Under pressure: Effect of first parity sows' early social experience and genetic line on behavior and lesions in a paired interaction test with an older sow

Linda Marie Backeman Hannius^*Linda Jane KeelingAnna Wallenbeck

• Swedish University of Agricultural Sciences, Uppsala, Sweden

Sows housed in groups navigate complex social relationships, and individual differences affect their engagement and stress responses. To explore what shapes these differences, this study examined how genetic background and early-life social experiences affect social behavior, response strategies and risk of injury among young sows during pairwise social interactions with unfamiliar older sows. Eighty first-parity sows of two genetic lines, Swedish Yorkshire and Dutch Yorkshire, were raised in either socially mixed litters (with access to non-littermates through shared piglet areas) or conventional pens. At ten weeks, they were either grouped with unfamiliar gilts or remained with their original littermates. After first weaning, each sow was introduced to an unfamiliar older sow in a 60-minute interaction test, with behaviors recorded and lesions assessed before and after. By coding both the initiator's behavior and the response, and linking these to lesion placement, more nuanced response patterns could be analyzed for the short-term development of agonistic encounters. Swedish Yorkshire sows exhibited more aggression, responded more forcefully, and sustained more injuries, particularly to the front and rear part, compared to Dutch Yorkshire sows, which showed less aggression and lesions. Early-life social mixing correlated with more affiliative behaviors and a tendency to not react in social interactions, suggesting greater social tolerance, though not reduced injury risk. Social mixing later in rearing was not associated with behavior or lesion counts. Interaction effects were sparse, indicating broadly similar behavior–lesion associations across lines and rearing treatments in this paired interaction test with two sows. A few behavior-specific associations were observed, such as a higher lesion risk among late-mixed sows showing defensive responses but these were limited in scope and not consistent across outcomes. Across treatments and lines, retreat was common among first-parity sows, which may offer short-term protection but could have longer-term costs in social learning and affiliation. While the controlled test setting allowed for a precise measurement of social tendencies, real-world group housing involves more complex interactions. These findings highlight the need to consider genetic lines and early experience in understanding social behavior and indicate that future studies should evaluate these effects under practical group housing conditions.