Large-scale phenotyping of animal behaviour traits is time consuming and has led to increased demand for technologies that can automate these procedures. Automated tracking of animals has been successful in controlled laboratory settings, but recording from animals in large groups in highly variable farm settings presents challenges. The aim of this review is to provide a systematic overview of the advances that have occurred in automated, high throughput image detection of farm animal behavioural traits with welfare and production implications. Peer-reviewed publications written in English were reviewed systematically following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. After identification, screening, and assessment for eligibility, 108 publications met these specifications and were included for qualitative synthesis. Data collected from the papers included camera specifications, housing conditions, group size, algorithm details, procedures, and results. Most studies utilized standard digital colour video cameras for data collection, with increasing use of 3D cameras in papers published after 2013. Papers including pigs (across production stages) were the most common (n = 63). The most common behaviours recorded included activity level, area occupancy, aggression, gait scores, resource use, and posture. Our review revealed many overlaps in methods applied to analysing behaviour, and most studies started from scratch instead of building upon previous work. Training and validation sample sizes were generally small (mean±s.d. groups = 3.8±5.8) and in data collection and testing took place in relatively controlled environments. To advance our ability to automatically phenotype behaviour, future research should build upon existing knowledge and validate technology under commercial settings and publications should explicitly describe recording conditions in detail to allow studies to be reproduced.

Social interactions between individuals living in a group can have both positive and negative effects on welfare, productivity, and health of these individuals. Negative effects of social interactions in livestock are easier to observe than positive effects. For example, laying hens may develop feather pecking, which can cause mortality due to cannibalism, and pigs may develop tail biting or excessive aggression. Several studies have shown that social interactions affect the genetic variation in a trait. Genetic improvement of socially-affected traits, however, has proven to be difficult until relatively recently. The use of classical selection methods, like individual selection, may result in selection responses opposite to expected, because these methods neglect the effect of an individual on its group mates (social genetic effects). It has become clear that improvement of socially-affected traits requires selection methods that take into account not only the direct effect of an individual on its own phenotype but also the social genetic effects, also known as indirect genetic effects, of an individual on the phenotypes of its group mates. Here, we review the theoretical and empirical work on social genetic effects, with a focus on livestock. First, we present the theory of social genetic effects. Subsequently, we evaluate the evidence for social genetic effects in livestock and other species, by reviewing estimates of genetic parameters for direct and social genetic effects. Then we describe the results of different selection experiments. Finally, we discuss issues concerning the implementation of social genetic effects in livestock breeding programs. In our opinion, this review demonstrates that selection for socially-affected traits, using methods that target both the direct and social genetic effects, is a promising, but sometimes difficult to use in practice, tool to simultaneously improve production and welfare in livestock.

Facial expressions convey information on emotion, physical sensations, and intent. The much debated theories that facial expressions can be emotions or signals of intent have largely remained separated in animal studies. Here we integrate these approaches with the aim to 1) investigate whether pigs may use facial expressions as a signal of intent and; 2) quantify differences in facial metrics between different contexts of potentially negative emotional state. Facial metrics of 38 pigs were recorded prior to aggression, during aggression and during retreat from being attacked in a dyadic contest. Ear angle, snout ratio (length/height) and eye ratio from 572 images were measured. Prior to the occurrence of aggression, eventual initiators of the first bite had a smaller snout ratio and eventual winners showed a non-significant tendency to have their ears forward more than eventual losers. During aggression, pigs’ ears were more forward orientated and their snout ratio was smaller. During retreat, pigs’ ears were backwards and their eyes open less. The results suggest that facial expressions can communicate aggressive intent related to fight success, and that facial metrics can convey information about emotional responses to contexts involving aggression and fear.