Positive reinforcement and the elimination of reinforced responses.

Pigeons pecked a key for food. The team set up two feeders. One gave food only if the bird paused pecking for a set time. The other dropped food on a fixed clock, no peck needed.

Both feeders gave the same total food. The question: does paying for 'not pecking' cut the peck rate more than free food alone?

Birds pecked less when silence bought the pellets. The 'differential-reinforcement-of-not-responding' schedule beat plain non-contingent food.

The drop was biggest in birds that already knew the drill.

Attwood et al. (1988) ran the same match-up twelve years later. They got the same loser: non-contingent food. Longer pauses still won, a clean conceptual replication.

Horner (1971) had already shown that free food alone lowers pecking. The 1976 paper tops that baseline by adding the 'no-peck' rule, pushing rates even lower.

Richardson (1973) looks like a mirror: DRL also cuts rate, but it pays the bird for waiting then pecking once. Here the bird is paid for never pecking during the window. Same toolbox, different wrench.

When you pick DRO over NCR, you are not just handing out free tokens—you are actively reinforcing the absence of the target. If your client already has some experience with reinforcement schedules, the pause you want may come faster under DRO than under a plain NCR clock. Try equating the reinforcer rate first; then let the contingency do the extra work.