Abstract:
For the two experiments that were conducted for this thesis, ten pigeons were trained to peck response keys for food reinforcers on various concurrent variable interval schedules. The purpose of Experiment 1 was to investigate the effects of signalling reinforcer ratios on local preference changes. Each session consisted of seven components that were separated by a 10-s blackout period and lasted until subjects had obtained ten reinforcers. Across conditions, a number of different combinations of the following component reinforcer ratios were arranged in random order without replacement: 27:1, 9:1, 3:1, 1:1, 1:3, 1:9, and 1:27. Components were signalled by differential red/yellow flash durations on both response keys. Generally, there was strong control by the component signals in that preference in components was ordinally related to the arranged reinforcer ratios, and there was also evidence of peak shift. The contingency-discriminability model (Davison & Jenkins, 1985; Davison & Nevin, 1999) was fitted to the present data and provided a promising description of the interactions between stimulus and reinforcer effects. Eight free parameters were fitted to predict 49 different combinations of reinforcer ratios of each of 12 experimental conditions. The results further confirm the usefulness of the model in a variety of different experimental procedures. Experiment 2 investigated the effects of sequences of reinforcers on local preference in concurrent variable-interval schedules. On both alternatives, reinforcers were scheduled dependently by involving a conditional probability that a reinforcer would be arranged on an alternative given that the previous reinforcer was on the same alternative. Across conditions, this conditional probability was varied from 0 to 1, while the overall reinforcer ratio was always 1:1. Thus, in some conditions, reinforcer locations changed frequently, whereas in others there tended to be very long sequences of same-key reinforcers. Immediately after reinforcers, probability of staying and visit length at the just-reinforced alternative was strongly affected by the probabilities that the following reinforcer was going to be on the same alternative. This experiment identified sequences of same-key reinforcers and conditional probabilities of reinforcer locations as a powerful controlling variable in local changes in preference. Reinforcers can therefore function as discriminative stimuli for the location of subsequent reinforcers.