Loose housing of small goat groups: Influence of visual cover and elevated levels on feeding, resting and agonistic behaviour

Frequent social conflicts among goats due to their strict rank relationships can pose problems in loose housing of goats by negatively influencing feeding and resting times of low-ranking group members. In this study, we tested whether enrichment of loose-housing pens with structural elements providing visual cover (partitions, lying niches) and elevated levels (platforms) is effective in reducing agonistic interactions and interruptions in feeding and resting behaviour in small groups of goats. The study involved 8 groups of goats consisting of 8-9 non-lactating females that differed in grouping age and the presence of horns in a two-by-two factorial design. We compared the feeding, resting and agonistic behaviour the goats showed in pens to which they were well habituated (original situation) with the behaviour they performed in these pens after enriching them with additional structural elements providing visual cover and elevated levels (enriched situation). Finally, the arrangement of the pens was restored to the original situation (restored situation). In each group, the behaviour of two low-ranking, two medium-ranking and two high-ranking goats was continuously recorded by video. Using linear mixed-effects models, variables of goat behaviour were analysed with situation, rank, grouping age and presence of horns as fixed explanatory variables. With most of the outcome variables, the effects of the housing situation interacted with those of the rankand/or the presence of horns. From the original to the enriched situation, the feeding bouts were generally longer. Furthermore, from the original to the enriched situation, fewer feeding bouts of low- and medium-ranking goats were interrupted by displacements, and fewer of these bouts were interrupted by medium- and high-ranking goats to initiate agonistic interactions. In addition, fewer resting bouts were interrupted by displacements received, and in medium- and high-ranking goats, the risk of an interruption whilst resting due to initiating an agonistic interaction decreased from the original to the enriched situation. The outcome variables' values in the restored situation hardly ever returned to the values of the original situation which is likely to indicate a carry-over effect of the enriched to the restored situation. In conclusion, our results show that enriching pens for small groups of goats with structural elements providing visual cover and elevated levels positively affects feeding, resting and agonistic behaviour. (C) 2009 Elsevier B.V. All rights reserved.