A comparison of procedural variations in teaching behavior chains: manual guidance, trainer completion, and no completion of untrained steps.

The team taught daily-living chains to children with autism. They used forward chaining but changed what happened to the steps the child had not yet learned.

One condition gave light hand guidance through the missed step. A second condition had the trainer finish the step for the child. A third condition simply skipped the step and moved on.

An alternating-treatments design rotated the three styles across sessions. The goal was to see which route taught the full chain fastest.

All three styles worked; every child learned the chain. Manual guidance was the speediest for five of the seven kids.

Skipping the untrained steps saved the most total session time, even though it was not always the fastest to mastery.

Ding et al. (2017) extends this idea. They grouped three or four steps into mini-clusters and still used forward chaining. An adult with autism learned to cook full recipes and kept the skill for weeks.

Lambert et al. (2016) also extend the method. They paired forward chaining with basketball. A teen first mastered each ball skill alone, then linked them into full offense and defense sequences.

Hoch et al. (2007) looks like a clash but is not. Their video-only prompting failed for three adults until the trainer stepped in and finished missed steps. That "trainer completion" matches the least efficient arm here, yet both studies agree: having the adult do the step can rescue learning when solo methods stall.

You can relax about unfinished steps. If session time is tight, skip the untrained steps and keep the flow. If you want the cleanest learning path, add light manual guidance. Either way the child still masters the chain, so pick the style that fits your schedule and your learner’s tolerance for help.