Forum Do rapid assays predict repeatability in labile (behavioural) traits? A reply to Biro Hannah A. Edwards a, * , Isabel S. Winney a , Julia Schroeder a, d , Hannah L. Dugdale a, b, c a Department of Animal and Plant Sciences, University of Sheffield, Sheffield, U.K. b Theoretical Biology, Centre for Ecological and Evolutionary Studies, University of Groningen, Groningen, The Netherlands c Behavioural Ecology and Self-organization, Centre for Ecological and Evolutionary Studies, University of Groningen, Groningen, The Netherlands d Evolutionary Biology, Max Planck Institute for Ornithology Seewiesen, Germany article info Article history: Received 13 July 2012 Initial acceptance 20 August 2012 Final acceptance 13 December 2012 Available online 31 January 2013 MS. number: 12-00541R Keywords: acclimation activity boldness exploratory behaviour habituation individual variation novel environment personality reaction norm repeatability When individuals react differently to a stimulus, and these behavioural trait differences are repeatable (consistent) across time and contexts, the variation is known as personality (e.g. Briffa & Weiss 2010). Using data from successive behavioural assays, Biro (2012) showed that within-individual behavioural responses during initial tests in a relatively unfamiliar envi- ronment were not repeatable. In contrast, later tests conducted in the same environment, which was relatively more familiar, produced repeatable behaviours. Biro therefore suggested that a large number of observations ( 10) per individual are required to acclimate and characterize labile behavioural states rigorously. We believe that Biro’s assay unintentionally conflates different types of behavioural traits. This issue is not discussed in the commentary. We also propose that rapid repeated behavioural assays habituate individuals and the results cannot be extrapolated to assaying in a novel and unfamiliar environ- ment. Finally, we suggest that Biro’s recommended number of repeat samples per individual is often not logistically feasible in wild populations. A behavioural trait can be measured along an axis with high and low estimates of the trait represented as extremes. Réale et al. (2007) suggested that behavioural traits can be divided into five guideline categories: (1) shyebold: an individual’s response to a risky situation in a familiar environment; (2) exploratione avoidance: an individual’s response to a novel situation, such as an unfamiliar environment or object; (3) activity: the activity level of an individual in a nonrisky and familiar environment; (4) aggres- siveness: an individual’s pugnacity towards conspecifics in a social setting; and (5) sociability: an individual’s nonaggressive tendency to seek or avoid conspecifics. As all these axes are sensitive to the context in which they are measured, experiments must be designed to ensure that the intended behavioural axis is assayed (Carter et al. 2012b; Réale et al. 2007). * Correspondence: H. A. Edwards, Department of Animal and Plant Sciences, University of Sheffield, Western Bank, Sheffield S10 2TN, U.K. E-mail address: hannah892@hotmail.com (H. A. Edwards). Contents lists available at SciVerse ScienceDirect Animal Behaviour journal homepage: www.elsevier.com/locate/anbehav 0003-3472/$38.00 Ó 2013 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.anbehav.2012.12.021 Animal Behaviour 85 (2013) e1ee3