Behavioural Traits Recorded in Gilts and Associations with Reproductive Performance as Group-Housed Sows

Abstract

Gilts from two maternal lines were recorded for flight time (FT, N=8854) and scored for the count of lesions resulting from fighting 24 hours after selection and mixing into new groups (N=3238). Anterior (ANT) and posterior lesion counts were scored on a progressive four point scale representing none to multiple lesions (0-3), and aggressive gilts (0/1 scores) were defined by ANT>1. Lesion counts over the whole body were subsequently rescored pre-farrowing (PFBLES). Genetic correlations were estimated between these behavioural traits and average daily gain (ADG), gilt removals without a farrowing event and first parity litter size (TB: total born; NBA: number born alive) and birth weight. All behavioural traits scored for gilts and pregnant sows prefarrowing were lowly to moderately heritable (h²<0.15), implying that selection could alter FT or reduce fighting behaviour and hence skin lesions at different time points. However, lesion scores for gilts were not highly correlated with later PBFLES, reflecting changes to individual participation in fighting behaviour over time. Skin lesion traits were also uncorrelated with FT. These traits therefore represent assessment of different behaviours. Flight time was genetically correlated with ADG (rg: 0.24±0.10) only. All behavioural traits were generally not significantly correlated (either genetically or phenotypically) with subsequent reproductive performance, implying a neutral association between behavioural phenotypes and selection criteria in maternal lines. However, low adverse phenotypic correlations between lesion and locomotion scores or gilt removals illustrate a detrimental impact of fighting on welfare. Using management strategies to reduce fighting and developing a better understanding of the genetic basis for long term behaviour remain important for ongoing improvement of welfare and performance of group housed sows.