On the dynamics of behavior allocation between simultaneously and successively available reinforcer sources.

The assumption of constant rates of extraneous reinforcement during changes in explicit reinforcer rates made by Herrnstein (1970) must logically be incorrect. An alternative theory that suggests constant extraneous-schedule values, rather than reinforcer rates, is suggested. In a test of this theory, six pigeons were trained on multiple variable-interval schedules with either no alternative reinforcers available, or with various variable-ratio or variable-interval schedules of reinforcement arranged concurrently, in 26 experimental conditions. A modification of McLean and White's (1983) multiple-schedule model that assumed a constant extraneous variable-ratio schedule fitted the obtained data, and those from two previous experiments, well. However, the model is dynamic, and allows the subject a range of adaptations to variations in arranged reinforcer rates. It is argued that this range of adaptations is both appropriate and necessary.