Analysis of the reinforcement and extinction components in DRO contingencies with self-injury.

Cohen et al. (1993) asked a simple question. Does DRO still work if you keep giving the reinforcer when self-injury happens?

They worked with three adults who hit or bit themselves. Attention kept the behavior alive. The team tried three plans: DRO plus extinction, extinction alone, and DRO without extinction.

Each plan was run until the data were clear. Then they switched to the next one.

DRO plus extinction cut self-injury fast. Extinction alone did the same.

DRO without extinction failed for two of the three adults. Their self-injury stayed high. The third person improved, but only after the DRO interval was stretched very long.

Briggs et al. (2019) seems to disagree. They showed DRA without extinction can lower escape-maintained destruction. The key difference is function. Briggs worked with escape; L et al. worked with attention. When the reinforcer is attention, withholding it appears critical.

Brown et al. (2020) later confirmed the same pattern. DRA without extinction kept more target behavior during treatment, even though resurgence later looked the same.

McClannahan et al. (1990) used a token-based DRO and also saw fast suppression. Their study supports the idea that timing and reinforcer type matter, but it did not test the extinction piece directly.

If your functional analysis says attention keeps self-injury alive, do not rely on DRO alone. Pair it with extinction or use extinction by itself. Check early; if behavior stays high, add extinction before thinning the schedule. This small step can save weeks of stalled treatment.