Behavioral Genetics in Pigs and Relations to Welfare - Genetics and the Behavior of Domestic Animals - page 406

involved in more aggressive interactions after mixing than gilts with
smaller loin-eye area. Social rank, recorded during feeding, was highly
heritable when estimated on entire males at a test station and the genetic
correlation between position in social rank order and growth rate was
high ( Jonsson and Jorgensen, 1989 ). Thus, selection for high growth rate
could result in increased aggressiveness, which also has been proposed
by Schinkel et al. (2003) . We recently found that pigs with a high
direct breeding value for growth rate are more successful during social
contests ( Canario et al., 2012 ). They have a higher genetic merit for
both initiating and winning fights and bullying than other pigs.
Furthermore, these pigs are less frequently bullied and have fewer lesions
in the rear.
Individual behavior is adapted to the group around the individual and
the performance of the individual is dependent on the group mates.
Selection for individual performance maximizes the individual's results but
not necessarily that of the group the animal is raised in. Bill Muir became
interested in aggressive behavior because of its negative influence on pro-
duction results. His research has shown alternative methods of selection,
improving both production and welfare. Muir developed a model to handle
an unfavorable correlation between growth rate and competitive behavior
without any need for behavioral observations. The model is called “the
group model”, “the social model”, or “the competitive model”. The quanti-
tative relation between individual and group productivity was first pre-
sented by Griffing (1967) who extended classic population genetic models
to include social effects. Based on Griffing's theory, Muir (2005) pre-
sented a social model that was further developed by Bijma et al. (2007)
to estimate social genetic parameters. The social model for genetic evalua-
tion of growth rate includes not only the “ordinary” direct genetic effect
on an individual's growth rate, but also the social genetic effect on all
group members' growth rates ( Figure 11.3 ).
With this approach, two breeding values are estimated for each animal,
one describing the animal's genetic ability to grow and the other describing
the animal's genetic ability to influence the growth of other animals in the
pen, the so called social breeding value. The correlation between the direct
(D) genetic effect and the social (S) genetic effect (r DS ) is negative (unfavor-
able) if the animals compete for limited resources such as food and space. If
that is the case, it cannot be recommended to select the animal with the high-
est breeding value for the direct effect since fast-growing animals are best at
the expense of their group members. Instead, animals with rather high breed-
ing values for both the direct and the social effect should be selected. If the
genetic correlation between the direct genetic effect and the social genetic
effect is positive, i.e. favorable, selection of the animal with the highest
breeding value for the direct effect will not harm group members. Even so,
using both direct and social breeding values in the selection can lead to a