Search

Non-human primates occasionally exhibit behaviours thought to occur only in captivity that are considered abnormal. In particular, hair-plucking behaviour occurs across many species of mammals and birds. This study was the first to assess the phenomenology, demography and aetiology of this behaviour in captive gorillas across the Association of Zoos and Aquariums (AZA) population. A survey was sent to 52 AZA institutions housing western lowland gorillas. Hair-plucking behaviour occurred in 15% of the surveyed population with 62% of institutions housing a hair plucker. Individuals were most likely to self-pluck using their fingers. Individuals that were exposed as youngsters to a hair-plucking group member were significantly more likely to develop the behaviour themselves. There was also a trend toward solitary individuals being more likely to perform this behaviour. Future research needs include identifying hormonal correlates to this behaviour, understanding its relationship to acute or chronic stressors, and examining what causes this behaviour to fluctuate in frequency once it has developed.

One way to address the problem of small sample size in zoo research is to conduct a study involving multiple institutions. However, when conducting behavioral research, it can be difficult for one investigator to travel to each institution or for staff at individual institutions to find time to collect behavioral data. This study validates keeper assessments of western lowland gorilla ( Gorilla gorilla gorilla ) activity levels against behavioral data collected systematically in the zoo setting. Our purpose in validating a keeper rating of inactivity was to increase efficiency of behavioral data collection and maximize the amount of activity information for a larger study on gorilla health and behavior. A questionnaire was sent to 15 Association of Zoos and Aquariums (AZA) institutions housing 63 gorillas (37 male, 26 female) gorillas. Behavioral data for the project were collected by staff at each institution and/or by the principal invesitgator (PI) using group scan sampling. Every data collector was reliability tested by video until they gained >90% agreement with the PI. Mean percent of time spent in a behavior was calculated from the scan data. In this case, the focus was on the time spent immobile (i.e., stationary) and time spent idle (i.e., at rest, performing no otherwise defined active behaviors). We examined the correlation between the keepers' ratings and time spent immobile and idle as calculated from the behavioral data. Behavioral data on individuals that spent more time immobile were significantly correlated with an increasingly inactive ranking by staff ( P =0.003). Time spent idle was significantly correlated with time spent immobile ( P =0.001) and time spent idle was also significantly correlated with an increasingly inactive ranking by staff ( P