Levels Of Predation Risk Research Articles

The evolutionary origins and ecological context of tool use in New Caledonian crows

New Caledonian (NC) crows Corvus moneduloides are the most prolific avian tool users. In the wild, they use at least three distinct tool types to extract invertebrate prey from deadwood and vegetation, with some of their tools requiring complex manufacture, modification and/or deployment. Experiments with captive-bred, hand-raised NC crows have demonstrated that the species has a strong genetic predisposition for basic tool use and manufacture, suggesting that this behaviour is an evolved adaptation. This view is supported by recent stable-isotope analyses of the diets of wild crows, which revealed that tool use provides access to highly profitable hidden prey, with preliminary data indicating that parents preferentially feed their offspring with tool-derived food. Building on this work, our review examines the possible evolutionary origins of these birds' remarkable tool-use behaviour. Whilst robust comparative analyses are impossible, given the phylogenetic rarity of animal tool use, our examination of a wide range of circumstantial evidence enables a first attempt at reconstructing a plausible evolutionary scenario. We suggest that a common ancestor of NC crows, originating from a (probably) non-tool-using South-East Asian or Australasian crow population, colonised New Caledonia after its last emersion several million years ago. The presence of profitable but out-of-reach food, in combination with a lack of direct competition for these resources, resulted in a vacant woodpecker-like niche. Crows may have possessed certain behavioural and/or morphological features upon their arrival that predisposed them to express tool-use rather than specialised prey-excavation behaviour, although it is possible that woodpecker-like foraging preceded tool use. Low levels of predation risk may have further facilitated tool-use behaviour, by allowing greater expenditure of time and energy on object interaction and exploration, as well as the evolution of a 'slow' life-history, in which prolonged juvenile development enables acquisition of complex behaviours. Intriguingly, humans may well have influenced the evolution of at least some of the species' tool-oriented behaviours, via their possible introduction of candlenut trees together with the beetle larvae that infest them. Research on NC crows' tool-use behaviour in its full ecological context is still in its infancy, and we expect that, as more evidence accumulates, some of our assumptions and predictions will be proved wrong. However, it is clear from our analysis of existing work, and the development of some original ideas, that the unusual evolutionary trajectory of NC crows is probably the consequence of an intricate constellation of interplaying factors.

Levels of Predation Modulate Antipredator Defense Behavior and Metamorphic Traits in the Toad Bufo melanostictus

The influence of the presence of nonlethal predators on antipredator defense behavior and metamorphic traits was studied in the toad Bufo melanostictus. Tadpoles reared with caged predaceous tadpoles of Hoplobatrachus tigerinus delayed metamorphosis and also emerged at a smaller size compared to those in a predator-free environment. Different predaceous cues (kairomones, alarm cues and dietary cues, derived after feeding the predator with conspecific prey) induced increase in stationary phase and swimming speed and decrease in swimming spurts and time used for feeding. The intensity of defense behaviors varied with the cue. They were more intense in response to cues of predators fed on conspecific prey item. The response to alarm cues or kairomones or predators fed on heterogeneric tadpole species was similar but lower in magnitude compared to that elicited in response to predators fed on conspecific prey item. Most intense defense behavior in response to the odors of the predator that consumed conspecific members is possibly caused by a combined effect of kairomones, alarm, and dietary cues. The findings suggest that B. melanostictus tadpoles are capable of assessing levels of predation risk and modulating the intensity of their defense behavior in accordance with the perceived threat. Interestingly, despite longer larval period, size at metamorphosis of the toadlets was smaller in the presence of caged predator.

Learning to Avoid Dangerous Habitat Types by Aquatic Salamanders,Eurycea tynerensis

AbstractThere should be intense selection for predation avoidance mechanisms when prey live in close proximity to their predators. Prey individuals that can learn to associate habitat features with high levels of predation risk should experience increased survival if they subsequently avoid those habitats. We tested whether or not habitat learning occurred in a benthic stream community consisting of adult Oklahoma salamander (Eurycea tynerensis) prey and a syntopic predatory fish, the banded sculpin (Cottus carolinae). We exposed individual salamanders to chemical stimuli from sculpin, non‐predatory tadpoles, or a blank control in training tanks containing either rocks or grass. Two days later, the salamanders were tested in tanks that offered a choice of rocks or grass. Salamanders showed significant avoidance of the habitat where they had previously encountered chemical cues from sculpin in comparison to the non‐predatory controls. Learning to avoid dangerous habitats may be particularly important for prey whose predators are visually cryptic ambush foragers, such as sculpin.

Sensory basis of vigilance behavior in birds: Synthesis and future prospects

Birds gather visual information through scanning behavior to make decisions relevant for survival (e.g., detecting predators and finding food). The goal of this study was (a) to review some visual properties involved in scanning behavior (retinal specialization for visual resolution and motion detection, visual acuity, and size of the blind area), and (b) hypothesize how the inter-specific variability in these properties may lead to different scanning strategies. The avian visual system has a high degree of heterogeneity in visual performance across the visual field, with some sectors providing higher levels of visual resolution and motion detection (e.g., retinal specializations) than others (e.g., peripheral retina and blind area). Thus, information quality will vary in different parts of the visual field, which contradicts some theoretical assumptions on information gathering. Birds need to move their eyes and heads to align the retinal specializations to different sectors of visual space. The rates of eye and head movements can then be used as proxies for scanning strategies. I propose specific predictions as to how each of the visual properties studied can affect scanning strategies in the context of predator detection in different habitat types and with different levels of predation risk. Establishing the degree of association between sensory specializations and scanning strategies can enhance our understanding of the evolution of anti-predator behavior.

Does fine sediment constrain salmonid alevin development and survival?

We applied a randomized factorial design with three sedimentation treatments (high, low, and zero sediment addition) and two predation-risk levels (predator present or absent) to assess whether the survival and emergence of brown trout (Salmo trutta) alevins were affected by sedimentation and (or) chemical cues from a predatory fish (burbot, Lota lota). In this laboratory experiment, survival was only related to the finest organic sediment fraction (<0.074 mm), which decreased embryo survival, although only constituting less than 1.5% of all sediments. Control fish tended to postpone emergence when exposed to predator odour, whereas fish in the high-sedimentation treatment showed no response to predators. Alevins that received high sedimentation had larger yolk sacs at emergence compared with control fish. Sedimentation may thus have serious fitness consequences on salmonids, as fry with larger yolk sacs are poor swimmers and therefore more vulnerable to predation or downstream displacement. Fine-sediment deposition from the catchment or stream banks may cause a serious threat to salmonid populations, and more effective erosion control is needed to reduce inputs of fine sediments to river habitats.

Indirect measures of reproductive effort in a resource-defense polygynous ungulate: territorial defense by male guanacos

Trade-offs between reproductive effort and activities that are likely to enhance survival have been studied extensively in harem defense ungulates. However, among resource defense systems, ecological determinants of the variation in male reproductive effort are less understood. The aim of this study was to assess the extent of the effort devoted to territorial defense by male guanacos and how this is affected by predation risk. Time allocation to different activities and displays of aggressive behavior were compared between territorial and bachelor male guanacos within two populations with contrasting levels of predation risk. Territorial males devoted almost twice the time than bachelors to vigilance and moving, at the expense of foraging time. Aggressive interactions were more frequent and severe for territorial males than for bachelors, based on observation. These differences were larger during the reproductive season but remained statistically significant after this period, highlighting the extended effort towards territorial defense beyond the peak of reproductive activity. In contrast to previous observations on females of the same populations, there was no effect of predation risk level or group size on male time allocation, suggesting that under the ecological conditions of the study, intra-sexual competition entails a stronger influence on male individual behavior than predation risk level.

Breeding of the Australasian Bittern (Botaurus poiciloptilus) in New Zealand

In Australasia, wetlands and the birds that inhabit them are under considerable threat. Australasian Bitterns (Botaurus poiciloptilus) are a potential indicator of wetland health but there is an urgent need to collect baseline ecological data and understand processes threatening the species so that conservation actions aimed at reversing apparent declines can be implemented. I collated nest records from New Zealand to determine breeding parameters and highlight areas requiring further research. Nests were recorded throughout the country, and all except one were in wetlands. Booming calls were recorded in most months but peaked between September and November. Breeding was observed from August to May, with most activity in November and December. Nests were in a wide variety of vegetation types, frequently Typha orientalis, and sites were usually surrounded by deep water. Breeding parameters appear similar to those of northern hemisphere species of Botaurus. Australasian Bitterns appear to have a high potential breeding capacity, with large clutches, fairly short incubation periods, and a long breeding season but there are no quantifiable data on nesting success. There are major gaps in our knowledge, and research on breeding should focus on assessing the fate of breeding attempts, potential influence of habitat loss and fragmentation on breeding success, vulnerability to predation by introduced mammals, the interaction between predation risk and water levels, and the roles of food and habitat in sustaining populations.

Predation risk shapes social networks in fission-fusion populations.

Predation risk is often associated with group formation in prey, but recent advances in methods for analysing the social structure of animal societies make it possible to quantify the effects of risk on the complex dynamics of spatial and temporal organisation. In this paper we use social network analysis to investigate the impact of variation in predation risk on the social structure of guppy shoals and the frequency and duration of shoal splitting (fission) and merging (fusion) events. Our analyses revealed that variation in the level of predation risk was associated with divergent social dynamics, with fish in high-risk populations displaying a greater number of associations with overall greater strength and connectedness than those from low-risk sites. Temporal patterns of organisation also differed according to predation risk, with fission events more likely to occur over two short time periods (5 minutes and 20 minutes) in low-predation fish and over longer time scales (>1.5 hours) in high-predation fish. Our findings suggest that predation risk influences the fine-scale social structure of prey populations and that the temporal aspects of organisation play a key role in defining social systems.

Moonlight and shelter cause differential seed selection and removal by rodents

Various environmental factors may influence the foraging behaviour of seed dispersers which could ultimately affect the seed dispersal process. We examined whether moonlight levels and the presence or absence of rodent shelter affect rodent seed removal (rate, handling time and time of removal) and seed selection (size and species) among seven oak species. The presence or absence of safe microhabitats was found to be more important than moonlight levels in the removal of seeds. Bright moonlight caused a different temporal distribution of seed removal throughout the night but only affected the overall removal rates in open microhabitats. Seeds were removed more rapidly in open microhabitat (regardless of the moon phase), decreasing the time allocated to seed discrimination and translocation. Only in open microhabitats did increasing levels of moonlight decrease the time allocated to selection and removal of seeds. As a result, a more precise seed selection was made under shelter, owing to lower levels of predation risk. Rodent ranking preference for species was identical between full/new moon in shelter but not in open microhabitats. For all treatments, species selection by rodents was much stronger than size selection. Nevertheless, heavy seeds, which require more energy and time to be transported, were preferentially removed under shelter, where there is no time restriction to move the seeds. Our findings reveal that seed selection is safety dependent and, therefore, microhabitats in which seeds are located (sheltered versus exposed) and moonlight levels in open areas should be taken into account in rodent food selection studies.

Response across a gradient: behavioural reactions of newly settled fish to predation cues

How individuals respond to predation threats will have a large influence on which individuals survive within a population. The magnitude and timing of these responses will be particularly important during periods of high predation susceptibility, such as that experienced by coral reef fishes immediately following settlement to the reef environment. Although reef fish are known to adopt certain antipredator behaviours when exposed to chemical alarm scents during this early period, the role of visual cues, and how the response varies with different levels of predation threat (both olfactory and visual), remains unknown. This study examined the behavioural response of a newly settled coral reef fish (Pomacentrus amboinensis) to different levels of an olfactory and visual predation threat. The concentration of a conspecific chemical alarm cue and the visual proximity to a potential predator (Pseudochromis fuscus) were manipulated in separate aquarium experiments. Behavioural responses were found to be threat sensitive in nature, with higher-level threat cues eliciting a more intense response. Although significant changes were observed, responses to visual cues were more inconsistent, while responses to extremely low chemical cue concentrations were marginal, indicating a possible threshold lower limit. These findings demonstrate the ability of newly settled fish to assess the level of predation risk using both visual and chemical cues, and respond appropriately.

The degree of response to increased predation risk corresponds to male secondary sexual traits

Secondary sexual traits are expected to reflect a balance between sexual selection and natural selection. We test the hypothesis that plasticity in sexual advertisement behaviors can influence this trade-off, allowing showier traits than expected for a given level of predation risk. Specifically, we tested whether the degree of behavioral plasticity exhibited in response to chemical cues of a co-occurring predatory wolf spider corresponds to courtship rate and the degree of ornamentation in male wolf spiders. Both ornamented (brush-legged) males and non-ornamented males decreased locomotion, decreased their likelihood to court, and increased their time to initiate courtship in response to predator cues. However, brush-legged males increased their time to initiate courtship more than did non-ornamented males, demonstrating a greater response to the risk of predation for the more ornamented males. Similarly, within brush-legged males, individuals with the highest courtship rates also showed the greatest degree of plasticity in time to initiate courtship across predation contexts, whereas behavioral plasticity was independent of courtship rate for non-ornamented males. We found no correlation between ornament size and plasticity in response to predator cues within brush-legged males. Ultimately, we suggest that our data provide support for the hypothesis that behavioral plasticity in response to predator cues may alter the trade-off between predation risk and sexual advertisement and may be more important for males with higher degrees of conspicuousness in ornamentation and courtship.

Behavioral responses to variable predation risk in the California spiny lobster Panulirus interruptus

Shelter dwelling and gregariousness are behavioral strategies used by benthic marine organisms to reduce the risk of predator-induced mortality. For spiny lobsters, the relative and syner- gistic success of shelter dwelling and aggregation in reducing mortality may vary with the perceived risk of predation as well as the availability and size distribution of conspecifics and shelters. We tested how these factors influence antipredator behavior in the California spiny lobster Panulirus interruptus in southern California kelp forests. We measured shelter and lobster characteristics, predator density, relative lobster mortality outside of shelter, and quantified lobster antipredator behaviors in (1) the La Jolla Ecological Reserve (LJER), a marine protected area with high densities of large fishes known to be lobster predators, and (2) the Point Loma kelp forest, a nearby, heavily fished area. Predator densities and the relative mortality of tethered juvenile and adult lobsters were much higher in the LJER than in Point Loma, as was mean lobster size. In response to higher preda- tion risk, solitary lobsters in the LJER used shelters more closely scaled to their body size than lob- sters in Point Loma, thereby restricting predator access to shelters. In addition, aggregations of 2 or more lobsters were more frequent than expected by chance in the LJER, but not in Point Loma. Lobsters in Point Loma commonly were found outside of shelter during the day, whereas no lobsters ever were observed outside of shelter during daylight hours in the LJER. Overall, our results indicate that California spiny lobsters modify their antipredatory behaviors with level of predation risk and that behaviors designed to minimize predator-induced mortality are similar between California spiny lobsters and other spiny lobster species.

Comparing habitat quality within and between environments using giving up densities: an example based on the winter habitat of white-tailed deer Odocoileus virginianus

We previously developed a model, based on the precepts of optimal patch use, to compare habitat quality both within and between environments. Here we illustrate the use of this model in a study estimating quality of winter habitats (deer yards) of white-tailed deer Odocoileus virginianus near the northern limit of their range by following their foraging behaviour. We compare giving up densities (GUDs), the amount of food remaining in a patch when a forager ceases foraging there, with and without the presence of supplemental food in order to draw inferences about the relative quality either of habitats within an environment or of distinct environments. We use our model to evaluate the impact of alterations to the winter habitat of deer at two distinct sites that differed in their level of predation risk and food availability. The first site, the Mont Rigaud deer yard, was surrounded by farm land and gradually-expanding suburbs. Predators were rare and food was available in winter either in farm fields or around private homes but deer browsing in the past had left little food in the forest. At the second site, the Calumet deer yard, deer experienced a higher predation risk and did not have access to supplemental food from farm fields or private homes. However, past browsing by deer had not drastically reduced food in the forest. We offered food to deer in four habitats per site (forest, forest edge, clearing, clearing edge) with four to six replicates per site and measured the GUDs after 24 h. Analysis of these data, interpreted according to our model, suggests that deer are more sensitive to metabolic costs at Mont Rigaud and food availability at Calumet; predation risk does not alter deer behaviour between the two sites. Within habitats, deer at Mont Rigaud reacted to clearings as though they imposed higher metabolic costs than the forest. They also reacted to an interaction in which missed food costs influenced GUD only when metabolic costs were not too high. Thus our model appears to provide a convenient tool for comparing habitat quality both within and between environments.

Behavioural adaptations of mussels to varying levels of food availability and predation risk

Blue mussels Mytilus edulis (n ¼ 14) were studied in the laboratory using Hall sensor systems to record their gaping behaviour when exposed to varying food rations and levels of predation risk. Mussel response to increasing daily algal ration was to increase mean gape angle per day and was associated with copious pseudofaeces production at excessive initial algal concentrations, e.g. 250 cells/ml. Mean gape angle decreased (backward S-shaped curve) when fed a fixed algal ration per day where simulated predation risk (introduced fresh mussel homogenate) increased in the water, presumably as an anti-predation strategy. However, this behaviour is presumed to lower feeding rates. There was a general positive relationship between both gape angle and the extent of valve movement (abduction/ adduction) per event and valve movement speed. However, the fastest valve adduction events, which resulted in valve closure, were recorded independent of gape angle and only when mussels were first exposed to high perceived predation risk. We interpret this as an appropriate response, with high energetic cost, as a first line of defence from predators such as starfish and crabs. Overall, mussel response to predation appears graded and complex, indicating a trade-off between maximizing feeding/pseudofaeces production and minimizing predation risk.

Non-lethal effects of predators on body growth and health state of juvenile lizards, Psammdromus algirus

Predation risk does not necessarily increase predation rates because prey may be able to behave differentially to cope with higher predation risk. However, antipredatory behaviors may be costly, leading to negative, although non-lethal, effects of predators on prey. We examined in outdoor enclosures whether an experimental increase in predation pressure, which did not increase direct mortality, but forced individuals to increase antipredatory behaviors, may have significant non-lethal effects on body growth and health state of juvenile lizards, Psammodromus algirus. Simulated persistent predator attacks resulted in slower rates of body size growth and body mass gain of juvenile lizards, which may greatly affect their future survival. However, juvenile lizards were able to maintain their initial body condition and immune response regardless of predation risk level. Moreover, our data suggested that experience of lizards with their home range “environment” might allow them to compensate the negative effects of temporal high predation risk on body condition when predation risk subsequently decreased. Finally, juvenile lizards with greater immune responses showed smaller increments in body size, but larger increments in body mass and body condition, at the end of the control treatment. In contrast, there was no relationship between immunity and growth after the experimental treatment.

Spatio-temporal overlap between Yellowstone bison and elk - implications of wolf restoration and other factors for brucellosis transmission risk

1. In the Greater Yellowstone Ecosystem, bison Bison bison and elk Cervus canadensis nelsoni act as hosts for Brucella abortus. The presence of B. abortus within wildlife populations is an important conservation issue because of the risk of brucellosis transmission from wildlife to cattle. 2. We investigated conditions facilitating contact between bison (40–60% seroprevalence) and elk on a shared winter range in the Madison headwaters area of Yellowstone National Park. We evaluated the effects of snow pack, season, elk and bison population sizes, and wolf Canis lupus predation risk on the degree of spatial overlap between bison and elk from 1991 to 2006. 3. Sixty-eight per cent of 10 093 independent elk observations occurred within the known bison wintering range, 29% occurred within the distribution of bison within the winter range at the time of sampling and 14% occurred within 100 m of bison. Spatial overlap between bison and elk measured across these three spatial scales increased with week of the season, snow pack, and on days when wolves were within the same drainage area as elk, but decreased with cumulative levels of wolf predation risk. 4. Wolves contributed to immediate, short-term responses by elk that increased spatial overlap with bison, but longer-term responses to wolves resulted in elk distributions that reduced spatial overlap with bison. Spatial overlap increased through the winter and peaked when late-term abortion events and parturition occurred for bison. 5. Synthesis and applications. Despite this high level of association, elk exposure to B. abortus in the Madison headwaters (2–4%) was similar to those in free-ranging elk populations that do not intermingle with bison (1–3%), suggesting that B. abortus transmission from bison-to-elk under natural conditions is rare. Our results suggest that risk-driven elk behavioural responses to wolves are unlikely to have important disease implications. Management of brucellosis in greater yellow stone ecosystem elk populations should focus on reducing elk-to-elk transmission risk and, wherever possible, curtailing practices that increase elk density and group sizes during the potential abortion period.

To be or not to be conspicuous: the effects of prey availability and predator risk on spider's web decoration building

Some spiders build conspicuous silk decorations, or stabilimenta, on their orb webs. The decorations are thought to function as visual signals, and there is evidence to support both prey attraction and predator defence hypotheses. This study was designed to test both hypotheses simultaneously in Cyclosa argenteoalba by examining whether spiders change web decorations according to perceived levels of prey availability and predation risk. Spiders built their webs with more thread when prey was experimentally supplied, but with less thread when exposed to the vibrating tuning fork that mimicked the wing beat of flying insect predators. The results indicate that spiders invest more in foraging activity in the presence of available prey, but invest less in response to perceived high levels of predation risk. Web decoration, on the other hand, was only affected by the perceived level of predation risk and not prey availability. Spiders built longer decorations when exposed to the tuning fork vibrations. The results indicate that silk decoration has a predator defence function, but not a prey attraction function in C. argenteoalba . They also suggest that orb web decoration by C. argenteoalba does not involve a trade-off between deterring predators and being avoided by prey.

Testing the effects of resource distribution and inherent habitat riskiness on simultaneous habitat selection by predators and prey

Theoretical models have extended the Ideal Free Distribution model to examine predator–prey systems having three trophic levels, when both predator and prey are allowed to move freely. One consistent prediction made by such models is that the spatial distribution of prey should be mainly determined by the inherent habitat riskiness (e.g. cover level), with prey avoiding the inherently riskier habitats regardless (or nearly so) of resource distribution. To test this prediction, we conducted laboratory experiments in which both predators (creek chub, Semotilus atromaculatus) and prey (northern redbelly dace, Phoxinus eos) were free to move between feeding patches differing in resource quantity and habitat riskiness. When alone, creek chub and northern redbelly dace both preferred the more food-rich patch. The spatial distribution of dace was also inversely related to that of creek chub in the presence of chub fenced into feeding patches. When the patches differed in both riskiness and resource quantity, the spatial distribution of dace was mainly influenced by resource distribution and, to a lesser extent, by the inherent habitat riskiness in the absence of predators. In the presence of creek chub, the dace significantly decreased their use of the inherently riskier patch compared to when predators were absent. However, contrary to the models' prediction, food distribution still significantly influenced dace distribution when predators were present. Finally, dace aggregated increasingly in a third, totally safe area (but one offering no food) as the number of chub present in both patches increased. Thus, the influence of resource distribution and inherent habitat riskiness on prey distribution seems to vary with the level of predation risk.

Alpine marmots (Marmota marmota) adjust vigilance behaviour according to environmental characteristics of their surrounding

Predation risk can strongly affect behaviour of preys, and natural selection should have favoured the evolution of behavioural plasticity in the strategies used to avoid predation. The level of predation risk can be affected by environmental characteristics and we expect individuals to adjust vigilance behaviour to the level of risk. In this study we investigate if Alpine marmots (Marmota marmota) living in an area delimited by the forest and with scarce visibility (Closed Site) differ in their vigilance behaviour from marmots inhabiting an open meadow further away from the forest edge and with greater visibility (Open Site). The rate of vigilance and the time spent vigilant in the Closed Site were higher than in the Open Site, while the mean duration of each vigilance bout did not differ between the two sites. Our results suggest that Alpine marmots adjust vigilance behaviour according to the environmental characteristics of their surroundings while there appears to exist an optimal duration of scan bouts which is independent of relative predation risk.

Optimal escape theory predicts escape behaviors beyond flight initiation distance: risk assessment and escape by striped plateau lizards Sceloporus virgatus

Abstract Escape theory predicts that flight initiation distance (FID = distance between predator and prey when escape begins) is longer when risk is greater and shorter when escape is more costly. A few tests suggest that escape theory applies to distance fled. Escape models have not addressed stochastic variables, such as probability of fleeing and of entering refuge, but their economic logic might be applicable. Experiments on several risk factors in the lizard Sceloporus virgatus confirmed all predictions for the above escape variables. FID was greater when approach was faster and more direct, for lizards on ground than on trees, for lizards rarely exposed to humans, for the second of two approaches, and when the predator turned toward lizards rather than away. Lizards fled further during rapid and second consecutive approaches. They were more likely to flee when approached directly, when a predator turned toward them, and during second approaches. They were more likely to enter refuge when approached rapidly. A novel finding is that perch height in trees was unrelated to FID because lizards escaped by moving out of sight, then moving up or down unpredictably. These findings add to a growing body of evidence supporting predictions of escape theory for FID and distance fled. They show that two probabilistic aspects of escape are predictable based on relative predation risk levels. Because individuals differ in boldness, the assessed optimal FID and threshold risks for fleeing and entering refuge are exceeded for an increasing proportion of individuals as risk increases.