Assessment & Research

Vocabulary of Autistic Preschool Children With Limited Language: Alignment With Early Word Inventories.

Kong et al. (2026) · Autism research : official journal of the International Society for Autism Research 2026
★ The Verdict

Minimally verbal preschoolers with autism already say fringe words that core AAC lists skip, so program their personal favorites first.

✓ Read this if BCBAs who program AAC for preschoolers with autism.
✗ Skip if Practitioners working only with fluent speakers over age 6.

01Research in Context

01

What this study did

Kong et al. (2026) compared the real words spoken by minimally verbal autistic preschoolers with two standard lists. One list was the MCDI, a parent checklist of early words. The other was a core AAC word list, the 300 words most often put on talkers.

Team recorded each child's spontaneous speech during play, then checked which words also appear on those lists.

02

What they found

Only part of the children's vocabularies showed up on either list. MCDI items covered some words, but the core AAC list missed many that the kids actually said.

The missing words were mostly 'fringe' items tied to the child's own interests—trains, slime, specific foods—not the generic want, go, more found on core boards.

03

How this fits with other research

Jiang et al. (2026) extends this idea. After parents were trained on low-tech AAC, the same kind of preschoolers started to use new words for joint attention and comments, not just requests. The words they chose matched the fringe items Eunji found, showing the lists are useful once kids get the tools.

Abdi et al. (2023) goes further. A 16-session language package that taught those personally chosen fringe words produced very large vocabulary gains. Together the three studies form a chain: inventory reveals gaps → pick child-specific words → teach them → child talks more.

Su et al. (2018) shows the pattern crosses languages. Mandarin-speaking autistic preschoolers also knew words that sat outside standard inventories, giving the same lesson: look at what the child already says, then build on it.

04

Why it matters

When you program an AAC device, start with the child's own fringe words instead of filling the home page with generic core words. Watch play, note the oddball nouns and verbs the child already uses, and put those on buttons first. This small shift can make the system immediately relevant and may jump-start broader language growth.

Free CEUs

Want CEUs on This Topic?

The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.

Join Free →
→ Action — try this Monday

Spend 10 minutes in free play, write down every unique word the child says, then add those five fringe words to the device before core ones.

02At a glance

Intervention
not applicable
Design
other
Sample size
66
Population
autism spectrum disorder
Finding
not reported

03Original abstract

There is a critical need to understand the early vocabulary of young children with autism who have limited language, defined in this study as producing fewer than 20 different spontaneous and functional spoken or augmented words, to better inform educational targets and vocabulary selection for spoken as well as augmentative and alternative communication (AAC) interventions, particularly given the lack of evaluation tools designed for children with limited language. The spontaneous words and gestures produced by 66 preschoolers with autism (ages 3.5-5) during a natural language sample are compared with words in two early vocabulary tools including the MacArthur Communicative Development Inventories (MCDI) and a list of core words compiled from research studies of early AAC vocabulary. Participants' expressive words and gestures were coded from the transcripts of 20-min natural language samples. Forty-nine children (74.24%) used spoken words, gestures, or a combination of both, with six children (9.09%) communicating using a speech-generating device (SGD). Spoken words were primarily used for commenting, while gestures, especially pointing, were used for requesting. Although more than half of the unique words expressed by the children during the natural language sample overlapped with those in the MCDI, only 32% of unique words expressed by the children overlapped with Laubscher's and Light's core word lists, suggesting that young children with autism who have limited language may use more fringe words related to their personal interests or experiences. The study's limitations as well as implications for vocabulary selection for AAC systems and intervention goals are discussed.

Autism research : official journal of the International Society for Autism Research, 2026 · doi:10.1002/aur.70216