Date of Award


Degree Name

Doctor of Philosophy



First Advisor

Dr. Jack L. Michael

Second Advisor

Dr. Alan Poling

Third Advisor

Dr. Bradley Huitema


Previous research has indicated that both absolute response rate and choice behavior differs under concurrent chains requiring different terminal-link responses, even when the reinforcement schedules are nominally identical. To date, no studies have investigated responding under concurrent chains with unequal schedules and different topographies arranged in the terminal links. Moreover, although the delay-reduction hypothesis has been widely tested using concurrent chains in which the same response topography is required in all links, the generality of the delay-reduction hypothesis in describing performance when different terminal-link topographies are required has yet to be examined. The present study was designed to address both of these issues.

Six pigeons responded under concurrent-chains schedules of food delivery. The initial links consisted of concurrent variable-interval 60 s schedules for all subjects. For three subjects, the required initial-link response topography was key pecking; for three others, the required initial-link topography was treadle pressing. Pecking was always required in the terminal link of one chain and treadling always required in the opposing terminal link. All subjects were exposed to five pairs of terminal-link variable-interval schedules over ten conditions. The pairs of variable-interval schedules employed were: 6 s vs. 54 s, 18 s vs. 42 s, 30 s vs. 30 s, 18 s vs. 42 s, and 54 s vs. 6 s.

Comparing data from nominally identical terminal-link schedules revealed that all subjects generally responded faster to the chain requiring terminal-link pecking than to the chain requiring terminal-link treadling. Regressing obtained choice proportions on their predicted values indicated that choice behavior was not well described by the delay-reduction hypothesis. Instead, a relative insensitivity to time to reinforcement and a strong bias towards the alternative correlated with terminal-link pecking were found. Differences between scheduled and obtained reinforcement schedules were found to be a major contributor to all of the present findings. Without exception, the time to reinforcement for chains requiring terminal-link pecking was shorter than that obtained for terminal-link treadling chains. Differences between the response forces associated with pecking and treadling were also identified as possible factors influencing choice but, because of methodological constraints, the effects of this factor were not empirically evaluated.

Access Setting

Dissertation-Open Access