Yolk testosterone affects growth and promotes individual-level consistency in behavioral lateralization of yellow-legged gull chicks Cristina Daniela Possenti a , Andrea Romano a, ⁎, Manuela Caprioli a , Diego Rubolini a , Caterina Spiezio b , Nicola Saino a , Marco Parolini a, ⁎ a Department of Biosciences, University of Milan, via Celoria 26, Milan, Italy b Research and Conservation Department, Parco Natura Viva, Verona, Italy abstract article info Article history: Received 21 July 2015 Revised 11 January 2016 Accepted 29 January 2016 Available online 1 February 2016 Behavioral lateralization is common in animals and may be expressed at the individual- and at the population- level. The ontogenetic processes that control lateralization, however, are largely unknown. Well-established sex-dependence in androgen physiology and sex-dependent variation in lateralization have led to the hypothesis that testosterone (T) has organizational effects on lateralization. The effects of T exposure in early life on lateral- ization can be efficiently investigated by manipulating T levels in the cleidoic eggs of birds, because the embryo is isolated from maternal and sibling physiological interference, but this approach has been adopted very rarely. In the yellow-legged gull (Larus michahellis) we increased yolk T concentration within the physiological limits and tested the effects on the direction of lateralization in two functionally fundamental behaviors (begging for parental care and escape to cover) of molecularly sexed hatchlings. We also speculated that T may intervene in regulating consistency, rather than direction of lateralization, and therefore tested if T affected the ‘repeatability’ of lateral preference in consecutive behavioral trials. T treatment had no effect on the direction of lateralization, but enhanced the consistency of lateral preference in escape responses. Sex did not predict lateralization. Neither behavior was lateralized at the population-level. We therefore showed for the first time in any species an effect of egg T on consistency in lateralization. The implications of the effect of T for the evolution of trade-offs in maternal allocation of egg hormones, and the evolutionary interpretations of findings from our studies on lateralization among unmanipulated birds are discussed. © 2016 Elsevier Inc. All rights reserved. Keywords: Larus michahellis Lateralization Maternal effect Population-level lateralization Testosterone Introduction Behavioral lateralization, whereby behavioral functions are consis- tently biased towards either the right or the left side generally as a result of asymmetric control by either brain hemisphere, is common among vertebrates (Adret and Rogers, 1989; Franklin and Lima, 2001; Reddon and Hurd, 2008; Rogers, 2002, 2008; Rutledge and Hunt, 2004; Ströckens et al., 2013; Vallortigara, 2000; Vallortigara and Rogers, 2005; see Pfannkuche et al., 2009), and has also been documented in invertebrates (Frasnelli, 2013). Behavioral lateralization in animal pop- ulations may be expressed at different levels (Rogers et al., 2013): its occurrence at the individual-level, meaning that individuals perform a behavior preferentially on either side, may or may not translate into population-level lateralization (i.e. alignment of the direction of the lateralization in the majority of individuals) depending on the relative frequency of individuals showing a specific lateral preference. In humans, for example, handedness is apparent at the individual-level and also at the population-level, because the frequency of right-handed people large- ly exceeds that of left-handers (Schaafsma et al., 2009). Conversely, in fid- dler crabs where antisymmetric distribution in the size of the chelae is associated with strongly lateralized socio-sexual behavior at the individu- al-level, no or low lateralization occurs at the population-level because right- and left-‘handed’ individuals occur at approximately the same fre- quency (Jennions and Backwell, 1996; Pratt et al., 2003). Thus, where no lateralization occurs at the population-level, individual-level lateraliza- tion may still exist. It has also been shown that lateralization may vary in ‘ strength’, with lateralized individuals differing in the intensity of the pref- erence for a specific side. Whether we see population-level lateralization or not will thus ultimately depend on genetic and epigenetic control of individual-level lateralization in combination with the frequency of indi- viduals that inherit or develop any specific lateral preference. An addition- al dimension of individual-level variation in lateralization, which has seldom been investigated, is the consistency of individuals in their lateral- ization, i.e. the extent to which any particular individual systematically prefers either side or shows no specific lateral preference. Any particular lateralized individual may show highly or, conversely, poorly consistent lateral preference (i.e. the variance in lateral preference may vary) or Hormones and Behavior 80 (2016) 58–67 ⁎ Corresponding authors at: Department of Biosciences, University of Milan, via Celoria 26, 20133 Milan, Italy. E-mail addresses: andrea.romano@unimi.it (A. Romano), marco.parolini@unimi.it (M. Parolini). http://dx.doi.org/10.1016/j.yhbeh.2016.01.007 0018-506X/© 2016 Elsevier Inc. All rights reserved. Contents lists available at ScienceDirect Hormones and Behavior journal homepage: www.elsevier.com/locate/yhbeh