A laboratory comparison of two variations of differential-reinforcement-of-low-rate procedures.

Jessel et al. (2014) tested two ways to run a DRL schedule in a lab. One group got spaced-responding DRL: each press had to wait a set time. The other group got full-session DRL: keep the total count under a limit for the whole period. All participants were neurotypical adults working for points.

The team tracked how many times each person pressed the button and whether they kept earning points or gave up.

Spaced-responding DRL kept the reward coming. People learned to wait and kept pressing at a nice slow pace. Full-session DRL backfired. Six out of ten people stopped pressing almost completely. The schedule turned into accidental extinction.

In plain words, the end-of-interval rule punished any early presses so hard that most folks just quit.

Becraft et al. (2018) ran the same two DRL types with preschoolers and added colored cards as signals. Kids in the signaled spaced-responding DRL did great; the full-session version still wobbled. Their data line up with Joshua et al. and show the risk is real outside the lab.

TCruz-Montecinos et al. (2024) moved DRL into a group of autistic boys. They used spaced-responding style and cut vocal disruptions every time. No extinction seen—another vote for the safer version.

Laposa et al. (2017) paired DRL with self-control lessons for detained teens. Full-session rules were not tested, but the omission-based DRL worked. Together the papers say: DRL works, but the spaced or omission shapes are kinder than the full-session cap.

If you use full-session DRL, watch for sudden silence. The client may have stopped talking, pressing, or walking because early responses burned the reward. Switch to spaced-responding DRL or add clear signals. You will keep the behavior alive while still trimming the rate.