BACK TO QUEBEC CITY 2002 TITLES
Determination of social
dominance in groups of pigs: methodology approach
C. Bousquet and M.C. Meunier-Salaün
INRA, Unité Mixte
de Recherches sur le Veau et le Porc, Saint-Gilles, France
Social hierarchy within groups of pigs has generally been determined at mixing or during competitive feeding situations, in pairs or within the group context. The social rank has been classically determined from aggressive and submissive acts during social confrontation and defined as a dominance index calculated in different ways, including the frequency of agonistic interactions, the proportion of wins and defeats or the number of conspecifics that each pig dominates in relation to group size. In most cases the definition of a dominance index postulates a linear hierarchy within groups. The aim of our study was to evaluate methods for the determination of the social status in groups of eight (N) growing pigs subjected to a food competition test after a 16h food deprivation. During the test lasting 30 min, all agonistic interactions exchanged between each pig were recorded (nature of interaction and the identity of the initiator and the recipient). We built a matrix based on the total number of interactions exchanged within each dyad (Tij, Tji). An individual score was calculated from the nature and number of acts exchanged within each dyad: 0 for the less aggressive animal, 1 for the more aggressive animal and 0.5 for each animal when agonistic behaviour was similar (Tij-Tji <3). Secondly, a dominance index was determined for each pig, by dividing the sum of its scores (maximum = N-1) by the number of conspecifics interacting with the considered animal. The resulting index for all groups was normally distributed and resulted in significant differences in the feeding pattern of growing pigs fed with an automatic feeder. From the matrix of the agonistic interactions within every dyad in each group, the “graph” theory (Roy, 1969) was also applied to determine the social status of each pig. The results showed the existence of triangular or circular relationships within the experimental groups.