Analysis of factors that affect responding in a two-response chain in children with developmental disabilities.

Kuhn et al. (2006) worked with three children who had developmental delays. Each child already knew a two-step chain, like touching a card then handing it to the teacher.

The team tested three ways to weaken the chain: extinction (no more praise), satiation (give lots of the reinforcer first), and unchaining (stop reinforcing only the first step).

Extinction and satiation knocked down both steps of the chain. The kids stopped doing the whole sequence.

Unchaining was pickier. It cut the first step but the second step kept going strong. The chain broke at the front link only.

Cengher et al. (2020) and Matson et al. (2011) show the bright side of extinction. When they withheld reinforcement, new responses popped out—like full sentences or brand-new words. Contrucci’s work adds the caution: if you just unchain, the last response may never change.

Takashima et al. (1994) saw the same variability boost when toy play was put on extinction. Together these studies say extinction can both shrink old chains and sprout new forms, while unchaining only prunes the first leaf.

Pinkston et al. (2018) looked at what does NOT matter during extinction—response force made no difference. Contrucci shifts the lens to what DOES matter: which link you target decides how much of the chain survives.

If you want the whole chain gone, use extinction or satiation. If you only need the first step to drop out—say, touching a hot stove—try unchaining while you keep the safer second step. Check the data: if the final response sticks around, move to full extinction or add differential reinforcement of an alternative response.