Spontaneous Play Profiles in Mandarin-Speaking Preschool Children With Autism, Developmental Delay, and Typical Development: A Fine-Grained Comparative Analysis.

Researchers watched 90 Mandarin-speaking preschoolers play alone for 15 minutes. Thirty kids had autism, 30 had general developmental delay, and 30 were typically developing.

Trainers coded every play move using a 7-level scale. They counted how long each child stayed at each level and how many different Varied Action Sequences each child showed.

Autistic kids played for shorter time and reached lower play levels than both other groups. The gap was largest at the top levels like pretend play.

The biggest red flag was Varied Action Sequences. Autistic children produced fewer and less creative action chains than both DD and TD peers.

Saban-Bezalel (2025) saw the same pattern in communication: DD kids looked worse when compared to same-age TD peers, but similar when matched for language level. Tongxin now shows play follows the same rule—benchmark DD kids against language peers, not age peers.

Ozonoff et al. (2008) found that babies later diagnosed with autism spin and rotate objects at 12 months. Tongxin’s preschoolers did not show more spinning; instead they showed less varied play. The two studies line up—early odd object use fades and lack of flexible play takes over.

Chiang (2008) showed that half of non-verbal autistic preschoolers use challenging behavior to request or reject. Tongxin adds a new piece: when these kids play alone, their actions are not just rare—they are repetitive and simple. Watch both the negative behavior and the missing play.

During assessment, count Varied Action Sequences for two minutes. Fewer than three different chains in Mandarin-speaking three-year-olds signals risk for autism over DD. For DD kids, compare play to language-matched peers, not age peers, to set fair goals. Add play-level targets after mand training—simple functional play first, then chained actions, then pretend.