Skip to main content
ARS Home » Midwest Area » West Lafayette, Indiana » Livestock Behavior Research » Research » Publications at this Location » Publication #223488

Title: The social behavior carried out by unacquainted sows on mixing may predict the likelihood of escalation into aggression

item Marchant, Jeremy
item Schenck, Erin
item Lay Jr, Donald

Submitted to: Journal of Animal Science
Publication Type: Abstract Only
Publication Acceptance Date: 4/2/2008
Publication Date: 7/7/2008
Citation: Marchant Forde, J.N., Garner, J.P., Schenck, E.L., Johnson, A.K., Lay, Jr, D.C. 2008. The social behavior carried out by unacquainted sows on mixing may predict the likelihood of escalation into aggression [abstract]. Journal of Animal Science. 86(E.Suppl2):352.

Interpretive Summary:

Technical Abstract: Aggression is a major problem when housing sows in groups. Aggression can increase injuries, stress, and cost of production, and decrease productivity. The aim of this project was to determine the behavioral sequences associated with fight and non-fight interactions when two unacquainted sows are mixed. Twelve unacquainted multiparous sows were mixed in pairs after weaning. Body weight, parity, litter of origin, and body length were measured one day prior to mixing. All sows were paired based on body weight and parity. Each pair was placed in a 64 ft2 pen, with 4 ft high solid walls to eliminate outside influences on behavior, for 1 hr during which behavior was video-recorded. All social interactions were extracted from the video data, using an ethogram of 10 behaviors to describe each interaction. A new interaction was defined as 5 or more seconds of non-contact from the end of the previous interaction. Aggressive interactions were defined as a bite to some part of the body. To identify behaviors predictive of aggression, we collated the first 20 behaviors in an observation, and the last 20 immediately preceding the first act of aggression, to form non-aggressive and aggressive behavioral profiles respectively. We predicted that changes between these profiles would identify behaviors associated with the onset of aggression. We used a split-plot GLM, blocked by observation, to compare the profiles. Profiles differed between the start of the observation and prior to the first fight (F9,95 = 4.89; P<0.0001). Bonferroni post hoc planned comparisons showed that “nose-to-nose contact” decreased immediately before a fight (P=0.0001; alpha = 0.005), and “no reaction to social behavior” was greater (P=0.0060; alpha = 0.005). These results suggest that nose-to-nose contact reduces the chance of aggression in sows, while a failure to respond to social behaviors increases it.