A comparison of several procedures for eliminating behavior.

The team worked with pigeons on a DRL schedule. Birds had to wait a set time between pecks to earn food.

They tested four ways to stop short, off-time pecks. The birds got either mild electric shock, food satiation, plain extinction, or a change in the key color.

Each bird tried all four methods. The researchers watched which method cut the off-time pecks fastest and kept the good, on-time pecks high.

Punishment won. Shock quickly stopped the short pecks and the birds kept the reinforced long pauses.

Satiation, extinction, and color change all worked slower. They also lowered the reinforced pecks, so the birds earned less food.

With punishment, DRL efficiency rose. The birds waited the right time and still got most reinforcers.

STEBBINROSS et al. (1962) showed punishment plus extinction gives strong control, but add reinforcement and control weakens. The 1963 study used punishment alone, so it kept tight suppression.

Harris et al. (1973) later repeated the idea with rats. Shock cut lever presses while licking stayed high, matching the pigeon result: punishment hits only the targeted response.

Schroeder et al. (1969) moved the test to adult humans. VI shock also cut button pressing, proving the effect crosses species and schedules.

Dunham et al. (1969) warned of side effects: punishing key pecks made off-key pecks rise. The 1963 study did not see this contrast, likely because DRL already spaced responses far apart.

If you need to erase quick, impulsive responses that spoil a timing-based program, mild punishment can act like a scalpel. Pair it with clear DRL rules and rich reinforcement for the correct wait. Watch for contrast in other topographies, but expect fast, clean suppression of the off-time bursts.