Two seconds is all it takes: European starlings (Sturnus vulgaris) increase levels of circulating glucocorticoids after witnessing a brief raptor attack Blake C. Jones a, ⁎, Adam D. Smith b , Sara E. Bebus a , Stephan J. Schoech a a Department of Biological Sciences, University of Memphis, Ellington Hall 239, 3700 Walker Ave, Memphis, TN 38152, USA b Department of Natural Resources Science, University of Rhode Island, 105 Coastal Institute in Kingston, Kingston, RI 02881, USA abstract article info Article history: Received 11 May 2015 Revised 27 October 2015 Accepted 28 October 2015 Available online 10 November 2015 Keywords: Corticosterone Stress response HPA axis Avian Depredation Researchers typically study “acute” activation of the hypothalamic–pituitary–adrenal (HPA) axis by measuring levels of circulating glucocorticoids in animals that have been exposed to a predator or a cue from a predator (e.g., odor), or have experienced a standardized capture-and-restraint protocol, all of which are many minutes in duration. However, exposure to predators in the “wild”, either as the subject of an attack or as a witness to an attack, is generally much shorter as most depredation attempts upon free-living animals last b 5 s. Yet, wheth- er a stimulus lasting only seconds can activate the HPA axis is unknown. To determine if a stimulus of a few sec- onds triggers a glucocorticoid response, we measured levels of corticosterone (CORT; the primary avian glucocorticoid) in wild-caught European starlings (Sturnus vulgaris) after they witnessed a brief (b 2–8 s) raptor attack upon a conspecific, a human “attack” (i.e., a researcher handling a conspecific), and an undisturbed control. Witnesses of a raptor attack responded with CORT levels comparable to that induced by a standardized capture- and-restraint protocol. Glucocorticoid levels of individuals following the control treatment were similar to base- line levels, and those that witnessed a human “attack” had intermediate levels. Our results demonstrate that witnessing a predator attack of very brief duration triggers a profound adrenocortical stress response. Given the considerable evidence of a role for glucocorticoids in learning and memory, such a response may affect how individuals learn to recognize and appropriately react to predators. © 2015 Elsevier Inc. All rights reserved. Introduction Levels of glucocorticoids (GCs; steroid hormones released by the ad- renal cortex) at baseline serve critical metabolic functions, but increase following activation of the hypothalamic–pituitary–adrenal (HPA) axis in response to real or perceived stressors, both acute and chronic (Charmandari et al., 2005). Chronic exposure to GCs can deleteriously affect health and cognitive function in humans and animals (Cohen et al., 2007; McEwen and Sapolsky, 1995; McEwen and Wingfield, 2003; Roozendaal et al., 2009; Sheriff et al., 2009). Acute activation of the HPA axis facilitates physiological and behavioral changes that redi- rect energy away from processes not essential for immediate survival, such as reproduction, and toward processes and behaviors that may en- hance survival, such as glucose mobilization, antipredator behaviors, and memory consolidation (reviewed by Sapolsky et al., 2000; Wingfield and Ramenofsky, 1999). The conventional methods used to study acute activation of the HPA axis are often understood to be a proxy for a life-threatening encounter with a predator, even if not explicitly stated as such (Wingfield and Ramenofsky, 1999). However, the duration of experimental “acute” stressors used to date are typically longer, by orders of magnitude in most cases, than the duration of an acute predation attempt experi- enced by free-living animals. Methods that have been used include pursuit by a human for 15 min (Rödl et al., 2007), exposure to static predator mounts for 15 to 60 min (Cockrem and Silverin, 2002; Silverin, 1998), exposure to a tame individual of a predatory species that does not exhibit depredation behavior for 30 min (Canoine et al., 2002), or forced proximity to a live predator for 5 to 60 min (Canoine et al., 2002; Figueiredo et al., 2003; Manogue et al., 1975; McIntyre et al., 1999; Narayan et al., 2013; Park et al., 2008). Pakkala et al. (2013) present a notable exception with their use of a relatively brief predation stressor. They measured HPA axis responsiveness in rock pi- geons (Columba livia) that were used as lure birds to attract and trap free-living raptors (see the Discussion section for further consideration of this study). The most widely used stressor for assessing acute HPA axis respon- siveness is the standardized capture-and-restraint protocol (Wingfield, 1994), in which an animal is captured and subsequently held in a cloth bag anywhere from 5 to 60 min while a series of blood samples are collected for later measure of GC (Astheimer et al., 1995; Small and Schoech, 2015; Wingfield et al., 1992). This method has been widely used to study the acute stress response of vertebrates Hormones and Behavior 78 (2016) 72–78 ⁎ Corresponding author at: Department of Biological Sciences, University of Memphis, Ellington Hall 239, 3700 Walker Ave, Memphis, TN 38152, USA. E-mail address: bcjones8@memphis.edu (B.C. Jones). http://dx.doi.org/10.1016/j.yhbeh.2015.10.017 0018-506X/© 2015 Elsevier Inc. All rights reserved. Contents lists available at ScienceDirect Hormones and Behavior journal homepage: www.elsevier.com/locate/yhbeh