Personality predicts social dominance in female zebra finches, Taeniopygia guttata, in a feeding context Morgan David a, b, * , Yannick Auclair a,1 , Frank Cézilly a a Equipe Ecologie Evolutive, UMR CNRS 5561 Biogéosciences, Université de Bourgogne b Groupe de Recherche en Ecologie Comportementale et Animale, Département des Sciences Biologiques, Université du Québec à Montréal, Canada article info Article history: Received 13 May 2010 Initial acceptance 9 July 2010 Final acceptance 5 October 2010 Available online 13 November 2010 MS. number: 10-00351R Keywords: behavioural syndrome competition dominance false discovery rate feeding success hierarchy life history productivity personality repeatability zebra finch Although personality has been defined as a suite of correlated behaviours, most studies of animal personality actually consider correlations between a few traits. We examined the repeatability and correlational structure of five potential personality traits (activity, neophobia, exploratory tendencies, risk-taking behaviour and obstinacy), in female zebra finches. In addition, we assessed to what extent personality influenced social dominance in a feeding context in this gregarious species. All personality traits were found to be highly repeatable within individuals. In addition, except for obstinacy, all of them were related to each other, thus defining a behavioural syndrome. Social dominance was predicted by personality, with proactive individuals being more likely to be dominant. Our results suggest that personality can be considered as a new static factor influencing within-group hierarchies. We finally discuss these results in terms of the consequences for the evolution of personalities and the need to take several traits into account to provide full descriptions of individual personality. Ó 2010 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved. Consistent individual differences in behaviour across various contexts and situations are of increasing interest to behavioural ecologists (Sih et al. 2004a; Réale et al. 2007). Several studies have shown that individual variation in personality can have fitness consequences (reviewed in Smith & Blumstein 2008). For instance, survival between two breeding seasons is related to personality in female red squirrels, Tamiasciurus hudsonicus (Boon et al. 2008), whereas the influence of personality on survival varies according to population density in the common lizard, Lacerta vivipara (Cote et al. 2008). Reproductive fitness can also be influenced by personality, as shown by recent evidence in fish (Wilson et al. 2010a), birds (van Oers et al. 2008) and mammals (Réale et al. 2009). In addition, individual variation in personality can be repeatable (Quinn et al. 2009; Wilson & Godin 2009) and linked to genetic variation (van Oers et al. 2004a, b; Fidler et al. 2007), suggesting that variation in personality can be exposed to natural selection. Assessing the ecological relevance of personality, both between and within species (Uher 2008), however, requires that personality be assessed in some standardized way. Although personality has been defined as a suite of correlated behaviours expressed either within a given behavioural context or across different contexts (Sih et al. 2004b), in practice most studies of animal personality consider correlations between a few traits, focusing eventually on only two aspects of personality such as, for instance, exploration and risk taking (Brodin 2009; Harcourt et al. 2009; but see Wilson et al. 2010b). This is in strong contrast to studies of personality in human beings, where personality is assessed from patterns of covariation of traits belonging to several broad factors (McCrae et al. 1998; Uher 2008). Recent studies of human personality, however, suggest that measuring only a few personality traits might be insufficient to predict complex behaviours (Paunonen et al. 2003; Ashton & Lee 2007; de Vries et al. 2009). In partic- ular, using a larger number of factors is recommended to reduce error in behaviour prediction and increase accuracy in behaviour explanation (Paunonen & Ashton 2001), particularly in the case of relatively complex behaviours, which can be considered multidi- mensional and multidetermined (Paunonen et al. 2003). Although various constraints can limit the number of personality dimensions that can be quantified using free-ranging individuals, * Correspondence: M. David, Equipe Ecologie Evolutive, UMR CNRS 5561 Biogéosciences, Université de Bourgogne, 6, bld Gabriel, 21000 Dijon, France. E-mail address: Morgan.David@u-bourgogne.fr (M. David). 1 Yannick Auclair is now at the Institute of Evolutionary Biology and Environ- mental Studies, c/o Verhaltensbiologie, University of Zürich-Irchel, Winter- thurerstrasse 190, 8057 Zürich, Switzerland. Contents lists available at ScienceDirect Animal Behaviour journal homepage: www.elsevier.com/locate/anbehav 0003-3472/$38.00 Ó 2010 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.anbehav.2010.10.008 Animal Behaviour 81 (2011) 219e224