Using a Lag Schedule of Reinforcement to Increase Response Variability in Children With Autism Spectrum Disorders

Olin et al. (2020) tested Lag 1 and Lag 2 schedules on social questions for children with autism. They added echoic prompts and small picture cards to help the kids think of new answers.

The team used a changing-criterion design. Each session the child had to give a different answer before earning praise or a token.

The kids quickly started giving fresh, non-repeated answers. The new answers stayed high even four weeks later.

No rote scripts were needed after the first few trials. The children answered on their own and smiled more during turn-taking.

Poppes et al. (2016) and Gardner et al. (2009) already showed that lag schedules boost verbal variety in autism. Olin adds echoic prompts and visuals, making the jump faster and friendlier.

Cohrs et al. (2017) used Lag 2 and Lag 4 after teaching three examples in group social-skills class. Olin skips the long exemplar training and still gets quick gains, showing the prompt package can stand alone.

Parilla et al. (2023) later moved the same idea into teens’ homes with simple scripted prompts. The chain shows the method keeps working as kids grow and as sessions move to the kitchen table.

You can run this in any clinic or home room. Pick a social question like “What did you do last night?” Set a Lag 1 rule: the child must say something new to get a point. Give a quick echoic hint or show a picture if they stall. Most kids need only one or two sessions to shift from “played iPad” to “built a castle, fed the dog, watched space videos.” Flexible answers help peers stay interested and cut down on robotic conversation, a common parent worry.

A lag schedule of reinforcement delivers reinforcement only when the current response differs from the previous responses. Under a Lag 1 schedule a response is reinforced if it differs from the one immediately before it; under Lag 2 it must differ from the previous two.

The purpose is to build response variability, which is useful for teaching flexible language, play, and social answers rather than rote, repeated responses. Variability itself becomes the reinforced dimension of behavior.

Lag schedules sit alongside standard reinforcement schedules but target novelty instead of rate. They are especially relevant when rote responding limits generalization.

An 11-year-old girl with autism gave little variability at baseline, repeating rote answers to social questions. Under Lag 1 and Lag 2 schedules, with echoic prompts, visual aids, and error correction, varied appropriate responding increased.

She emitted 11 novel prompted responses and 13 spontaneous responses, and retained the skill at a four-week maintenance probe, supporting lag schedules as a tool for durable response variability.