Social models fail to induce diet and feeding site avoidance in naı¨ve yearling steers

Social learning can be of critical importance to cattle grazing rangeland environments with high variability of food resources across space and time. Experienced individuals can greatly facilitate foraging decisions (what to eat and where to eat) of naı¨ve peers in such settings. We conducted an experiment with cattle to investigate strength and persistence of socially induced food and feeding site avoidance behaviours. Sixteen naı¨ve yearling steers were paired with 16 social models that had either not been trained (control) or been trained with an emetic (LiCl), electrical shock or both to avoid: (a) an unsafe high-quality food (LiCl); (b) an unsafe high-quality feeding site (shock); or (c) both the unsafe high-quality food and the unsafe high-quality feeding site (LiCl1shock). Ten-minute trials were conducted in an experimental arena containing three artificial feeding sites each consisting of groups of bowls with either high- (HQ) or moderate-quality (MQ) foods (HQ5barley and oat grain; MQ5Bermuda grass hay). Unsafe high-quality (UHQ, surrounded by traffic cones) and safe moderate-quality (SMQ) feeding sites consisted of nine rubber bowls containing either HQ or MQ foods. The safe high-quality (SHQ) feeding site consisted of two groups of eight bowls containing HQ food, which surrounded the UHQ and SMQ feeding sites. Social models did not induce diet and feeding site avoidance behaviours in naı¨ve steers; they exerted small and transient changes in the feeding behaviour of their naı¨ve counterparts. Consequences to the individual outweighed social influences; when naı¨ve animals experienced the same punishment contingencies as their social models, their behavioural patterns closely resembled those of their social model. Conditioned food and location aversions via LiCl were apparently influenced by prior exposure to target foods and the experimental arena. Conversely, conditioned feeding site avoidance via shock was apparently not influenced by prior exposure to target foods or the experimental arena