Auto-shaping of the pigeon's key-peck.

The team put hungry pigeons in a box with a small round key.

Every so often the key lit up for eight seconds, then food appeared no matter what the bird did.

The birds never had to peck. Light and food were simply paired.

After about the pairings every bird began to peck the lit key.

Birds that saw random light or random food never pecked.

The response grew even though pecking did not produce extra food.

Hartmann et al. (1979) later showed the same pecking keeps going even when it cancels food.

This extends the 1968 finding: Pavlovian control is so strong that birds keep pecking even when it costs them a meal.

Schwartz et al. (1971) got much faster shaping in rats, but the response was lever-pressing and the food moved the lever.

The rat study is a conceptual replication: both show new moves appearing quickly with little trainer effort, just different species and topographies.

Shimp et al. (1974) kept the key-peck but swapped food for shock-avoidance.

Pecking still held steady, showing the same response can be maintained by very different consequences.

Your non-contingent reinforcement might accidentally create new behaviors.

If you pair a card-flash with candy while teaching a child, the child may start tapping the card even though you never asked for that tap.

Watch for these surprise moves and decide quickly: shape them, block them, or change the pairing.