmatics, phonology, semantics, and syntax (Ganz et al., 2022; Reichle, 2019). Researchers have found that communication skills can greatly impact academic, behavioral, social, and postsecondary outcomes (Carter et al., 2012; Chiang, 2008; Matson et al., 2013; Park et al., 2012; Pillay & Bronlow, 2017). Fortunately, augmentative and alternative communication (AAC) has been effectively used to increase communication for individuals with intellectual and developmental disabilities (IDD; Crowe et al., 2022). Most often, individuals with autism are only taught to request using single words or short phrases using AAC devices (Ganz et al., 2017; Muharib et al., 2018; Tincani et al., 2020). Another way to expand communication through AAC is to teach sentence structure. Researchers have used an intervention package consisting of response prompting, sentence frames, and technology like AAC to teach students with autism and CCN to construct sentences (Pennington et al., 2021; Pennington, Flick, et al. 2018; Pennington, Foreman, et al. 2018; Pennington & Rockhold, 2018). Additionally, matrix training has been used as a generative framework to increase language for individuals with autism who use vocal speech (Frampton et al., 2016, 2019; Jimenez-Gomez et al., 2019; Kohler & Malott, 2014) and AAC (Marya et al., 2021; Naoi et al., 2006; Nigam et al., 2006; Tönsing et al., 2014). This study examined the effects of matrix training, response prompting, and sentence frames on sentence writing for four students, ages 10–18 with ASD and CCN in a specialized private school located in the southeastern United States. Three teachers, ages 23–46 served as the interventionists in the study. A series of A-B designs with modifications was used to examine the effects of the intervention package on the percentage of trained and untrained correct sentences, percentage of subject-verb combinations, and the percentage of correct word selections. Teachers presented photos of subject-verb combinations for students to write about using pre-programmed arrays with words and symbol supports on speech-generating devices. Overall, results indicated that across all interventions, there were no effects on the percentage of trained and untrained correct sentences and subject-verb combinations for all participants. Two students, however, increased their percentage of correct word selections. Overall, teachers found the intervention acceptable and beneficial for students in the classroom. Furthermore, three of four students preferred this writing intervention over their typical writing instruction in the classroom. Implications of this study provide several considerations for practitioners who would like to use matrix training to teach subject-verb combinations and/or sentence writing with students who have autism and CCN.