Harnessing variability in temperament to improve outcomes in experimental biology

Animal personality, also sometimes called temperament, has been investigated in domesticated and non-domesticated species, mammal and non-mammal animals, from foxes and marsupials to horses and sheep. Standardized behavioural tests can be used to assess the phenotype of temperament and have identified large variability within populations of many species. Here, we present three examples that illustrate the importance of temperament variability in animal biology. The first example shows that while there were differences in temperament between five breeds of sheep that were raised and flocked together since weaning, there was a similar variability in temperament phenotype within each breed. The second example demonstrates that extreme temperament phenotypes, labelled calm or nervous, can affect how sheep perceive, habituate and anticipate a stressful situation to which they have been exposed previously. Our last example shows how the inclusion of individual temperament score can improve the predictive power of Bayesian models of the response of cortisol to an immune challenge. Classifying individuals based on their temperament, based on their phenotypic traits or genotyping, could offer a way to include the variability in temperament in experimental design. We conclude that the inclusion of temperament in experimental design could improve our understanding of animal biology. This article is part of the theme issue 'Embracing variability in comparative physiology: why it matters and what to do with it'.