The interconnection of hierarchy, affiliative behaviours, and social play shapes social dynamics in Maremmana beef cattle

Dynamic Multi-Behaviour, Orientation-Invariant Re-Identification of Holstein-Friesian Cattle

To perform reliable animal re-identification, most available algorithms require standardised animal poses. However, this lack of versatility prevents widespread application of these algorithms in behavioural research and commercial environments. To circumvent this, we incorporated information about the orientation and behaviour of the animals in an embedding-based algorithm to re-identify Holstein-Friesian cattle. After all, the orientation and behaviour of an animal determine which body parts of an animal are visible from the camera's perspective. We evaluated our approach using a dataset with more than 11,000 instance segments of Holstein-Friesian cattle, but our methodology is readily generalisable to different animal species. Our results show that incorporation of informative metadata parameters in the re-identification procedure increases the rank-1 re-identification accuracy from 0.822 to 0.894, corresponding to a 40% reduction in the number of incorrectly identified animals.

Automated, longitudinal measures of drinking behavior provide insights into the social hierarchy in dairy cows

Dairy cows compete for feed and water access on commercial farms. In this study we used EloSteepness to assess the summed Elo winning probabilities (i.e., dominance) of 87 cows housed in a dynamic group and compared the resulting social hierarchies based on their steepness (i.e., the average degree of differences in winning probability between adjacently ranked individuals in the group, ranging from 0 to 1). We identified a hierarchy at the drinker with a steepness of 0.55 ± 0.02 (SD), whereas the hierarchy detected at the feeder during the same time period was 0.45 ± 0.02, indicating smaller dominance differences among cows when competing for feed compared with competing for water. Individual cows' winning probabilities at the feeder and drinker were moderately correlated (rs = 0.55), and cows at the lower and upper ends of the hierarchy showed good agreement. We compared the drinker hierarchy between hot (i.e., temperature-humidity index [THI] ≥72) and normal (i.e., THI <72) periods. The hierarchy steepness was similar in both hot (0.54 ± 0.03) and normal conditions (0.56 ± 0.03), and there was a strong correlation in cows' individual winning probabilities across these periods (rs = 0.87). Cows with higher winning probability visited the drinker less frequently (hot: rs = -0.40, normal: rs = -0.33) but had a higher average daily water intake (hot: rs = 0.38, normal: rs = 0.37). We also found evidence that individual cows' drinking times differ depending on their winning probability; cows with lower winning probability shifted their drinking times to before or after the visit peak after milking. Automatically identifying cows with consistently high or low winning probabilities using drinkers may help inform grouping decisions and water provision on farms.

Oppositions, joints, and targets: the attractors that are the glue of social interactions

Social interactions are often analyzed by scoring segments of predefined behavior and then statistically assessing numerical and sequential patterns to identify the structure of the encounters. However, this approach can miss the dynamics of the animals' relationship over the course of the encounter, one that often involves invariant bonds, say a nose-to-nose orientation, with many different movements performed by both partners acting to counteract each other's attempts to break or maintain the relationship. Moreover, these invariant bonds can switch from one configuration to another during an interaction, leading from one stable configuration to another. It is this stepwise sequence of configurational stabilities that lead to functional outcomes, such as mating, aggression, or predation. By focusing on the sequence of invariant relational configurations, the deep structure of interactions can be discerned. This deep structure can then be used to differentiate between compensatory movements, no matter how seemingly stereotyped they may appear, from movement patterns which are restricted to a particular form when more than one option is available. A dynamic perspective requires suitable tools for analysis, and such tools are highlighted as needed in describing particular interactions.

Animal play and evolution: Seven timely research issues about enigmatic phenomena

The nature of play in animals has been long debated, but progress is being made in characterizing play and its variants, documenting its distribution across vertebrate and invertebrate taxa, describing its mechanisms and development, and proposing testable theories about its origins, evolution, and adaptive functions. To achieve a deeper understanding of the functions and evolution of play, integrative and conceptual advances are needed in neuroscience, computer modeling, phylogenetics, experimental techniques, behavior development, and inter- and intra-specific variation. The special issue contains papers documenting many of these advances. Here, we describe seven timely areas where further research is needed to understand this still enigmatic class of phenomena more fully. Growing empirical and theoretical evidence reveals that play has been crucial in the evolution of behavior and psychology but has been underestimated, if not ignored, in both empirical and theoretical areas of evolutionary biology and neuroscience. Play research has important ramifications for understanding the evolution of cognition, emotion, and culture, and research on animals can be both informative and transformative.