“Other” behavior and the <scp>DRO</scp>: The roles of extinction and reinforcement

Hangen and team compared two ways to cut problem behavior: DRO and plain extinction.

They worked with children and adults who had intellectual or developmental disabilities.

Each person got both treatments while the researchers counted the target behavior and any other behaviors that popped up.

DRO knocked down the target behavior, but it also made new "other" behaviors bloom in every participant.

Extinction gave mixed results: sometimes other behaviors rose, sometimes they did not.

The data back the idea that extinction makes behavior vary, and DRO can accidentally strengthen whatever is left.

Bensemann et al. (2015) ran a near-copy study and saw the same DRO side effect, so the finding is holding up across years.

Horner (1971) first showed DRO beats extinction alone for speed, but the 2020 data add a warning: faster suppression can feed untargeted responses.

Logue et al. (1986) found sensory extinction only cuts untargeted stereotypy when it looks like the treated behavior; Hangen’s results widen the lens—any response class can bloom if it escapes the DRO contingency.

If you use DRO, graph more than the problem behavior. Track anything the client can do instead—silly noises, tapping, new scripts. When other behaviors climb, tighten the DRO interval or add an alternative-behavior program so the child has a clear, reinforced path to success.