Contents lists available at ScienceDirect Behavioural Processes journal homepage: www.elsevier.com/locate/behavproc To be so bold: boldness is repeatable and related to within individual behavioural variability in North Island robins Ruchuan He a , Emilio Pagani-Núñez a,b , Clément Chevallier c , Craig R.A. Barnett a,d, ⁎ a College of Forestry, Guangxi University, Nanning, People’s Republic of China b State Key Laboratory of Biocontrol, College of Ecology and Evolution/School of Life Sciences, Sun Yat-sen University, Guangzhou, People’s Republic of China c Centre d’Étude Nordique, Université du Québec à Rimouski, Rimouski, Québec, Canada d Department of Zoology, Graduate School of Science, University of Kyoto, Kyoto, Japan ARTICLE INFO Keywords: Boldness Intraindividual variability (IIV) North Island robins Repeatability Risk Unpredictability ABSTRACT Behavioural research traditionally focusses on the mean responses of a group of individuals rather than variation in behaviour around the mean or among individuals. However, examining the variation in behaviour among and within individuals may also yield important insights into the evolution and maintenance of behaviour. Repeatability is the most commonly used measure of variability among individuals in behavioural research. However, there are other forms of variation within populations that have received less attention. One such measure is intraindividual variation in behaviour (IIV), which is a short-term fluctuation of within-individual behaviour. Such variation in behaviour might be important during interactions because it could decrease the ability of conspecific and heterospecific individuals to predict the behaviour of the subject, thus increasing the cost of the interaction. In this experiment, we made repeated measures of the latency of North Island robins to attack a prey in a novel situation (a form of boldness) and examined (i) repeatability of boldness (the propensity to take a risk), (ii) IIV of boldness, and (iii) whether there was a significant relationship between these two traits (a behavioural syndrome). We found that boldness was highly repeatable, that there were high levels of IIV in boldness, and that there was a negative relationship between boldness and IIV in boldness. This suggests that despite high levels of repeatability for this behaviour, there were also still significant differences in IIV among different individuals within the population. Moreover, bolder individuals had significantly less IIV in their boldness, which suggests that they were forming routines (which reduces behavioural variability) compared to shyer individuals. Our results definitively demonstrate that IIV itself varies across individuals and is linked with key behavioural traits, and we argue for the importance of future studies aimed at understanding its causes and consequences for behavioural interactions. 1. Introduction Individual animals often show consistency in their behavioural expression that is unrelated to time or context (Sih et al., 2010), which we call personality. Personality traits are often also related to other traits in predictable ways and have been linked with other phenotypic traits (i.e. behavioural syndromes, Biro and Stamps, 2008). Many authors argue that the existence of animal personalities is problematic for behavioural evolution because they restrict behavioural flexibility. However, there is growing evidence for the fitness benefits of main- taining personality differences within populations (Dingemanse and Réale, 2005; Smith and Blumstein, 2008). For example, in great tits (Parus major), exploration and boldness were related to food avail- ability. In winters with low food competition (during mast seeding), fast exploring males and slow exploring females had the highest over winter survival rates. Conversely, in years with higher over winter competi- tion, the trend was reversed (Dingemanse and Réale, 2005). Therefore, the interactions between personality, environment, and fitness are complex and in need of further study. Behavioural consistency is a cornerstone of personality research and one of the commonest ways to assay behavioural consistency is to calculate the level of individual repeatability for a behavioural trait (R) within the population. Repeatability is a measure that quantifies an individual’s change in the expression of a trait over time. Expressed mathematically, repeatability is the ratio of between-group (or between individual) variation and the total phenotypic variation (Bell et al., 2009). It can be likened to the expected within individual correlation among measurements and is sometimes referred to as the intra-class http://dx.doi.org/10.1016/j.beproc.2017.04.014 Received 15 December 2016; Received in revised form 5 April 2017; Accepted 24 April 2017 ⁎ Corresponding author at: Department of Zoology, Graduate School of Science, University of Kyoto, Kyoto Japan. E-mail address: optimalforager@hotmail.com (C.R.A. Barnett). Behavioural Processes 140 (2017) 144–149 Available online 25 April 2017 0376-6357/ © 2017 Elsevier B.V. All rights reserved. MARK