Review: Peer social bonds in farm animals

Peer social bonds may be defined as relationships involving frequent and non-random (i.e. preferential) mutual affiliative associations and interactions between peers (excluding parental and sexual relationships) that are consistent across a biologically relevant time frame and behavioural states, and lead to stress modulation. Hence, these bonds can be defined based on the following four criteria: (1) social preference, (2) consistency of preferences, (3) stress modulation, and (4) symmetry or reciprocity of behaviours. These have been studied in wild animals, particularly primates, and shown to provide fitness benefits. In farm animals, behavioural studies increasingly report social bonds. Yet, it is unclear whether there is sound evidence to draw such conclusions. This systematic review aims to discuss the current knowledge of peer social bonds in farm animals, further assessing the evidential robustness of the literature. We also identify key knowledge gaps and provide recommendations for future studies. The retrieved literature included 47 papers covering a range of species including cows (n = 18), pigs (n = 9), sheep (n = 6), horses (n = 4), donkeys (n = 1), goats (n = 3), chickens (n = 3), and quails (n = 3). In some farm animal species (pigs, cows, horses), there is evidence of variation in strength (e.g. frequency) and/or non-randomness of interactions and associations between dyads, potentially representing peer bonds. However, there is currently insufficient research on the other criteria to draw conclusions on the existence of peer social bonds under farm conditions. Overall, studies face methodological challenges in the assessment of social bonds, which is further complicated by housing and husbandry conditions on commercial farms. This review shows that various methods have been used to assess social relationships, but that many of these have shortcomings, and lack the nuance to make conclusions on peer social bonds. Hence, we provide recommendations to address the identified methodological gaps. Research on peer social bonds in farm animals is in its infancy, and we hope this work inspires other scientists to address this important topic diligently, while taking into consideration the implications of research conclusions on animal management in practice.