The association between social structure and development of tail biting was evaluated in this study. Pigs with intact tails (n = 96, initial weight = 25.0 ± 3.0 kg) were assigned to treatment based on their litter origin: littermates, non-littermates, and half-group littermates (4 pigs per litter). Four pens (8 pigs/pen) of each litter origin treatment were studied for 14 weeks. Tail injury was assessed weekly. Pigs with visible blood were considered victims of tail-biting. Behavior of pigs was video-recorded between 0900h and 1500h at 4-week intervals starting one week after study initiation. Video-recordings were viewed continuously to identify tail-biting events and tail biters in each pen. A pig that bit the tail of another pig causing reactions of the recipient was identified as a tail-biter. Pigs that were neither biters nor victims were classified as “others”. For social network analysis, video-recordings were scanned at 10-min intervals to register pigs that were lying together (1) or not (0) in binary matrices. Half weight association index was used for social network construction. Social network analysis was performed using the UCINET software. Litter origin treatment did not affect growth performance. Fifty nine percent of the littermate treatment (Chi-square = 11.6; P = 0.02), 34% of the half-group littermate treatment, and 22% of the non-littermate treatment were identified as victims. Littermate treatment had lower network density (Table 1) and fewer social ties than the non-littermate treatment, indicating that littermates were less socially connected with each other at the pen level. At the dyad level, the littermate treatment formed more social ties between biters and victims, and fewer ties between victims and others than non-littermates (Table 1). These results suggest that more social ties between biters and victims, and fewer social ties between victims and others may predispose littermates to development of tail biting.
展开▼
