Reconsidering coprophagy as an indicator of negative welfare for captive chimpanzees

Abstract

For captive chimpanzees, ‘abnormal’ behaviours include behaviours observed only in captivity (i.e. species-atypical behaviours) and those that are performed at higher rates in captivity compared to in the wild. Both types are used as metrics for evaluating captive primates’ welfare. However, categorizing all abnormal behaviours together ignores variation in their etiologies, which limits our ability to understand them and provide useful interventions. Coprophagy (deliberately eating faeces) is an intriguing abnormal behaviour because, unlike many abnormal behaviours, it is performed at higher rates among captive chimpanzees that were mother reared, compared to those that were human reared, and it has been proposed that it represents a socially learnt ‘cultural’ behaviour. Furthermore, coprophagy is observed among both wild and captive animals, although at higher rates in captivity. Typically, coprophagy is classed with other abnormal behaviours by those evaluating captive chimpanzee welfare, but such categorization has arisen from a top–down approach based on a priori assumptions. To apply a bottom–up approach, which would allow us to identify relations between behaviours in chimpanzees’ repertoire, in this study we ran a principal components analysis on the behaviours performed by 60 captive chimpanzees, to determine whether coprophagy should be classified with other abnormal behaviours. The principal components analysis revealed seven factors that we termed social, aggressive, playful, active, feed, abnormal and self-directed. Furthermore, the analysis revealed that coprophagy loaded onto the ‘social’ factor, which included positive social behaviours, and not onto the ‘abnormal’ factor, which included other abnormal behaviours. Supporting previous research, we also found that those chimpanzees that were mother-reared showed higher rates of coprophagy than those that were human reared; there was a significant positive correlation between time spent with conspecifics during the first four years of life and the rate of coprophagy performed by the subjects as adults (r=0.575, N=60, P<0.001). We discuss these results giving consideration to the practical applications for monitoring and evaluating captive chimpanzee welfare and also from a theoretical perspective about the social learning mechanisms that may underpin the transmission of coprophagy among captive chimpanzees.