MULTIPLE SCALES OF FEAR: FORAGING BEHAVIOR OF WHITE-NAPED JAYS IN SEMIARID LANDSCAPES

MULTIPLE SCALES OF FEAR: FORAGING BEHAVIOR OF WHITE-NAPED JAYS IN SEMIARID LANDSCAPES | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL Oikos This is a preprint and has not been peer reviewed. Data may be preliminary. 24 April 2025 V1 Latest version Share on MULTIPLE SCALES OF FEAR: FORAGING BEHAVIOR OF WHITE-NAPED JAYS IN SEMIARID LANDSCAPES Authors : Maria Carolina Venáncio , Luiz Mestre , and Lorenzo Zanette 0000-0001-6497-3555 [email protected] Authors Info & Affiliations https://doi.org/10.22541/au.174549801.12515474/v1 253 views 170 downloads Contents Abstract Material and methods Results Risky Times Hypothesis Habitat Complexity Risk Mediation Hypothesis Many-Eyes Hypothesis Discussion Supplementary Material References Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Animals must constantly balance the need to find resources with the risk of predation. Not only avoiding direct encounters with predators but also assessing the overall risk of their environment using cues, social information or habitat traits at multiple spatial and temporal scales. Although such multiscale understanding of the landscape of fear has been recognized, few studies have concomitantly measured how habitat traits at different scales affect risk perception (direct or indirect). Here, we conducted a set of field-based giving-up density experiments to study risk perception on white-naped jays living in semi-arid thorn forests in northeastern Brazil. We recorded data from 23 groups of jays, ranging from 2 to 15 individuals per group, exposed to simulated predators in areas with varying habitat complexity at both local and landscape scales. Overall, our findings support the hypotheses of risky times and habitat complexity risk mediation. White-naped jays exhibited reduced food consumption in the presence of a predator and displayed increased vigilance while consuming less food in more complex habitat patches with dense canopy cover in regenerated forest areas. Finaly, we found no evidence supporting the many-eyes hypothesis; larger groups of white-naped jays did not reduce vigilance. Instead, vigilance was influenced by habitat characteristics like canopy cover and the differences between managed and regenerated forests. These findings underscore the dependency of risk perception on habitat complexity across various scales, indicating that simplifying habitats may create a less fearful environment, thereby increasing prey vulnerability by diminishing antipredator behaviours. Animals must constantly balance the need to find resources with the risk of predation. Not only avoiding direct encounters with predators but also assessing the overall risk of their environment using cues, social information or habitat traits at multiple spatial and temporal scales. Although such multiscale understanding of the landscape of fear has been recognized, few studies have concomitantly measured how habitat traits at different scales affect risk perception (direct or indirect). Here, we conducted a set of field-based giving-up density experiments to study risk perception on white-naped jays living in semi-arid thorn forests in northeastern Brazil. We recorded data from 23 groups of jays, ranging from 2 to 15 individuals per group, exposed to simulated predators in areas with varying habitat complexity at both local and landscape scales. Overall, our findings support the hypotheses of risky times and habitat complexity risk mediation. White-naped jays exhibited reduced food consumption in the presence of a predator and displayed increased vigilance while consuming less food in more complex habitat patches with dense canopy cover in regenerated forest areas. Finaly, we found no evidence supporting the many-eyes hypothesis; larger groups of white-naped jays did not reduce vigilance. Instead, vigilance was influenced by habitat characteristics like canopy cover and the differences between managed and regenerated forests. These findings underscore the dependency of risk perception on habitat complexity across various scales, indicating that simplifying habitats may create a less fearful environment, thereby increasing prey vulnerability by diminishing antipredator behaviours. Keywords: antipredator behaviour, vigilance, predation risk, giving-up density, Cyanocorax cyanoopogon. Introduction Balancing foraging and safety is a fundamental problem for most animals (McNamara and Houston, 1987; Sih, 1980; Sih, 1992). Commonly, behavioural changes offer the fastest route to solve it (Lima, 1998). Strategies to minimize direct contact with predators include behaviours that are context and state dependent, and may affect fitness directly or not (Liesenjohann et al., 2015; Lima, 1987, 1998). More generally, prey responses are shaped by the overall risk of predation, not only by direct encounters with predators (Cresswell, 2008; Lima, 1998; Nelson et al., 2004). Indirect (non-consumptive) effects are determined by perceived predation risk, i.e. the individual’s assessment of the likelihood of a predator attack, which is based on experience and imperfect knowledge of the environment (Gaynor et al., 2019a; Laundre et al., 2010). In the seminal study illustrating the concept, Laundré, Hernández and Altendorf (2001) described an increase in antipredator vigilance coupled with a decrease in foraging time among elks ( Cervus elaphus ) exposed to reintroduced wolves ( Canis lupus ) following a 50-year absence in Yellowstone National Park, USA. In addition to expanding to different predator-prey systems, subsequent work has addressed both temporal and spatial variability in the landscape of fear (Abu Baker & Brown, 2010; Gigliotti et al., 2020; Plamer et al., 2022; Sih & McCarthy, 2002) and showed that perceived risk can drive adaptations in morphology (Abbey-lee et al., 2016; Eklöv & Svanbäck, 2006), behaviour(Bishop & Byers, 2015; Magnhagen & Borcherding, 2008) and life history traits (Pettorelli et al., 2011). Environmental characteristics such as habitat complexity can directly influence the predator’s ability to find or kill prey, while also affecting the efficiency of behaviours to detect and evade predators (Atuo & O’Connell, 2017; Smith et al., 2019; Wheatley et al., 2020). Although dense vegetation might serve as a refuge for prey species of several groups: mammals (Cherry et al., 2015; Wheeler & Hik, 2014), fish (Werner et al., 2015) and birds (Rodríguez et al., 2001; von Post et al., 2012), it can also reduce visibility, which in turn could increase predation risk (Camp et al., 2012). The landscape of fear model predicts that prey should avoid predators moving through a mosaic of high to low-risk habitats (Brown & Laundre, 1999; Laundre et al., 2014). If a predator species and a prey species rely on the same habitat features or the landscape is homogeneously composed by high-risk patches, then prey species can either increase antipredator behaviours because long-term risk is predictable in certain habitat types, or the predator can increase encounter rates because predators are able to readily predict the locations of prey (Schmidt & Kuijper, 2015). Thus, areas of dense vegetation might serve as a strong proxy for long-term high predation risk in systems with ambush predators and induce increased antipredator behaviours to reduce the risk of detection or capture. Perception of predation risk and display of antipredator behaviours can depend on a variety of behavioural and environmental factors. Animals might exhibit more antipredator behaviours when exposed to an immediate risk (i.e., the ”risky times” hypothesis; Creel et al., 2008), or they can vary their level of antipredator response depending on habitat characteristics (i.e. ‘the habitat complexity risk mediation hypothesis’; Gigliotti et al., 2020). Additionally, the amount of risk of predation constantly perceived in a habitat can affect how animals respond to short acute pulses of predation risk, either by exhibiting higher responses to short acute pulses of predation risk in habitats with constant high predation risk or by exhibiting the highest response to acute pulses of risk of predation when habitat overall predation risk is low (‘predation risk allocation’ hypothesis; Dröge et al., 2017; Sih & McCarthy, 2002). Most of our understanding of risk perception comes from studies that investigate the role of either spatial or temporal variation in predation risk (Gaynor et al., 2019a; Verdolin, 2006). Less research has been conducted on reactive antipredator behaviours in relation to habitat complexity and short pulses of acute predation risk (but see: Gigliotti et al., 2021). Animals can reduce predation risk through behaviours such as minimizing the amount of time spent in risky areas (Eccard & Liesenjohann, 2008; Thaker et al., 2010), increasing vigilance behaviour (Lima, 1987), fleeing from risky situations (Cooper, 2003; Stankowich & Coss, 2007), or aggregating in larger groups (Grand & Dill, 1999). These antipredator behaviours can be beneficial for avoiding predation but can also come at energetic or fitness costs because of trade-offs with foraging or parental care (Lima, 1987). Larger groups may decrease risk of predation by dilution of the risk, increased predator detection or greater confusion of a predator during attack (Lima, 1995a; Roberts, 1996; Treisman, 1975). Therefore, animals commonly decrease their vigilance when living in groups (see Grand & Dill, 1999; Lima, 1995b, 1995a). This inverse relationship between group size and time spent scanning the environment is often attributed to the antipredator function of vigilance. Even if risk does not change with group size, each individual can reduce their personal investment in vigilance at no increased risk to themselves as more members join the group because probability of detection is maintained by the ‘many eyes’ effect (Pulliam, 1973). This is especially important for cooperative breeding animals because individuals that live and forage in groups benefit from increased vigilance and early response to threats (Bell et al., 2009; Griesser et al., 2006; Groenewoud et al., 2016). Giving up densities (GUDs) provide an index of the relative value of a given foraging patch. An animal is more likely to give up a food resource at a patch when higher quality food is available elsewhere or when there are safer foraging areas available elsewhere (Brown, 1992). Giving-up density is the remaining quantity of food in a given patch when the costs of foraging equals the benefits and the forager quits a patch (Brown & Kotler, 2004; Wheeler & Hik, 2014). Higher GUDs (i.e., greater density of food remaining) in a foraging patch or a time period relative to another may indicate that the surrounding habitat has higher quality food or that a given patch is a riskier foraging location (i.e., high predation risk) than elsewhere. Therefore, the amount of food that foragers leave in a patch (i.e., the GUD) reflects the perceived cost of foraging at that patch, such that a lower GUD indicates a lower net cost. White-naped jays in the Brazilian semiarid offer an ideal system to investigate risk perception associated to habitat complexity and the efficiency of antipredator behaviours. White-naped jays are social birds that live in groups formed by adults, young birds and offsprings of the year and occupy year-round territories in the Caatinga (Anjos & Shibatta, 2010), a mosaic of seasonally dry tropical forest with recurrent food and water shortage (Pennington et al., 2009). White-naped jays often forage at ground level (Barros et al., 2014), where they are exposed to predators like snakes and medium size cats (e.g. Puma yagouaroundi and Leopardus wiedii , pers. obs.). Adults are highly responsive to threat and when facing a potential predator respond by approaching and mobbing it, despite the potential costs of such tactics, which include, the risk of being exploited by other group members, the loss of foraging opportunities, and death of a group member (Curio, 1978; Griesser & Ekman, 2005). White-naped jays occupy habitats in a large range of complexity and group sizes varies among different habitats. These life-history traits allow us to investigate how habitat complexity affect risk perception, by using jays’ investment in antipredator behaviours, such as group size and time allocated to vigilance, to identify riskier habitats. Here we used giving-up density experiments in natural habitats varying in complexity and manipulated short-term predation risk to investigate how white-naped jays perceive and avoid predation risk. Specifically, we tested the following hypotheses, and associated predictions, related to risk perception: (1) Risky times hypothesis: white-naped jays antipredator behaviours depend on short-term predator cues. We predict that jays would quit foraging earlier (higher GUD) in response to cues of a simulated predator (2) Habitat complexity risk mediation hypothesis: white-naped jays antipredator behaviours depend on a combination of habitat complexity and short-term predator cues. We predict that jays will consume less food (higher GUD), invest more time in vigilance in areas of dense vegetation, and that this effect will be magnified in the presence of a simulated predator. (3) Many-eyes hypothesis: white-naped jay antipredator vigilance depend on group size. We predict that larger groups will allocate less time in antipredator vigilance. Material and methods Study Site We studied white-naped jay responses to predation risk in two study sites 240 km apart, in the state of Ceará, northeastern Brazil: site 1 (“Fazenda Experimental Vale do Curu”; 3°49’5.89”S; 39°20’17.81”W) and site 2 (“Fazenda Não Me Deixes”; 4°49’8.00”S; 38°58’10.37”W). Experiments were conducted in 2015 and 2018. The dominant vegetation on both study sites is Seasonally Dry Tropical Forest (Caatinga), it is an assortment of physiognomies ranging from shrub-lands to taller woodlands and forests, characterized by species of thorny deciduous trees and annual plants (Pennington et al., 2009). The climate is semi-arid, dry, with highly seasonal and irregular rainfall and two delimited seasons: a short rainy season and a long dry season with less than 100 mm (Pennington et al., 2009). Severe water and food restrictions are common during the dry season, especially in farmed areas with low plant diversity akin to other semi-arid regions around the globe (Derroire et al., 2016). At site 1 (hereafter, managed forest) we selected two fragments of recently managed forest, that have not been disturbed for at least 15 years (de Oliveira et al. 2023). The vegetation is characterized by a high density of small, still-growing trees with low heights (< 2.8 m), small stem diameters (< 31 cm), and mostly within the same-age class, a crowded canopy and little understory vegetation (de Oliveira et al. 2023). At site 2 (hereafter, regenerated forest) we selected two fragments at advanced stages of regeneration (at least 25 years), with a high number of tall trees (>10 m) with large stem diameters (>59 cm), and overall lower plant density and higher diversity (de Oliveira et al. 2023). It is characterized by abundant coarse woody debris resulting in more vertical structure, increased understory, and multiple-age classes of trees. Information on previous anthropogenic activities and a complete description of the vegetation at the study sites can be found at de Oliveira et al. (2023) and Gomes et al. (2019). Habitat Complexity Analysis Habitat complexity was estimated using an information-based metric of complexity based on Shannon’s equation for entropy. The mean information gain index (MIG) was calculated using the amount of spatial heterogeneity in an and the fraction of aspatial heterogeneity (Proulx & Parrott, 2008, 2009). MIG values range from zero, for completely uniform spatial patterns across pixels (order), which would represent a single colour, to one for completely random patterns (disorder) (Proulx & Parrott, 2008, 2009). Therefore, MIG measures increase linearly with increasing disorder in the system (Parrott, 2010). Both complete ordered systems (MIG = 0) and complete random systems (MIG = 1) are simple in the sense that they can be easily described by a probability distribution function that can reproduce the main characteristics of the system (Parrott, 2010). Hence, MIG values between the two extremes of order or disorder are associated with more spatially complex data (Parrott, 2010). Therefore, images of undifferentiated, uniform habitats have low MIG, while images of random or highly differentiated habitats have higher MIG values. MIG has been found to be positively correlated with species richness in terrestrial and aquatic habitats ranging from old-grown forests (Proulx & Parrott, 2008) and urban areas (Suarez-Rubio et al., 2018) to coral reefs (Mellin et al., 2012; Mellin et al., 2015; Tanner et al., 2015). To estimate habitat complexity at each foraging patch, we took digital photographs (EOS Rebel T5, EF-S, f3 5-5.6, Canon Inc., Tokyo, Japan). The images were taken from July to August 2015 during the dry season, on sunny days with open sky, always between 10:00 h to 12:00 h to minimize variations in light condition. The camera was installed at two adjacent sides of the quadrant with a tripod placed at a fixed height of 1 m above the ground. Images were taken using automatic mode (focal length of 55.0 mm), aperture diameter 5.6 mm and with the camera pointing to the centre of the foraging patch with the (imaginary) horizon parting the scene in half horizontally. Additionally, one image was taken with the camera installed at the centre of the quadrat facing upwards, to estimate canopy cover. Suarez-Rubio et al. (2018) demonstrated that average MIG of all images taken at a patch positively correlates with canopy cover, tree density, vertical heterogeneity, vegetation height, and number of trees; while average MIG of images of the side of the patch correlates with vegetation vertical heterogeneity; and MIG of the canopy image alone positively correlates with canopy cover and vegetation heigh. Following Suarez-Rubio et al., (2018), we calculated the average MIG of all images of one patch (MIGall, hereafter overall complexity) to estimate overall structural complexity; the average MIG of images parting the scene horizontally of a patch (MIGside, hereafter vertical heterogeneity) to estimate vegetation vertical heterogeneity; and we calculated MIG for the image of the canopy (MIGtop, hereafter canopy cover), to estimate canopy cover. Following Proulx and Parrot (2008) we compressed each image to 3.2 MP in JPEG format, and we converted the RGB (Red, Green, Blue) colour space of all images to hue, saturation, and value (HSV) to separate the pure colour component (hue) from chroma (saturation) and intensity (value). Conversion is necessary given the considerable overlap of transmittances among the three spectral bands (RGB), which is not present in the HSV space (Suarez-Rubio et al., 2018). Additionally, HSV reproduces more effectively how the human brain represents colour without the within and among variation in the RGB color space (Mellin, C., Parrott, L., Andréfouët, S., Bradshaw, C. J., MacNeil, M. A., & Caley, 2012). We used the value component of the images for analysis, which Proulx and Parrot (2008, 2009) have identified as a robust value for quantifying structural complexity in natural scenes. We calculated MIG for all images with Matlab (The MathWorks Inc., 2022; code available at http://complexity.ok.ubc.ca/projects/measuring- complexity). Giving-up Density Setup We tested hypotheses related to white-naped jay perception of predation risk by conducting a series of giving-up density experiments. We visited each study site repeatedly (84 visits/site) to select patches to perform giving-up density experiments. Only foraging patches that jays were seen feeding for three consecutive days were selected. We selected 10 foraging patches at each site. At each patch a 10 m2 quadrat was marked with a yellow string and a plastic white tray (48 × 34 × 2.3 cm) was placed at the centre. Trays were placed on the ground, always under a shrub, at least 20 m away from the forest border to minimize the effects of possible confounding factors such as weather and increased predation risk through border effect. All experiments were recorded with a digital camera (Sony DSC-H400, 20.4 MP Sony Corporation) on a tripod placed at a fixed height of 40 cm above the ground and installed 5 m away from the tray. We started recording when white-naped jays arrived at the patch. Experiments were 30 minutes duration each. One of us (M.C.B.) monitored the birds with a binocular (Nikon Monarch 7) at least 15m away from the feeder. To estimate group size, we used the maximum number of jays seen simultaneously at the patch during each trial. Foraging Threshold Analysis To assess how habitat complexity influences the antipredator behaviour and foraging tradeoff, we conducted GUD experiments for four consecutive days at N = 9 patches randomly selected at the managed forest site during the dry season (August 2018). Our goal was to separate the effects of hunger from the availability of resources in the habitat and at the tray. This experiment consisted of allowing jays to feed freely for 30 minutes a day, for 4 consecutive days at each patch. After ending the experiment, we collected and weighed all remaining seeds in each tray to the nearest 0.1 g, using a Rhino BAPRE3 electronic balance to obtain GUDs. Risky Times Hypothesis To assess how short-term predation risk influenced antipredator behaviours in white-naped jays, we conducted GUD experiments in the dry season of 2019 at N = 9 patches (same used in foraging threshold analysis). Our goal was to test if jays were responsive to a potential predator and imminent risk. We used 10 randomly selected feeding patches at site 1 and used a rubber rattle snake to cause a short pulse of acute risk. Three treatments were employed: (a) a rubber snake (Safari Ltd Incredible Creatures Eastern Diamondback Rattlesnake), (b) a tree branch (neutral object) and (c) a control (a nylon string). In each patch, a different treatment was applied each day during three consecutive days in random order. For the predator and neutral object tests the rubber snake and the tree branch, respectively, were placed on the ground 5m away from the feeder, covered in forest litter, and tied to a nylon string. After 10 minutes of the arrival of the jays at the feeder the rubber snake or branch was exposed by pulling the nylon string. After 10 minutes of exposure the object was slowly pulled out of the proximity (ca. 20 m) of the tray which was then observed for another 10 minutes and the trial was ended. After each experiment, we collected and weighed all remaining seeds in each tray to the nearest 0.1 g, using a Rhino BAPRE3 electronic balance to obtain the GUDs. Habitat Complexity Risk Mediation Hypothesis We assessed how long-term risk mediated by habitat complexity might influence perception of short-term predation risk in the dry season of 2015 at both study sites (N=20). Our goal was to separate the effect of a potential predator (artificial snake) from that of the risk associated to habitat complexity. At each site, experiments were carried out in two consecutive days. In the first day foraging behaviour was record without interference. In the second day we exposed foragers to a rubber snake. The rubber snake was placed on the ground 5m away from the feeder, covered in forest litter, and tied to a nylon string. After 10 minutes of the arrival of the jays at the feeder the rubber snake was exposed by pulling the nylon string. After 10 minutes of exposure the snake was slowly pulled out of the proximity (ca. 20 m) of the tray which was then observed for another 10 minutes and the trial was ended. All remaining seeds in each tray were collected and weighed to the nearest 0.1 g, using a Rhino BAPRE3 electronic balance to obtain the GUDs. Video Analysis We analysed all videos from the habitat complexity risk mediation hypothesis setup using Windows Movie Maker (version 16.4, Microsoft Corporation, 2012). Video recordings were analysed to quantify the overall time spent at the foraging patch, and the proportion of time spent with vigilance and feeding. Following Altmann (1974) focal samples were carried out to assess white-naped jays’ vigilance. We recorded the head position of each jay feeding at the tray at 2 s intervals. We considered jays to be vigilant when they move their heads-up and when they turned their head horizontally (Jones et al., 2007). Head turns were classified as a distinct lateral movement of the head when the head was raised, head movements when the head was down were excluded as these were classified as food-searching movements less related to vigilance (Jones et al., 2007). For each trial, we calculated the proportion of time vigilant in blocks of ten minutes. This means that, for the predator presence trials we calculate proportion of time vigilant before, during and after the predator was visible. Jays were not individually marked; therefore, all vigilance measurements were calculated as the sum of time spent vigilant by any jay at the tray divided by total time the patch was used by the group. Statistical Analysis To test the difference of habitat complexity from the two study sites, we used Mann-Whitney tests to compare overall complexity (MIGall) and vegetation vertical heterogeneity (MIGside) between sites. We used Student’s t Test to compare canopy cover (MIGtop) between sites. For foraging threshold analysis, we used a one-way analysis of variance (ANOVA) to compare GUD, to assess the potential effect of hunger state on several days of GUD experiments. Because groups of different sizes might have explored the patch in different trials, we used a one-way ANOVA with group size as factor, to test how different group sizes affect GUD. For our risky times analysis, we used general linear mixed effect models (GLMM) with gaussian family distribution, with a random effect of patch identity (ID) to compare GUD of each trial (predator present, branch, control). Our goal was to assess how short-term predation risk influenced white-naped jays’ foraging behaviour. Group size (max. no. of jays feeding at the tray) and treatment (predator present, tree branch and control) were used as predictor variables. Finally, we tested the habitat complexity risk mediation hypothesis with generalized linear mixed models (GLMM) with Beta family error distribution to account for the proportional data we collected (Brooks et al., 2017) to assess how risk mediated by habitat complexity might influence perception of short-term predation risk and antipredator behaviours. We used data from all trials at the two study sites (N =20). GUD were used as response variable while: site, MIG, presence of predator, group size and proportion of time vigilant were included as explanatory variables. We did not include proportion of time feeding because it is strongly correlated with proportion of time vigilant (cor = 0.749, t = 6.974, df = 38, p-value <0.0001). We also included the two-way interactions between site and MIG; site and proportion of time vigilant; group size and proportion of time vigilant as explanatory variables. Patch identity was included as a random intercept factor to account for our repeated temporal GUD measurements per patch. As all three measurements of MIG are correlated, each type of MIG (all, sides and top) was used in a different set of GLMMs, repeating all other variables. We tested the many-eyes hypothesis with a set of GLMMs to assess the effects of habitat complexity and group size on the proportion of time vigilant was analysed with. The overall proportion of time vigilant was used as a response variable, MIG and group size were used as predictor variables. The interactions between group size and site, MIG and group size were also included as predictor variables to account for the potential effects of different habitat scales on risk perception. Patch was included as a random factor to account for repeated temporal measurements at each patch. All sets of GLMM were fitted using glmmTMB R package (Brooks et al., 2017) and model residuals were inspected using simulation-based method implemented in the “DHARMa” package v. 0.4.3 (Hartig and Lohse, 2021). Finally, we used “car” package v. 3.0-11 (Fox & Weisberg, 2020) to assess homoscedasticity and independence through graphical diagnostic. Model selection was performed using “MuMIn” package v. 1.43.17 (Bartón, 2015) to find the subset of predictors that best explained the data. This procedure ranks all possible candidate models by AICc values (Burnham & Anderson, 2002, 2004). We considered competing models within a ΔAICc < 4 as important in explaining GUD. We evaluated model-averaged estimates for variables of interest in competing models and calculated unconditional standard errors and 95% confidence limits (Burnham & Anderson, 2004; Nakagawa & Schielzeth, 2013). Following model averaging, we obtained p- values based on top models within a ΔAICc < 4 (Grueber et al., 2011). All statistical analysis were performed in R v. 3.5.1 (R Core Team, 2018) and assessed significance using a value of 0,05. Ethical notes For this study we documented the natural foraging behaviour of animals in a natural environment. There was no physical treatment, aggressive interaction, or risk of injury for the animals. Results We collected data in 20 foraging patches of white-naped jays and conducted 94 GUD trials. We used 36 GUD trials to test foraging thresholds, 18 trials to test our predictions derived from the risky times hypothesis and 40 trials to test predictions from the habitat complexity risk mediation hypothesis. We tested 8 groups of jays ranging from 3 to 8 individuals on site 1 in 2015 (mean ± S.E. = 4.70 ± 0.42) and 9 groups ranging from 15 to 2 jays in 2018 (mean ± S.E. = 6.38 ± 3.04). At site 2, we tested 6 groups ranging from 4 to 8 jays (mean ± S.E. = 4.65 ± 0.20) in 2015. The locations where we conducted trials varied in terms of vegetation vertical heterogeneity and canopy cover. Our foraging threshold analysis to control for the potential effect of state on GUD, showed that group size significantly affected the amount of food consumed (F = 31,82 p <0,01 df =1) and that there was no relationship between number of days jays fed on the tray and the amount of food consumed (F = 0,56 p = 0,65 df = 3, Figure S1). Hence, this variable (day of trial) was not considered in further analysis. Therefore, GUD is a reliable measurement of white-naped jays foraging investment for up to three consecutive days. Habitat Structure As expected, patch habitat complexity differed significantly between sites, in conformity with the difference of regeneration process of each site. Vegetation from patches from the recently managed forest (site 1) were more vertically heterogenous (MIGside; N = 20, W = 276, p-value = 0.04) and canopies were more open (MIGtop; N = 20, t = -4.62, df = 35.78, p<0.01) than patches from the regenerated forest (site 2). Interestingly, overall habitat complexity (MIGall) did not differ between patches from the recently managed forest and the regenerated forest (N = 20, W = 228, p-value = 0.45), which suggests that both are heterogeneous landscapes. The managed forest site is composed by patches within a wide range of MIG values, suggesting that at the landscape scale it is more heterogeneous, and the vertical structure of woody vegetation varies strongly, ranging from patches with simplified vegetation, such as uniformly still-growing trees without understory, to patches with complex, multi- layered vegetation filled with trees and shrubs of different sizes. On the contrary, the regenerated forest site is mainly composed by patches of high habitat complexity with a high number of tall trees, abundant coarse woody debris increased understory, and dense canopy. FIGURE 1. Comparisons of patch habitat complexity based on values of mean information gain (MIG) at two study sites in north-eastern Brazil. A) MIGall values summarizing overall habitat complexity did not differ between patches from the recently managed forest and the regenerated forest (N = 20, W = 228, p-value = 0.45). B). MIGside values summarizing vertical habitat structure. Patches in the recently managed site were more vertically heterogenous (N = 20, W = 276, p-value = 0.04) than patches from the regenerated forest site. C) MIGtop values used to estimate canopy cover. Patches from the recently managed site had more variable open canopies than patches from the regenerated forest site (N = 20, t = -4.62, df = 35.78, p<0.01). Risky Times Hypothesis White-naped jays changed foraging behaviour following predator presence (i.e. rubber snake) and branch (n = 9, mean GUD ± SE =42.11 ±1.74 and 44.00 ±1.84 g, respectively) compared to control trials (n = 9, mean GUD ± SE =30.11 ±5.03 g, Figure S2). Specifically, our results confirmed that white-naped jays use antipredator strategies in response to short-term pulses of acute risk. These results, combined with the consistently jumpy and more agitated behaviour of the jays recorded in the presence of the rubber snake and the branch, when they were pulled out, confirm that both stimuli are aversive and jays classified them as threats (McCune et al., 2018). Habitat Complexity Risk Mediation Hypothesis The presence of predator significantly decreased the amount of food consumed on all patches. However, as predicted by the habitat complexity risk mediation hypothesis, white-naped jays from different sites responded differently to a pulse of acute risk. White-naped jays from the regenerated forest site consumed less food than jays in the managed forest site (t = -3.70, df = 25.82, p-value = 0.00, Figure 2; Table 1), both in the presence (Site1: mean ± sd = 43.24 ± 5.47; Site 2: mean ± sd = 36.13 ± 10.51) and in the absence of predator (Site1: mean ± sd = 39.41 ± 7.02; Site 2: mean ± sd = 19.59 ± 14.44). As expected, a short pulse of risk elicited different antipredator strategies depending on the environmental features of the study site. Table 1. Summary results of generalized linear mixed models used to analyse the effect of site features, predator presence, vigilance, and patch habitat complexity (MIGtop, MIGall and MIGside) on the fraction of food left by the jays on experimental trays (GUD) Data was collected on 20 foraging patches in two study sites in the Brazilian semi-arid region. Estimated values are the conditional average of parameters and confidence interval obtained GLLMM model selection. Model with MIGall Intercept 1.70 -2.69 6.09 0.45 Site -0.53 -4.49 3.42 0.79 Group Size -0.31 -0.95 0.33 0.35 Presence of predator 1.14 0.25 2.04 0.01 Proportion of time vigilant -4.84 -12.76 3.08 0.23 Site:proportion of time vigilant 6.77 0.72 12.81 0.03 Site:presence of predator -1.01 -2.19 0.17 0.09 Group size:proportion of time vigilant 1.46 -0.89 3.82 0.22 MIGall -2.64 -9.05 3.78 0.42 Site:MIGall 8.04 -0.58 16.66 0.07 Model with MIGside Intercept 1.41 -2.94 5.76 0.53 Site -0.23 -4.02 3.55 0.90 Group size -0.30 -0.92 0.32 0.35 Presence of predator 1.14 0.25 2.03 0.01 Proportion of time vigilant -4.98 -13.19 3.23 0.23 Site:proportion of time vigilant 6.69 0.69 12.70 0.03 Site:presence of predator -1.00 -2.18 0.18 0.10 Group size:proportion of time vigilant 1.46 -0.89 3.82 0.22 MIGside 0.92 -2.54 4.38 0.60 Model with MIGtop Intercept 2.32 -1.98 6.62 0.29 Site -1.28 -5.26 2.70 0.53 Group size -0.29 -0.88 0.30 0.33 MIGtop -3.91 -7.90 0.07 0.05 Presence of predator 1.17 0.27 2.07 0.01 Proportion of time vigilant -4.62 -12.48 3.24 0.25 Site:proportion of time vigilant 6.67 0.82 12.51 0.03 Site:MIGtop 7.13 0.26 14.01 0.04 Site:presence of predator -1.03 -2.19 0.13 0.08 Group size:proportion of time vigilant 1.45 -0.72 3.61 0.19 Figure 2. Relationship between white-naped jays foraging behaviour (GUD) and study site, in the presence or absence of a predator. Experiments occurred at two study-sites in the Brazilian semi-arid: managed forest is a fragment of Seasonally Tropical Dry Forest of at least 15 years in regeneration process; Regenerated Forest is a fragment of Seasonally Tropical Dry Forest of at least 25 years of regeneration. Similarly, the interaction between site and overall patch habitat complexity and site and canopy cover also had a negative effect on the amount of food consumed (higher GUD values), suggesting that the interplay between patch habitat complexity, specially canopy cover, and site’s environmental features affect risk perception in white-naped jays. Hence, as predicted by the habitat complexity risk mediation hypothesis, our results indicate that both habitat complexity and the presence of a predator affected risk perception in white-naped jays. There was no direct effect of overall patch habitat complexity and patch vegetation vertical heterogeneity on the amount of food consumed at the patches (Table 1). However, as predicted by the habitat complexity risk mediation hypothesis, jays at patches with a less dense canopy cover in the managed forest site, consumed more food relative to patches with a denser canopy from the regenerated forest site (Table 1, Fig.3). Canopy cover differed significantly between sites (t = -4.62, df = 35.78, p values and patches have a more open canopy (Mean MIGtop ± sd= 0.35 ± 0.10) compared to patches at the regenerated forest site (Mean MIGtop ± sd = 0.49 ± 0.08, Fig. 2). Similarly, GUD values were more variable (mean GUD = 27.86± 14.93, N = 20), and jays consumed more food in patches of more open canopy cover without a model predator in the managed forest site. In contrast, jays in the regenerated forest site consumed less food and did not adjust their intake according to canopy cover (mean GUD = 41.33 ± 6.44, N = 20, Fig.2) suggesting that the risk of foraging in habitats with dense canopy is constantly higher. [1]¿p#1 Figure 3. Relationship between white-naped jays foraging behaviour (GUD) and canopy cover (Mean information gain -, MIGtop), in the presence or absence of a predator. Experiments with white-naped jays occurred at two study-sites in Bazilian semi-arid: managed forest is a fragment of Seasonally Tropical Dry Forest of at least 15 years in regeneration process, with a more open canopy (mean ± sd= 0.35 ± 0.10); Regenerated Forest is a fragment of Seasonally Tropical Dry Forest of at least 25 years of regeneration with patches with denser canopies (mean ± sd = 0.49 ± 0.08). As expected, the interplay of site and proportion of time vigilant significantly affected the amount of food consumed (p = 0.03, Table 1). At the managed forest, groups of jays presented a wider range of values of vigilance and GUD. On the contrary, white-naped jays at the regenerated forest maintained a higher threshold of vigilance independently of the presence of predator, which resulted on higher and less variable GUD values suggesting that the effect of risk mediated by habitat complexity was constant and strong in that site (Figure 4). FIGURE 4. Giving-up density (GUD) as a function of proportion of time that groups of White-naped jays spent vigilant at 20 experimental patches in Brazilian semi-arid in the presence and in the absence of predator. Many-Eyes Hypothesis There was no significant relationship between group size and change in white-naped jays’ vigilance (Table 2). Contrary to what we expected in the many-eyes hypothesis, larger group sizes do not reduce their overall proportion of time spent with vigilance. However, our results show that the interplay between site features and habitat complexity affect changes in white-naped jays’ vigilance. Specifically, the interaction of overall habitat complexity and canopy cover affected vigilance (Table 2). Table 2. Summary of results of generalized linear mixed models including patch indentity as a random factor for proportion of time vigilant as a function of study site, presence of predator, group size and patch habitat complexity (i.e., estimated by MIGtop, MIGall and MIGside). Data was collected on 20 foraging patches at two study sites on the Brazilian semi-arid region for the white-naped jay ( Cyanocorax cyanopogon ). Estimated values are the conditional average of parameters and confidence interval obtained GLLMM model selection. Models with MIGall Intercept 0.41 -0.09 0.91 0.11 Site -0.24 -1.05 0.57 0.56 MIGall 0.19 -0.92 1.29 0.74 Site:MIGall 1.22 0.02 2.42 0.05 Group size -0.02 -0.05 0.01 0.23 Presence of predator -0.05 -0.14 0.04 0.29 Site:presence of predator 0.08 -0.01 0.17 0.09 Site:group size -0.04 -0.10 0.03 0.30 Group size:presence of predator 0.03 -0.01 0.06 0.16 Model with MIGside Intercept 0.43 0.03 0.83 0.03 Site -0.03 -0.58 0.52 0.92 MIGside 0.18 -0.71 1.08 0.69 Site:MIGside 0.89 -0.20 1.98 0.11 Group size -0.02 -0.05 0.01 0.22 Presence of predator -0.03 -0.21 0.14 0.70 Site:presence of predator 0.08 -0.01 0.17 0.08 Group size:presence of predator 0.03 0.00 0.07 0.08 MIGside:presence of predator -0.43 -0.84 -0.03 0.04 Model with MIGtop Intercept 0.39 0.14 0.65 0.00 MIGtop 0.30 -0.36 0.96 0.37 Presence of predator -0.06 -0.18 0.06 0.35 Site -0.17 -0.77 0.43 0.58 Site:MIGtop 1.11 0.05 2.18 0.04 Group size -0.02 -0.04 0.01 0.29 MIGtop:presence of predator 0.25 -0.15 0.64 0.23 Site:presence of predator 0.08 -0.01 0.17 0.10 Group size:presence of predator 0.03 -0.01 0.06 0.17 Specifically, the interaction of overall habitat complexity and canopy cover affected vigilance (Table 2). White-naped jays from the managed site had a wide range vigilance time and spent more time vigilant on patches with dense vegetation, higher overall habitat complexity (Figure 5) and dense canopy cover (Figure 5). On the contrary, white-naped jays from the regenerated forest site seems to show a higher threshold of vigilance independently of patch overall habitat complexity (Figure 5) and canopy cover (Figure 5). This suggest that patches at the regenerated forest site have a higher and constant predation risk. Moreover, these results support that white-naped jays use vigilance as an antipredator strategy to mitigate risk when foraging. Figure 5. Relationship between white-naped jays’ vigilance and A) habitat complexity B) canopy cover, on two study-sites in Brazilian semi-arid: managed forest is a fragment of Seasonally Tropical Dry Forest of at least 15 years in regeneration process; Regenerated Forest is a fragment of Seasonally Tropical Dry Forest of at least 25 years of regeneration. Discussion We found that white-naped jay foraging behaviour varied based on two scales of habitat characteristics: site and patch habitat complexity. White-naped jays consumed less food on patches with dense vegetation and dense canopy cover in the regenerated forest, which indicates and overall higher predation risk mediated by habitat complexity. Our results show that changes on foraging behaviour were, at least partially, consequence of the amount of time invested in vigilance. Our results also indicate that jay´s investment in vigilance is reactive in the managed forest, i.e. is changed according to the current level of risk. In contrast, in the regenerated forest, is predictive, i.e. jays maintain a threshold of vigilance independent of small-scale habitat features and the presence of a predator. Our results highlight the need to consider different scales of habitat when analysing risk perception in natural habitats. We found that white-naped jays exhibited a strong antipredator behavioural response to short-term risk, as expected by the risky times hypothesis. Similar evidence has been found for mammals (Gigliotti et al., 2021), birds (Griesser & Nystrand, 2009) and invertebrates (Steinhoff et al., 2020). White-naped jays are active foragers that exploit all vegetation strata, but spend considerable time foraging on the forest ground (Barros et al., 2014); thus, it makes sense that when faced with visual and sound cues of an ambush predator, such as a snake, white-naped jay exhibited higher vigilance and rapidly abandoned the patch. When faced with an immediate threat, the benefits of antipredator behaviour likely outweigh any potential costs (Lima & Dill, 1990). Additionally, the strongest antipredator response to the presence of a fake predator (higher GUD values and higher vigilance times) were observed on white-naped jays from patches with high habitat complexity and dense canopy in the regenerated forest site. This suggests that patches with dense vegetation and dense canopy are perceived as riskier patches, like other studies with prey species that are consumed by ambush predators (Kotler et al., 1993, 2004). Given that jays visually identify ambush predators and display mobbing behaviour towards them, it is likely that patches with dense canopy are perceived as riskier habitats, rather than the habitat with more abundant refuges and escape opportunities. It is likely that a dense canopy cover, like in patches from the regenerated older forest, decreases the light in the environment and the efficiency of vigilance. Similarly, gerbils consumed less food and spent more time vigilant than foraging, when they used patches with higher predation risk (i.e. in the presence of predators and in bright moonlight) (Embar et al., 2011). In the same experiment Embar et al. (2011) demonstrated that when gerbils had their sightline blocked, the efficiency of vigilance to detect predators was reduced, causing gerbils to use their foraging patch less due to the perceived high risk of predation (i.e., bright moonlight). Moreover, in the presence of a predator, the gerbils response to short-term risk was more severe – harvesting less food and spending less time in vigilance – in the patches with the increased risk (Embar et al., 2011). Additionally, our results support the risky times risky places hypothesis. We found that white-naped jays responded differently to short-term risk based both on small-scale (i.e., patch canopy cover) and large-scale (i.e., site) habitat characteristics. White-naped jays consumed more food at patches with more open canopies than jays at patches with dense canopy. Moreover, we observed a difference in the pattern of food consumption related to site. Instead of maintaining a low baseline of foraging intake, jays at patches in the managed forest adjusted their foraging behaviour based on the interplay of canopy cover and presence of a predator. On the contrary, white-naped jays from patches in the regenerated forest maintained low food intake despite the presence of a predator. While it might seem counter-intuitive at first, simulated short pulses of acute risk had a weaker effect on white-naped jays’ behaviour than constant risk associated with habitat structure. Indeed, jays reduced foraging in the presence of the predator in both sites, however this behaviour does not suffice to explain the overall difference in the amount of food consumed in each site. This finding is in line with results from a meta-analysis by Verdolin (2006), although considering mainly rodents, it revealed that habitat characteristics elicit stronger behavioural adjustments in prey organisms than cues of live predators. Brown & Kotler (2004) suggests that given the large fitness cost of predation, some prey animals may have evolved a tendency to perceive a higher probability of predation, and to prioritize safety instead of foraging, maintaining a baseline antipredator behaviour, which supports our assumption that patches in the regenerated forest with denser canopy cover are perceived as higher predation risk habitats by the jays. Similarly, gerbils display a stronger antipredator response in the presence of a predator in the patches with the increased risk, harvesting less food and spending less time and vigilance (Embar et al., 2011). By perceiving high risk as a default, prey may exhibit more homogenous antipredator behavioral responses when compared with the heterogeneous landscape of predation risk (Gaynor et al., 2019b). Additionally, we found that white-naped jays differentially modified vigilance based on the interplay between patch habitat complexity, canopy cover and site. Jays at patches in the managed forest adjusted their vigilance to periods of short-term risk (i.e., presence of predator), while jays in the regenerated forest maintained a high investment in vigilance even in the absence of predator. There are two possible explanations for these differences First, at the landscape level the regenerated forest is more homogeneous than the managed forest. Therefore, most of the patches from the regenerated forest have similar habitat complexity and canopy cover. Hence, as reported for temperate forests, predators and resources are likely to be predictably distributed among patches (Schmidt & Kuijper, 2015), and jays maintain higher levels of vigilance in response to the increased risk associated to vegetation. Given that jays visually identify ambush predators and display mobbing behaviour towards them, it is likely that patches with dense canopy are perceived as riskier habitats, rather than safer places with more abundant refuges and escape opportunities. It is likely that a dense canopy cover, like in patches from regenerated older forests, decreases the light in the environment and the efficiency of vigilance, which increases the difficulty in finding ambush predators that benefit from concealment in dense vegetation. Hence, a constant high investment in antipredator behaviour might may be required and less time will be available for foraging. Second, the regenerated forest in relatively older and may have a higher density of jay predators. Usually more complex habitats are associated with higher species diversity (Tews et al., 2004; Yang et al., 2015). and higher levels of trophic interactions (Kadmon & Allouche, 2007). As opposed to research on other social birds, we did not find support to the many-eyes hypothesis. Although vigilance time was not affected by group size, we cannot entirely exclude other antipredator benefits of group size, such as increased predator detection and risk dilution. Differences found in white-naped jays’ vigilance levels among sites may be explained by differences in territory features such as risk of predation and food availability, hence the cost of vigilance in terms of lost feeding time might vary, causing the intensity of the group-size effect to vary (Sansom et al., 2008). Enhanced safety from predators by cooperative defence and dilution effect may be the primary benefits of sociality for cooperative birds, especially in high-risk habitats (e.g. Guindre-Parker and Rubenstein 2020). Because we found variation in vigilance behaviour to different sites and levels of habitat complexity, our results suggest the importance of considering the context of habitat mediated risk, rather than attempting to simplify the complexity of factors that affect vigilance. Our results offer insight into how social birds might perceive predation risk and how risk and antipredator behaviours are related to habitat complexity. Many studies use foraging behaviour and vigilance as a proxy for predation risk (Bedoya-Perez et al., 2013; Moll et al., 2017). However, our research adds to a growing body of literature that indicates that predation risk needs to be considered across multiple spatial and temporal scales because behavioural responses do not always match the direct risk of predation and habitat complexity can influence these responses (Gaynor et al., 2019; Moll et al., 2017). Our results also suggest that habitat complexity might serve as a better proxy for long-term predation risk. Inferences into the mechanisms underlying variance in antipredator behaviours should be tempered by a caveat inherent in our design. The shape of foraging-predation risk trade-off is also influenced by variation in local resource abundance (Houston, McNamara, & Hutchinson, 1993), which we did not explicitly account for and may be the ultimate driver of short-term behavioural responses (Naman, 2019). While these uncertainties cannot be resolved with the data at hand, they do not change the main conclusions of our study so much as offer alternative mechanisms for them. Ultimately, perceived risk still altered foraging behaviour leading to different strategies related to habitat complexity. Thus, our study provides an important in situ demonstration in a tropical social bird of the effect of predation risk mediated by habitat complexity on foraging-safety trade-off. Supplementary Material File (oik-11614-file003.docx) Download 21.90 KB References 1. Abbey-lee, R. N., Mathot, K. J., & Dingemanse, N. J. (2016). Behavioral and morphological responses to perceived predation risk : a field experiment in passerines. 27, 857–864. https://doi.org/10.1093/beheco/arv228 Crossref Google Scholar 2. Abu Baker, M. A., & Brown, J. S. (2010). Islands of fear: Effects of wooded patches on habitat suitability of the striped mouse in a South African grassland. Functional Ecology, 24(6), 1313–1322. https://doi.org/10.1111/j.1365-2435.2010.01757.x Crossref Google Scholar 3. Altmann, J. (1974). Observational Study of Behavior : Sampling Methods Author ( s ): Jeanne Altmann. Behaviour, 49(3), 227–267. http://www.jstor.org/stable/4533591 Google Scholar 4. Anjos, L. Dos, & Shibatta, O. A. (2010). Sobre a ocupação de habitat de algumas gralhas neotropicais na América do Sul, gênero Cyanocorax. Semina: Ciências Biológicas e Da Saúde, 31(2), 153. https://doi.org/10.5433/1679-0367.2010v31n2p153 Crossref Google Scholar 5. Atuo, F. A., & O’Connell, T. J. (2017). The landscape of fear as an emergent property of heterogeneity: Contrasting patterns of predation risk in grassland ecosystems. Ecology and Evolution, 7(13), 4782–4793. https://doi.org/10.1002/ece3.3021 Crossref Google Scholar 6. Barros, R. A. M., Costa, C. A., & Pascotto, M. C. (2014). Diet and feeding behavior of the White-naped Jay, Cyanocorax cyanopogon (Wied, 1821) (Aves, Passeriformes, Corvidae) in a disturbed environment in central Brazil. Brazilian Journal of Biology, 74(4), 899–905. https://doi.org/10.1590/1519-6984.09313 Crossref Google Scholar 7. Bedoya-Perez, M. A., Carthey, A. J. R., Mella, V. S. A., McArthur, C., & Banks, P. B. (2013). A practical guide to avoid giving up on giving-up densities. Behavioral Ecology and Sociobiology, 67(10), 1541–1553. https://doi.org/10.1007/s00265-013-1609-3 Crossref Google Scholar 8. Bell, M. B. V., Radford, A. N., Rose, R., Wade, H. M., & Ridley, A. R. (2009). The value of constant surveillance in a risky environment. Proceedings of the Royal Society B: Biological Sciences, 276(1669), 2997–3005. https://doi.org/10.1098/rspb.2009.0276 Crossref Google Scholar 9. Bishop, M. J., & Byers, J. E. (2015). Predation risk predicts use of a novel habitat. Oikos, 124(9), 1225–1231. https://doi.org/10.1111/oik.01967 Crossref Google Scholar 10. Bonte, D., Van Dyck, H., Bullock, J. M., Coulon, A., Delgado, M., Gibbs, M., Lehouck, V., Matthysen, E., Mustin, K., Saastamoinen, M., Schtickzelle, N., Stevens, V. M., Vandewoestijne, S., Baguette, M., Barton, K., Benton, T. G., Chaput-Bardy, A., Clobert, J., Dytham, C., … Travis, J. M. J. (2012). Costs of dispersal. Biological Reviews, 87(2), 290–312. https://doi.org/10.1111/j.1469-185X.2011.00201.x Crossref Google Scholar 11. Brooks, M. E., Kristensen, K., van Benthem, K. J., Magnusson, A., Berg, C. W., Nielsen, A., Skaug, H. J., Mächler, M., & Bolker, B. M. (2017). glmmTMB balances speed and flexibility among packages for zero-inflated generalized linear mixed modeling. R Journal, 9(2), 378–400. https://doi.org/10.32614/rj-2017-066 Crossref Google Scholar 12. Brown, Joel S, Laundre, G. (1999). Brown et al 1999) the ecology of fear_optimal foraging, game theory and trophic interactions.pdf. Google Scholar 13. Brown, J. S., & Kotler, B. P. (2004). Hazardous duty pay and the foraging cost of predation. Ecology Letters, 7(10), 999–1014. https://doi.org/10.1111/j.1461-0248.2004.00661.x Crossref Google Scholar 14. Burnham, K. P., & Anderson, D. R. (n.d.). Model Selection and Multimodel Inference. Google Scholar 15. Burnham, K. P., & Anderson, D. R. (2004). Multimodel inference: Understanding AIC and BIC in model selection. Sociological Methods and Research, 33(2), 261–304. https://doi.org/10.1177/0049124104268644 Crossref Google Scholar 16. Camp, M. J., Rachlow, J. L., Woods, B. A., Johnson, T. R., & Shipley, L. A. (2012). When to Run and When to Hide: The Influence of Concealment, Visibility, and Proximity to Refugia on Perceptions of Risk. Ethology, 118(10), 1010–1017. https://doi.org/10.1111/eth.12000 Crossref Google Scholar 17. Cherry, M. J., Conner, L. M., & Warren, R. J. (2015). Effects of predation risk and group dynamics on white-tailed deer foraging behavior in a longleaf pine savanna. Behavioral Ecology, 26(4), 1091–1099. https://doi.org/10.1093/beheco/arv054 Crossref Google Scholar 18. Cooper, W. E. (2003). Effect of risk on aspects of escape behavior by a lizard, Holbrookia propinqua, in relation to optimal escape theory. Ethology, 109(8), 617–626. https://doi.org/10.1046/j.1439-0310.2003.00912.x Crossref Google Scholar 19. Creel, S., Winnie, J. A., Christianson, D., & Liley, S. (2008). Time and space in general models of antipredator response: tests with wolves and elk. Animal Behaviour, 76(4), 1139–1146. https://doi.org/10.1016/j.anbehav.2008.07.006 Crossref Google Scholar 20. Cresswell, W. (2008). Review: non-lethal effects of predation in bird. Ibis, 150, 3–17. https://doi.org/10.1111/j.1474-919X.2007.00793.x Crossref Google Scholar 21. Curio, E. (1978). The Adaptive Significance of Avian Mobbing: I. Teleonomic Hypotheses and Predictions. Zeitschrift Für Tierpsychologie, 48(2), 175–183. https://doi.org/10.1111/j.1439-0310.1978.tb00254.x Crossref Google Scholar 22. de Oliveira, C. C., Martins, F. R., Souza, B. C., de Sa Barretto Sampaio, E. V., Loiola, M. I. B., & Soares, A. A. (2023). Resilience and successional trends of woody vegetation in seasonally dry tropical forests. Forestry: An International Journal of Forest Research, 96(5), 740–753. https://doi.org/10.1093/forestry/cpad010 Crossref Google Scholar 23. Derroire, G., Balvanera, P., Castellanos-castro, C., Kennard, D. K., Lebrija-trejos, E., Leiva, J. A., Powers, J. S., Rico-gray, V., Tigabu, M., & John, R. (2016). Resilience of tropical dry forests – a meta-analysis of changes in species diversity and composition during secondary succession. Oikos. Google Scholar 24. Dröge, E., Creel, S., Becker, M. S., & M’soka, J. (2017). Spatial and temporal avoidance of risk within a large carnivore guild. Ecology and Evolution, 7(1), 189–199. https://doi.org/10.1002/ece3.2616 Crossref Google Scholar 25. Eccard, J. A., & Liesenjohann, T. (2008). Foraging decisions in risk-uniform landscapes. PLoS ONE, 3(10), 1–4. https://doi.org/10.1371/journal.pone.0003438 Crossref Google Scholar 26. Eklöv, & Svanbäck. (2006). Predation Risk Influences Adaptive Morphological Variation in Fish Populations. The American Naturalist, 167(3), 440. https://doi.org/10.2307/3844765 Crossref Google Scholar 27. Embar, K., Kotler, B. P., & Mukherjee, S. (2011). Risk management in optimal foragers: The effect of sightlines and predator type on patch use, time allocation, and vigilance in gerbils. Oikos, 120(11), 1657–1666. https://doi.org/10.1111/j.1600-0706.2011.19278.x Crossref Google Scholar 28. Fox, J., & Weisberg, S. (2020). Using car and effects Functions in Other Functions. Cran, 3, 1–5. Google Scholar 29. Gaynor, K. M., Brown, J. S., Middleton, A. D., Power, M. E., & Brashares, J. S. (2019a). Landscapes of Fear: Spatial Patterns of Risk Perception and Response. Trends in Ecology and Evolution, 34(4), 355–368. https://doi.org/10.1016/j.tree.2019.01.004 Crossref Google Scholar 30. Gaynor, K. M., Brown, J. S., Middleton, A. D., Power, M. E., & Brashares, J. S. (2019b). Landscapes of Fear: Spatial Patterns of Risk Perception and Response. Trends in Ecology and Evolution, 34(4), 355–368. https://doi.org/10.1016/j.tree.2019.01.004 Crossref Google Scholar 31. Gigliotti, L. C., Slotow, R., Hunter, L. T. B., Fattebert, J., Sholto-Douglas, C., & Jachowski, D. S. (2020). Habitat complexity and lifetime predation risk influence mesopredator survival in a multi-predator system. Scientific Reports, 10(1), 1–10. https://doi.org/10.1038/s41598-020-73318-3 Crossref Google Scholar 32. Gigliotti, L. C., Slotow, R., Sholto-Douglas, C., de Vos, C., & Jachowski, D. S. (2021). Short-term predation risk and habitat complexity influence cheetah antipredator behaviours. Animal Behaviour, 178, 175–184. https://doi.org/10.1016/j.anbehav.2021.06.003 Crossref Google Scholar 33. Gomes, F. M., Oliveira, C. C. De, Rocha Miranda, R. Da, Costa, R. C. Da, & Loiola, M. I. B. (2019). Relationships between soil seed bank composition and standing vegetation along chronosequences in a tropical dry forest in north-eastern Brazil. Journal of Tropical Ecology, 35(4), 173–184. https://doi.org/10.1017/S0266467419000130 Crossref Google Scholar 34. Grand, T., & Dill, L. (1999). The effect of group size on the foraging behaviour of juvenile coho salmon: reduction of predation risk or increased competition? Animal Behaviour, 58(2), 443–451. https://doi.org/10.1006/anbe.1999.1174 Crossref Google Scholar 35. Griesser, M., & Ekman, J. (2005). Nepotistic mobbing behaviour in the Siberian jay, Perisoreus infaustus. Animal Behaviour, 69(2), 345–352. https://doi.org/10.1016/j.anbehav.2004.05.013 Crossref Google Scholar 36. Griesser, M., & Nystrand, M. (2009). Vigilance and predation of a forest-living bird species depend on large-scale habitat structure. Behavioral Ecology, 20(4), 709–715. https://doi.org/10.1093/beheco/arp045 Crossref Google Scholar 37. Griesser, M., Nystrand, M., & Ekman, J. (2006). Reduced mortality selects for family cohesion in a social species. Proceedings. Biological Sciences / The Royal Society, 273(1596), 1881–1886. https://doi.org/10.1098/rspb.2006.3527 Crossref Google Scholar 38. Groenewoud, F., Frommen, J. G., Josi, D., Tanaka, H., Jungwirth, A., & Taborsky, M. (2016). Predation risk drives social complexity in cooperative breeders. Proceedings of the National Academy of Sciences, 113(15), 4104–4109. https://doi.org/10.1073/pnas.1524178113 Crossref Google Scholar 39. Grueber, C. E., Nakagawa, S., Laws, R. J., & Jamieson, I. G. (2011). Multimodel inference in ecology and evolution: Challenges and solutions. Journal of Evolutionary Biology, 24(4), 699–711. https://doi.org/10.1111/j.1420-9101.2010.02210.x Crossref Google Scholar 40. Guindre-Parker, S., & Rubenstein, D. R. (2020). Surntvival benefits of group living in a fluctuating environme. American Naturalist, 195(6), 1027–1036. https://doi.org/10.1086/708496 Crossref Google Scholar 41. Hartig F, Lohse L (2021) DHARMa: Residual Diagnostics for Hierarchical (Multi-Level/Mixed) Regression Models Google Scholar 42. Jones, K. A., Krebs, J. R., & Whittingham, M. J. (2007). Vigilance in the third dimension: head movement not scan duration varies in response to different predator models. Animal Behaviour, 74(5), 1181–1187. https://doi.org/10.1016/j.anbehav.2006.09.029 Crossref Google Scholar 43. Kadmon, R., & Allouche, O. (2007). Integrating the Effects of Area, Isolation, and Habitat Heterogeneity on Species Diversity: A Unification of Island Biogeography and Niche Theory. The American Naturalist, 170(3), 443–454. https://doi.org/10.1086/519853 Crossref Google Scholar 44. Kotler, B. P., Brown, J. S., & Bouskila, A. (2004). Apprehension and time allocation in gerbils: The effects of predatory risk and energetic state. Ecology, 85(4), 917–922. https://doi.org/10.1890/03-3002 Crossref Google Scholar 45. Kotler, B. P., Brown, J. S., Slotow, R. H., Goodfriend, W. L., & Strauss, M. (1993). Nordic Society Oikos The Influence of Snakes on the Foraging Behavior of Gerbils. Source: Oikos, 67(2), 309–316. Google Scholar 46. Laundré, J. W., Hernández, L., & Altendorf, K. B. (2001). Wolves, elk, and bison: reestablishing the “landscape of fear” in Yellowstone National Park, U.S.A. Canadian Journal of Zoology, 79(8), 1401–1409. https://doi.org/10.1139/cjz-79-8-1401 Crossref Google Scholar 47. Laundre, J. W., Hernandez, L., & Ripple, W. J. (2010). The Landscape of Fear: Ecological Implications of Being Afraid~!2009-09-09~!2009-11-16~!2010-02-02~! The Open Ecology Journal, 3(3), 1–7. https://doi.org/10.2174/1874213001003030001 Crossref Google Scholar 48. Liesenjohann, T., Liesenjohann, M., Trebaticka, L., Sundell, J., Haapakoski, M., Ylönen, H., & Eccard, J. A. (2015). State-dependent foraging: lactating voles adjust their foraging behavior according to the presence of a potential nest predator and season. Behavioral Ecology and Sociobiology, 69(5), 747–754. https://doi.org/10.1007/s00265-015-1889-x Crossref Google Scholar 49. Lima, S. L. (1987). Vigilance while feeding and its relation to the risk of predation. Journal of Theoretical Biology, 124(3), 303–316. https://doi.org/10.1016/S0022-5193(87)80118-2 Crossref Google Scholar 50. Lima, S. L. (1995a). Back to the basics of anti-predatory vigilance: the group-size effect. Animal Behaviour, 49(1), 11–20. https://doi.org/10.1016/0003-3472(95)80149-9 Crossref Google Scholar 51. Lima, S. L. (1995b). Collective detection of predatory attack by social foragers: fraught with ambiguity? Animal Behaviour, 50(4), 1097–1108. https://doi.org/10.1016/0003-3472(95)80109-X Crossref Google Scholar 52. Lima, S. L. (1998). Nonlethal effects in the ecology of predator-prey interactions. BioScience, 48(1), 25–34. https://doi.org/10.2307/1313225 Crossref Google Scholar 53. Magnhagen, C., & Borcherding, J. (2008). Risk-taking behaviour in foraging perch: does predation pressure influence age-specific boldness? Animal Behaviour, 75(2), 509–517. https://doi.org/10.1016/j.anbehav.2007.06.007 Crossref Google Scholar 54. McCune, K., Jablonski, P., Lee, S. I., & Ha, R. (2018). Evidence for personality conformity, not social niche specialization in social jays. Behavioral Ecology, 29(4), 910–917. https://doi.org/10.1093/beheco/ary055 Crossref Google Scholar 55. McNamara, J. M., & Houston, A. I. (1987). Starvation and Predation as Factors Limiting Population Size. Ecological Society of America, 68(5), 1515–1519. Google Scholar 56. Mellin, C., Parrott, L., Andréfouët, S., Bradshaw, C. J., MacNeil, M. A., & Caley, M. J. (2012). Multi-scale marine biodiversity patterns inferred efficiently from habitat image processing Multi-scale marine biodiversity patterns inferred efficiently from habitat image processing. Ecological Applications, 22:3(January 2016), 792–803. https://doi.org/10.2307/23213917 Crossref Google Scholar 57. Mellin, C., Parrott, L., Andréfouët, S., Bradshaw, C. J. A., Aaron, M., Caley, M. J., Applications, S. E., & April, N. (2015). Ecological Society of America Multi-scale marine biodiversity patterns inferred efficiently from habitat image processing inferred patterns biodiversity from habitat image processing efficiently. 22(3), 792–803. Google Scholar 58. Moll, R. J., Redilla, K. M., Mudumba, T., Muneza, A. B., Gray, S. M., Abade, L., Hayward, M. W., Millspaugh, J. J., & Montgomery, R. A. (2017). The many faces of fear: a synthesis of the methodological variation in characterizing predation risk. Journal of Animal Ecology, 86(4), 749–765. https://doi.org/10.1111/1365-2656.12680 Crossref Google Scholar 59. Nakagawa, S., & Schielzeth, H. (2013). A general and simple method for obtaining R2 from generalized linear mixed-effects models. Methods in Ecology and Evolution, 4(2), 133–142. https://doi.org/10.1111/j.2041-210x.2012.00261.x Crossref Google Scholar 60. Nelson, E. H., Matthews, C. E., & Rosenheim, J. A. (2004). Predators reduce prey population growth by inducing changes in prey behavior. Ecology, 85(7), 1853–1858. https://doi.org/10.1890/03-3109 Crossref Google Scholar 61. Barton, K. (2020). Multi-Model Inference. R Project.’ Google Scholar 62. Palmer, Meredith S. et al. (2022). Dynamic landscapes of fear: understanding spatiotemporal risk. Trends in Ecology & Evolution, 37(10), 911 - 925 Google Scholar 63. Parrott, L. (2010). Measuring ecological complexity. Ecological Indicators, 10(6), 1069–1076. https://doi.org/10.1016/j.ecolind.2010.03.014 Crossref Google Scholar 64. Pennington, R. T., Lavin, M., & Oliveira-Filho, A. (2009). Woody plant diversity, evolution, and ecology in the tropics: Perspectives from seasonally dry tropical forests. Annual Review of Ecology, Evolution, and Systematics, 40, 437–457. https://doi.org/10.1146/annurev.ecolsys.110308.120327 Crossref Google Scholar 65. Pettorelli, N., Coulson, T., Durant, S. M., & Gaillard, J. M. (2011). Predation, individual variability and vertebrate population dynamics. Oecologia, 167(2), 305–314. https://doi.org/10.1007/s00442-011-2069-y Crossref Google Scholar 66. Proulx, R., & Parrott, L. (2008). Measures of structural complexity in digital images for monitoring the ecological signature of an old-growth forest ecosystem. Ecological Indicators, 8(3), 270–284. https://doi.org/10.1016/j.ecolind.2007.02.005 Crossref Google Scholar 67. Proulx, R., & Parrott, L. (2009). Structural complexity in digital images as an ecological indicator for monitoring forest dynamics across scale, space and time. Ecological Indicators, 9(6), 1248–1256. https://doi.org/10.1016/j.ecolind.2009.03.015 Crossref Google Scholar 68. Pulliam, H. R. (1973). On the advantages of flocking. Journal of Theoretical Biology, 38(2), 419–422. https://doi.org/10.1016/0022-5193(73)90184-7 Crossref Google Scholar 69. Roberts, G. (1996). Why individual vigilance declines as group size increases. Animal Behaviour, 51(June 1995), 1077–1086. https://doi.org/10.1006/anbe.1996.0109 Crossref Google Scholar 70. Rodríguez, A., Andrén, H., & Jansson, G. (2001). Habitat-mediated predation risk and decision making of small birds at forest edges. Oikos, 95(3), 383–396. https://doi.org/10.1034/j.1600-0706.2001.950303.x Crossref Google Scholar 71. Schmidt, K., & Kuijper, D. P. J. (2015). A “death trap” in the landscape of fear. Mammal Research, 60(4), 275–284. https://doi.org/10.1007/s13364-015-0229-x Crossref Google Scholar 72. Sih, A. (1980). Optimal behavior: can foragers balance two conflicting demands? Science (New York, N.Y.), 210(4473), 1041–1043. https://doi.org/10.1126/science.210.4473.1041 Crossref Google Scholar 73. Sih, A., & McCarthy, T. M. (2002). Prey responses to pulses of risk and safety: testing the risk allocation hypothesis. Animal Behaviour, 63, 437–443. https://doi.org/10.1006/anbe.2001.1921 Crossref Google Scholar 74. Sih, A. (1992). Prey uncertainty and the balancing of antipredator and feeding needs. American Naturalist, 139(5), 1052–1069. https://doi.org/10.1086/285372 Crossref Google Scholar 75. Smith, J. A., Donadio, E., Pauli, J. N., Sheriff, M. J., Bidder, O. R., & Middleton, A. D. (2019). Habitat complexity mediates the predator–prey space race. Ecology, 100(7), 1–9. https://doi.org/10.1002/ecy.2724 Crossref Google Scholar 76. Stankowich, T., & Coss, R. G. (2007). Effects of risk assessment, predator behavior, and habitat on escape behavior in Columbian black-tailed deer. Behavioral Ecology, 18(2), 358–367. https://doi.org/10.1093/beheco/arl086 Crossref Google Scholar 77. Steinhoff, P. O. M., Warfen, B., Voigt, S., Uhl, G., & Dammhahn, M. (2020). Individual differences in risk-taking affect foraging across different landscapes of fear. Oikos, 129(12), 1891–1902. https://doi.org/10.1111/oik.07508 Crossref Google Scholar 78. Suarez-Rubio, M., Ille, C., & Bruckner, A. (2018). Insectivorous bats respond to vegetation complexity in urban green spaces. Ecology and Evolution, 8(6), 3240–3253. https://doi.org/10.1002/ece3.3897 Crossref Google Scholar 79. Tanner, J. E., Mellin, C., Parrott, L., & Bradshaw, C. J. A. (2015). Fine-scale benthic biodiversity patterns inferred from image processing. Ecological Complexity, 22, 76–85. https://doi.org/10.1016/j.ecocom.2015.02.009 Crossref Google Scholar 80. Tews, J., Brose, U., Grimm, V., Tielbörger, K., Wichmann, M. C., Schwager, M., & Jeltsch, F. (2004). Animal species diversity driven by habitat heterogeneity/diversity: The importance of keystone structures. Journal of Biogeography, 31(1), 79–92. https://doi.org/10.1046/j.0305-0270.2003.00994.x Crossref Google Scholar 81. Thaker, M., Vanak, A. T., Owen, C. R., Ogden, M. B., & Slotow, R. (2010). Group dynamics of zebra and wildebeest in a woodland savanna: Effects of predation risk and habitat density. PLoS ONE, 5(9), 1–6. https://doi.org/10.1371/journal.pone.0012758 Crossref Google Scholar 82. Treisman, M. (1975). Predation and the evolution of gregariousness. I. Models for concealment and evasion. Animal Behaviour, 23(PART 4), 779–800. https://doi.org/10.1016/0003-3472(75)90106-2 Crossref Google Scholar 83. Verdolin, J. L. (2006). Meta-analysis of foraging and predation risk trade-offs in terrestrial systems. 457–464. https://doi.org/10.1007/s00265-006-0172-6 Crossref Google Scholar 84. von Post, M., Borgstrom, P., Smith, H. G., & Olsson, O. (2012). Assessing habitat quality of farm-dwelling house sparrows in different agricultural landscapes. Oecologia, 168(4), 959–966. https://doi.org/10.1007/s00442-011-2169-8 Crossref Google Scholar 85. Werner, E. E., Gilliam, J. F., Hall, D. J., & Mittelbach, G. G. (2015). An Experimental Test of the Effects of Predation Risk on Habitat Use in Fish Published by : Ecological Society of America AN EXPERIMENTAL TEST OF THE EFFECTS OF PREDATION RISK ON HABITAT USE IN FISH ’. 64(6), 1540–1548. Google Scholar 86. Wheatley, R., Pavlic, T. P., Levy, O., & Wilson, R. S. (2020). Habitat features and performance interact to determine the outcomes of terrestrial predator–prey pursuits. Journal of Animal Ecology, 89(12), 2958–2971. https://doi.org/10.1111/1365-2656.13353 Crossref Google Scholar 87. Wheeler, H. C., & Hik, D. S. (2014). Giving-up densities and foraging behaviour indicate possible effects of shrub encroachment on arctic ground squirrels. Animal Behaviour, 95, 1–8. https://doi.org/10.1016/j.anbehav.2014.06.005 Crossref Google Scholar 88. Yang, Z., Liu, X., Zhou, M., Ai, D., Wang, G., Wang, Y., Chu, C., & Lundholm, J. T. (2015). The effect of environmental heterogeneity on species richness depends on community position along the environmental gradient. Scientific Reports, 5(October), 1–7. https://doi.org/10.1038/srep15723 Crossref Google Scholar Information & Authors Information Version history V1 Version 1 24 April 2025 Copyright This work is licensed under a Non Exclusive No Reuse License. Collection Oikos Keywords antipredator behaviour cyanocorax giving-up density predation risk vigilance Authors Affiliations Maria Carolina Venáncio Universidade Federal do Ceará View all articles by this author Luiz Mestre Universidade Federal do Paraná View all articles by this author Lorenzo Zanette 0000-0001-6497-3555 [email protected] Universidade Federal do Ceará View all articles by this author Metrics & Citations Metrics Article Usage 253 views 170 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Maria Carolina Venáncio, Luiz Mestre, Lorenzo Zanette. MULTIPLE SCALES OF FEAR: FORAGING BEHAVIOR OF WHITE-NAPED JAYS IN SEMIARID LANDSCAPES. Authorea . 24 April 2025. DOI: https://doi.org/10.22541/au.174549801.12515474/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.174549801.12515474/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9ffc539bbac9300f',t:'MTc3OTQ1ODIxMQ=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();