Context-dependent responses to novelty in Rainbow trout (Oncorhynchus mykiss), selected for high and low post-stress cortisol responsiveness D. Basic a, ⁎, S. Winberg b , J. Schjolden a , Å. Krogdahl a , E. Höglund c a Department of Basic Sciences and Aquatic Medicine, Norwegian School of Veterinary Science, P.O. Box 8146 Dep, Ullevålsveien 72, N-0033, Oslo, Norway b Department of Neuroscience, Uppsala University, P.O. Box 572, SE-751 23 Uppsala, Sweden c National Institute of Aquatic Resources, Technical University of Denmark, P.O. Box 101, DK-9850 Hirthals, Denmark abstract article info Article history: Received 20 June 2011 Received in revised form 4 December 2011 Accepted 19 December 2011 Keywords: Stress coping style Selection line Cortisol Behavior flexibility Novelty Context Previous studies in a rainbow trout model, selectively bred for high (HR) and low (LR) post stress plasma cortisol levels, have yielded data that are indicative of contrasting stress coping styles. Fish from the HR line have been suggested to display a more diverse behavioral repertoire in challenging situations than the LR counterpart. The present study addressed whether such variation in behavioral flexibility traits was evident in different ex- perimental settings using these selection lines. The fish were subjected to three sets of challenges (novel object test, resident–intruder test and confinement stressor test), all which were repeated a week later. Introducing a novel object evoked a divergent behavioral response in association with feeding: fish from the LR line displayed consistently suppressed feed intake while the HR fish remained unaffected. This observation was found to be re- peatable along with attack latency and movement activity from the resident–intruder and confinement stressor tests. These results indicate that the behavioral responses in this animal model are context-dependent and shed new light on the expression of behavioral flexibility. © 2011 Elsevier Inc. All rights reserved. 1. Introduction It is well known that individuals within a population often respond differently to challenges [1–4]. This intraspecific variation often com- prises suites of behavioral and physiological traits that appear to be con- sistent over time as well as across situations. Behavioral ecologists and ethologists often refer to “behavioral syndromes”, “temperament” or “personality” to characterize this phenomenon, whereas researchers in stress physiology have promoted the term “coping styles” to describe such consistency in physiological and behavioral traits [1,2,5–10]. The behavioral responses to a challenge vary along a proactive– reactive continuum. Proactive individuals are characterized by high levels of aggression, active avoidance behavior and they more readily take risks. In contrast, reactive individuals show low levels of aggres- sion, respond with immobility and are also less susceptible to take risks, but their behavioral repertoire are more flexible, to aversive stim- uli (reviewed by Koolhaas et al. [10]). Both empirical studies [11,15–18] and the theoretical framework [6,9] suggest that behavioral flexibility is an important underlying component of coping styles. In a recent review, Coppens et al. [9] expanded the concept of stress coping, and suggested that the extent to which behavior is governed by environmental stimuli is an integral component of stress coping styles. A low behavioral flexibility can thus be attributed to individuals who pay little attention to changes in the environment and are prone to routine formation in coping with challenges. Behaviorally flexible individuals on the other hand seem to be highly attentive and able to readjust their behavior in the same circumstances. The field of research regarding genetic/non-genetic factors in- volved in shaping personality traits is quite complex. For this reason, many investigators are using animal models, generated through se- lective breeding for specific phenotypes, in order to gain more insight on the causal mechanisms underlying individual variation. Over the last 15 years, several studies have emerged which are consistent with the hypothesis that stress coping styles, as it is described for mammals [5], also are present in fish [20–25,33]. The establishment of selection lines in rainbow trout (Oncorhynchus mykiss), with con- sistently high (HR) or low (LR) post-stress cortisol levels [26] has pro- vided a unique model for further exploring the heritable variation in physiological and behavioral traits in teleosts. For example, behavioral studies from these trout lines have revealed that LR fish not only become socially dominant over HR in dyadic encounters, but also resume feeding earlier and display greater boldness after transfer to a novel environment [20,24,27]. Although it has been suggested that stress reactivity and con- comitant differences in glucocorticoid release may not always co-vary with the coping style-axis [10], the HR/LR trout model shows remarkable similarities with the characteristics of proactive/reactive coping. In addi- tion to the reported differences in social dominance and boldness, a re- cent study performed by Ruiz-Gomez et al. [18] indicates that the trout lines also differ in the propensity to follow and develop routines. The Physiology & Behavior 105 (2012) 1175–1181 ⁎ Corresponding author at: Department of Basic Sciences and Aquatic Medicine, Norwegian School of Veterinary Science, P.O. Box 8146 Dep, N-0033, Oslo, Norway. Tel.: +47 22964992. E-mail address: dean.basic@nvh.no (D. Basic). 0031-9384/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.physbeh.2011.12.021 Contents lists available at SciVerse ScienceDirect Physiology & Behavior journal homepage: www.elsevier.com/locate/phb