Skip to main content
ARS Home » Midwest Area » West Lafayette, Indiana » Livestock Behavior Research » Research » Publications at this Location » Publication #264515
Title: Action-reaction: using Markov analysis to elucidate social behavior when unacquainted sows are mixed

Author
item Marchant, Jeremy
item GARNER, JOSEPH - Purdue University
item Lay Jr, Donald
item JOHNSON, ANNA - Iowa State University

Submitted to: International Society of Applied Ethology
Publication Type: Abstract Only
Publication Acceptance Date: 4/1/2011
Publication Date: 7/4/2011
Citation: Marchant Forde, J.N., Garner, J.P., Lay Jr, D.C., Johnson, A.K. 2011. Action-reaction: using Markov analysis to elucidate social behavior when unacquainted sows are mixed. International Society of Applied Ethology. Proceedings.

Interpretive Summary:

Technical Abstract: Sows fight when mixed but there is little information on the detailed social behaviors performed by unacquainted sows at mixing. This project aimed to determine the sequence of behaviors leading up to aggression when unacquainted sows were mixed in pairs or in two established sub-groups of three, in different environments. Experiments 1 and 2 used Yorkshire × Landrace sows mixed indoors. Expt. 1 used 11 sow pairs mixed into a 6.2 m**2 pen. Expt. 2 used 14 groups of 3 sows, with 2 groups mixed into a 19.2 m**2 pen. Experiments 3 and 4 used Berkshire sows mixed outdoors. Expt. 3 used 16 sow pairs mixed into a 5000 m**2 grassy paddock. Expt. 4 used 28 groups of 3 sows, with 2 groups mixed into a 5000 m**2 grassy paddock. In all experiments, behavior was recorded continuously for 60 min post-mixing. All-occurrences sampling was used to extract social interactions with each initiator behavior being followed by a recipient behavior, giving a sequence of reciprocated behaviors for each interaction. After extraction, the data were subject to a 1st order Markov transitional analysis to distinguish meaningful behavioral pairs. When mixed in pairs indoors, only 14.0% of withdrawal behaviors and 19.6% of walking behaviors were followed by break - i.e. the interaction ending, whereas outdoors, the same behaviors were followed by break 47.0% and 40.0% of times respectively. Indoors, 32.3% of bites were preceded by no response/pause behaviors, as opposed to 18.4% outdoors. With sows mixed in groups, 46.5% of no response/pause behaviors were followed by bites indoors, whereas, 30.4% were followed by bites outdoors. In the other direction, 39.7% of bites indoors were preceded by no response/pause behaviors, as opposed to 8.5% outdoors. Break was preceded by withdrawal/head tilt 61.1% of times outdoors but only 15.6% of times indoors. Overall, the results indicate that, indoors, ignoring or trying to avoid the other sow did not necessarily diffuse the interaction, given the limited space, whereas outdoors, sows were able to use the space to effectively end aggression. The information derived from this study furthers our understanding of factors that may promote or ameliorate aggression in different mixing environments.