Generalized imitation and the mand: inducing first instances of speech in young children with autism.

The team worked with preschoolers with autism who had never spoken.

Each session started with a quick burst of easy motor imitations.

Right after the child copied five hand or body movements, the adult gave a single spoken model like "ball."

They tracked whether the child then made any speech sounds during play.

Every child began to vocalize within the first few sessions.

The sounds turned into true mands, such as saying "juice" to get juice.

Three months later the kids were still using their new words without prompts.

Hansen et al. (2019) ran the same motor-before-vocal order and also saw more vocal copying.

Their goal was compliance, not first words, so the tactic works for both aims.

Sun et al. (2025) flipped the direction: the adult copied the child’s sound.

That method produced even bigger gains and gives you a second tool to use after the child starts talking.

Neimy et al. (2020) and Pelaez et al. (2011) showed that simple imitation works with babies, proving the idea stretches across ages.

You now have a two-step starter kit for non-vocal learners.

Begin with five fast motor imitations your client can already do.

Immediately model the target word and wait.

Once speech emerges, switch to contingent imitation to keep the momentum going.

The whole sequence takes under a minute and needs no extra materials.