Group size and phenotypic appearance: Their role on the social dynamics in pullets

- We studied the effects of group size and phenotypic diversity on social behaviour.
- Three replications per group size and phenotypic treatment were used.
- Birds in groups of 10 interacted more than those in larger groups of 20-40.
- Phenotype determined the direction of active aggression in groups of 20 birds.
- No other effects of phenotypic diversity were found.

Non-caged production systems offer greater freedom of movement and behavioural opportunities to pullets, which may also include the occurrence of undesired behaviours. The incidence of such behaviours may be affected by group size but also by the group memberś phenotype. This study was designed to explore the effects of group size and phenotypic appearance in the social dynamics of pullets. A total of 1050 day old Hy-line Brown was randomly assigned to 45 pens at 3 group sizes (GS), 10, 20 and 40 birds (constant density 8 hens/m2). For all GS treatments, the phenotypic appearance (PA) of different bird proportions was modified with a black mark at the back of their head. Two types of populations were studied: homogenous (0, 100%) with all pen members presenting the same marked (M) or unmarked (U) phenotype, and heterogeneous (30, 50, and 70%) were the two phenotypes (M and U) coexisted in the same pen. All pens were observed during rearing on alternate weeks. Aggressive and affiliative social interactions performed among group members were registered together with their PA (M or U). The observed and expected mean frequencies for all phenotype combinations involved in the social interaction were calculated: MM, MU, UU and UM, where the giveŕs phenotype is indicated by the first letter and the receiveŕs phenotype by the second letter. Data were analysed with mixed model ANOVAs that included GS and PA as fixed factors for the general model, and GS, PA and type of interaction (MM, MU, UU, UM) to determine the direction of the interaction in the case of heterogeneous groups. Both aggressive (threats) and affiliative interactions (exploratory pecking and beak pecking) were more frequent in small (10) than in large (20, 40) GS (P < 0.05) regardless of PA (PA, P > 0.05; GS × PA, P > 0.05). Aggressive interactions however, occurred at a higher than expected frequency in heterogeneous GS 20, and were predominantly from U towards M birds (UM; P < 0.05), as compared to MU or UU. The significant interaction among GS × PA × Type of interaction for affiliative behaviours did not show a clear relationship. Overall, the results obtained showed that both aggressive and affiliative interactions occurred at a higher frequency in the smallest GS 10. Although the effects of PA were not as clear as expected, these results provide evidence of the directionality of the aggressive interactions from U towards M birds especially at intermediate GS 20.