Case study: Modifying repetitive behavior in a polar bear (Ursus maritimus).

Abstract

Stereotypic pacing in zoo-housed carnivores has been intensively researched for decades, revealing that it is a subtly complex behavior with a highly variable etiology. While pacing can be an indicator of poor welfare, it can also be anticipatory in nature, or even have a completely different motivating operation. In this case study, the pacing of a zoo-housed polar bear, "Tundra," was maintained through inadvertent reinforcement. Baseline data revealed that in 86% of afternoon observations, Tundra paced when the door to shift him off-exhibit was opened. Upon changing the contingencies of reinforcement, there was a change in his behavior; initially with a shift in time, and then a shift in duration. Using a strategy of Differential Reinforcement of an Incompatible behavior, the pacing bout duration significantly decreased from a median of 4 min 13 s to 2 min 26 s. Due to the potential for repetitive walking to negatively influence captive carnivore welfare and the high individual variability of this behavior, investigation into the ontogeny and motivation for these behaviors should continue.