How Studying Interactions Between Animal Emotions, Cognition, and Personality Can Contribute to Improve Farm - Genetics and the Behavior of Domestic Animals - page 92

animals are sufficiently similar to be described by the same label ( Gosling and
John, 1999; Gosling, 2001; Mehta and Gosling, 2006 ). There are many defini-
tions of “personality”; the one proposed by Pervin and John (1997; cited
by Gosling, 2001 ), “consistent patterns of feeling, thinking, and behaving”,
incorporates three different elements, namely emotion, cognition, and behav-
ior. Likewise, whereas dimensions of personality are generally discussed
in terms of coping style or fearfulness, it has been proposed that there is
currently no single concrete method for assessing a propensity to feel and
express emotions ( Gosling, 2001 ).

Individual Variability in Emotional Behaviors
To qualify for the label “personality,” individual differences have to be
consistent over time and across situations ( Erhard and Schouten, 2001 ),
which are typically social isolation, novelty, suddenness, predator response (in
many cases humans may evoke the same reactions as predators). Comprehensive
reviews of the literature on fear tests used on farm animals ( Boissy, 1998;
Dodd et al., 2012; Forkman et al., 2007; Jones and Boissy, 2010 ) already exist
and we will just present some newer findings in the field in this section.
Individual differences in fearfulness are not just interesting from a theoretical or
animal welfare point of view. High levels of fear can make animals difficult and
dangerous to handle.

Consistency of Fear-Related Responses Across
Different Frightening Situations
In animal experimentation, this element of personality is referred to as the
'validity' of a behavior test ( Forkman et al., 2007 ). It is based on the notion
that a personality trait will be expressed in a similar way in different situa-
tions within the same context. For example, sheep that were prepared to
move away from their group to explore an unknown environment in a behav-
ior test were found to have a lower group-cohesion nine weeks later when
grazing undisturbed. They were less easily frightened, recovered more easily
from a fright ( Sibbald et al., 2009 ), and split into sub-groups more easily
( Michelena et al., 2009 ) than sheep that explored less. While not all behavior
variables may be correlated across tests, there is often sufficient evidence
to suggest a common trait, in that individuals that react more strongly in one
situation will also react more strongly in others. In horses, for instance,
the frequency of licking/nibbling a novel object, the time to put one foot on
a novel area and to eat from a bucket placed just behind it, and the flight
distance and the time to eat under an opened umbrella were found to be
interrelated and consistent over time ( Lansade et al., 2008a ).
Not all reports show a strong consistency, however. This can have several
reasons: maybe the test is not appropriate, or the situations belong to