We determined whether a socially induced diet preference could ameliorate a conditioned food aversion in lambs. To do so, one group of lambs was conditioned to avoid the shrub Cercocarpus montanus by giving them lithium chloride, a toxin that produces strong food aversions, following shrub ingestion. On Day 1 of conditioning, lambs ingested an average of 29 bites of C. montanus, but by Day 2 the number of bites had decreased to two and lambs took even fewer bites on Days 3-5. Following conditioning, averted lambs spent a substantial amount of time foraging on C. montanus when foraging with non-averted conspecifics in pasture trials (10% vs. 19% of time spent ingesting C. montanus) and in pen trials (25 vs. 35 bites). We also examined the effects of group composition during conditioning (heterogeneous vs. homogeneous) on the use of C. montanus during pasture and pen trials. In the heterogeneous groups, lambs being averted to C. montanus were exposed to the shrub in the presence of lambs not being averted to C. montanus. In the homogeneous groups all lambs were averted to C. montanus. Lambs in the heterogeneous groups took numerically more bites of C. montanus than lambs in the homogeneous groups during conditioning (25 vs. 17 bites day-1), but the trend was not significant (P > 0.05). When foraging on pastures, however, lambs in the heterogeneous groups spent more (P < 0.05) time browsing C. montanus than lambs in the homogeneous groups during Trials 2 (11.4% vs. 3.7%), 3 (16.7% vs. 3.2%), and 4 (42.9% vs. 28.1%), but not during Trial 1 (7.6% vs. 3.9%). Lambs in the heterogeneous groups took numerically more bites of C. montanus than lambs in the homogeneous groups during a pen trial following the pasture trials (35 vs. 25 bites), but the trend was not significant (P > 0.05). The capacity of social interactions to ameliorate food aversions suggests that social influences are a major determinant of diet selection in sheep.
