Choice during a three-alternative rapidly changing procedure with varying reinforcement rates

Abstract

Six pigeons were used in a rapidly changing procedure to identify the local processes operating during choice behaviour. The procedure used three alternatives, and in two conditions, the overall reinforcer rate was varied within an experiment session. This made the procedure unusual but enabled tests that are otherwise not possible. Due to this, two types of tests were conducted. One type was to demonstrate that choice during the procedure was typical of other rapidly changing procedures. Results indicate that is was, but differed when the reinforcer rate was variable. The other type of tests examined properties of reinforcement and matching behaviour. One test looked at the information provided by reinforcement, but was inconclusive. Another test focused on the constant-ratio rule, one of the assumptions of the generalised matching equation. Results indicated that the rule does not hold when varying the reinforcer rate. A third test focused on the role of the reinforcer rate during matching procedures. Altering the rate caused subjects to shift their responses over time. In addition, the reinforcer rate altered the data collected between reinforcers; this has implications for experiments that alter the overall reinforcer rate.