Two different kinds of key peck in the pigeon: some properties of responses maintained by negative and positive response-reinforcer contingencies.

Duncan et al. (1972) worked with pigeons in a lab.

They watched the same key-peck move under two rules.

One rule gave food only if the bird pecked. The other gave food if the bird did NOT peck.

Birds gave short, quick pecks when food came for not pecking.

When food came for pecking, the same birds used longer, stronger pecks.

Same body part, same key, but two clear shapes.

Smith et al. (1975) later showed the short pecks cluster right after schedule changes. This supports the idea of two peck classes and helps explain behavioral contrast.

Gentry et al. (1980) used differential reinforcement to stretch or shrink peck length. Their work proves you can shape the long-peck class on purpose, backing the 1972 split.

Shimp (1971) found two separate time-gaps between pecks under mixed rules. The 1972 paper moves the same logic from timing to movement size.

The study reminds us that one movement can hide two operants. Check the contingency, not just the form. If a child’s hand-flap looks the same but serves different functions, treat each function as its own response class. When you shift reinforcement rules, watch for micro-changes in force or length—they may signal which class is active.