Using video modeling for generalizing toy play in children with autism.

The team used short videos to teach toy play to children with autism. Each video showed a peer playing with one toy in simple steps.

They ran a multiple-baseline design across toys. Kids watched, then got the same toy to play with. Later they tested if play moved to new, untrained toys.

Video modeling quickly boosted independent play and cut repetitive actions. Generalization only happened when the new toy looked or worked like the trained one.

For example, play spread from a red car to a blue car, but not to a toy piano.

Gutierrez et al. (2016) repeated the idea and found video prompting works even without voice-over. This backs up the core method while showing you can skip the audio track.

Haring (1985) got broad generalization across toy sets using stimulus-equivalence drills instead of videos. The 2007 study narrows the rule: generalization is free only within similar-looking toys when you use video modeling.

Lainé et al. (2011) extend video modeling to slow-motion clips for kids with severe autism. Together, the papers say video modeling works across severity levels and can be tweaked for better fit.

Pick training toys that share color, shape, or function with the ones you want the child to use later. One short video can unlock a whole class of similar toys and cut stereotypy at the same time.