Date of Award

Winter 1995

Project Type


Program or Major


Degree Name

Doctor of Philosophy

First Advisor

William M Baum


The distribution of behavior by organisms in choice situations is of long-standing interest to psychologists. The generalized matching relation accurately predicts choice between concurrent variable-interval schedules of reinforcement. Researchers have assumed, on weak grounds, that the effect of the changeover requirement on sensitivity to reinforcement--the exponent in the generalized matching equation--was consistent. This experiment considered the effects of the changeover requirement by parametrically manipulating the fixed-ratio schedule required to switch alternatives. Pigeons pecked either of two side-response keys in a standard three-key operant chamber for food, delivered according to independent variable-interval schedules. No changeover delay was used, instead completion of five fixed-ratio schedules (FR 0, 2, 6, 12, or 20) on the center-response key alternated the active side key. Five reinforcer ratio (1:1, 1:2, 2:1, 1:4, and 4:1) were paired with most FR schedules. A matching relation analysis indicated that for two pigeons response-allocation sensitivity generally overmatched for all but the FR 0 condition, which undermatched. The other two pigeons' sensitivity increased to overmatching when FR 12 was in force. Excepting FR 0 conditions, time-allocation sensitivity, on the other hand, decreased from extreme overmatching toward matching as the changeover requirement increased. Reliable changes in response rates to the two alternatives account for the results.

A positive relation between the conditional probability of switching and run length is reported. That is, the greater the number of consecutive pecks to an alternative, the greater the likelihood of switching. This result suggests that behavior is controlled in part by local reinforcement contingencies. I speculate that factors that increase visit duration may increase local control of switching.

The procedure encourages a foraging interpretation. The FR changeover requirement can be considered functionally equivalent to travel between patches. An analysis of visit measures supported earlier evidence that residence measures increase as travel between patches increases. These results together with the matching results suggest that behavior ecologists and operant psychologists are working on similar problems and the traditional tools of operant psychology can be used to simulate travel, an important component of foraging in the wild.