Varying reinforcer dimensions during differential reinforcement without extinction: A translational model

The team tested a lab model of DRA without extinction. They kept the problem response on its old payoff while making the new response pay better, faster, or tastier.

College students pressed buttons for points. One button still gave small, delayed, bland points. The other gave bigger, sooner, or tastier points. The researchers watched which button the students chose.

All upgrades cut problem button presses. Upgrading treat quality gave the biggest drop. Bigger piles and faster delivery helped too, but quality ruled.

When the good button paid favorite snacks, students almost forgot the old one. The study shows you can shrink problem behavior without turning the old payoff off.

Kittler et al. (2004) used extinction plus lean DRA and still got fast results. The new study says you can skip extinction if you juice up the alternative reward instead.

Drifke et al. (2020) later moved the same DRA logic into FCT with kids who have ID. They kept communication high during delay fading by requiring the alternative response, just like the lab model.

Stevens et al. (2018) stretched the idea to children with autism. A fixed-lean schedule kept problem behavior low, matching the magnitude tweak in the lab. Together the papers show the trick works from college kids to clinical caseloads.

Sometimes you can’t block the old reinforcer—safety, staffing, or consent says no. This paper gives you a backup plan: make the good behavior pay better, not zero. Start with better quality treats, then add size or speed if needed. You get fewer problem responses without extinction bursts or ethical headaches.