Varying language register according to listener needs in speakers with autism spectrum disorder.

Volden et al. (2007) watched high-functioning autistic youth talk to different listeners. They wanted to see if the kids changed their language to fit each listener's needs. The team compared the autistic speakers to matched peers without autism.

Each child told a story to a little kid and then to an adult. The researchers scored how well the speaker shifted vocabulary, sentence length, and clarity.

The autistic group only partly adjusted their language. They used simpler words for the child, but they missed other cues. Controls tuned both words and tone far better.

Overall, the autistic speakers' register shifts were weaker and less complete.

Ikeda et al. (2024) extends this picture. They showed that autistic children can learn which register goes with which listener, but they do not grasp that the wrong tone can upset people. The 2024 study adds Williams syndrome and typical kids, confirming the social-emotional gap is autism-specific.

Järvinen-Pasley et al. (2008) looks at the flip side: understanding. They found autistic children pass single-word prosody tests yet fail sentence-level intonation tasks. Together with Joanne et al., this suggests trouble lives on both the speaking and listening sides of register.

Edgerton et al. (2017) offers hope. One child quickly learned to raise conversational volume when given voice-app feedback, a wrist cue, and praise. The result shows you can teach vocal adaptation, even if it does not come naturally.

Your client may sound polite but still miss the hidden rules of register. Add explicit lessons: model how tone changes with listener age, role-play little-kid vs. adult chats, and give immediate feedback. Track both clarity and emotional fit. The Joanne study warns that partial change is not enough; pair peer modeling with reinforcement so shifts become accurate and complete.