Abstract
In Experiment I, three pigeons were shaped to use tokens by the method of successive approximations. They quickly learned to obtain food by dropping tokens into an opening on the floor using their beak. In Experiment II, a pigeon was trained to peck a key on fixed-ratio schedules of token reinforcement in which tokens were immediately exchangeable for food. As the number of responses required for a token was increased from one to 20,post-reinforcement pause was prolonged while running rate of key pecking did not change.
