Human activity and habitat type affect perceived predation risk in urban white‐footed mice (Peromyscus leucopus)

Predation risk is one of the largest costs associated with foraging in small mammals. Small mammals select microhabitat features such as tree and shrub canopy cover, woody debris and vegetative ground cover that can lower the risk of detection from predators and provide greater protection if discovered. Small mammals also increase foraging activity and decrease selection for cover when cloud cover increases and moon illumination is less. Often researchers assume small mammals in urban areas respond to these cues in the same manner as in natural areas, but these cues themselves are altered in urban zones. In this study, we investigated how Amur honeysuckle (Lonicera maackii) and coarse woody debris (CWD) affected giving‐up density (GUD) in white‐footed mice (Peromyscus leucopus). Each of three habitat treatments (open flood channel, the edge and interior of the honeysuckle patch) contained cover treatments with coarse woody debris present or absent. The six treatment combinations were compared to environmental variables (temperature, humidity and illumination) and habitat variables to test their effect on GUD. Peromyscus leucopus foraged to lower densities in areas with CWD present and also under the honeysuckle canopy, using this invasive shrub to decrease predation risk, potentially increasing survivability within this urban park. Increased human presence negatively affected foraging behavior across treatments. Human presence and light pollution significantly influenced P. leucopus, modifying their foraging behavior and demonstrating that both fine‐ and coarse‐scale urban factors can affect small mammals. Foraging increased as humidity increased, particularly under the honeysuckle canopy. Changes in illumination due to moonlight and cloud cover did not affect foraging behavior, suggesting urban light pollution may have altered behavioral responses to changes in light levels. Lonicera maackii seemed to facilitate foraging in P. leucopus, even though it adversely affects the plant community, suggesting that its impact may not be entirely negative.     

When perception reflects reality: Non‐native grass invasion alters small mammal risk landscapes and survival

Modification of habitat structure due to invasive plants can alter the risk landscape for wildlife by, for example, changing the quality or availability of refuge habitat. Whether perceived risk corresponds with actual fitness outcomes, however, remains an important open question. We simultaneously measured how habitat changes due to a common invasive grass (cheatgrass, Bromus tectorum) affected the perceived risk, habitat selection, and apparent survival of a small mammal, enabling us to assess how well perceived risk influenced important behaviors and reflected actual risk. We measured perceived risk by nocturnal rodents using a giving‐up density foraging experiment with paired shrub (safe) and open (risky) foraging trays in cheatgrass and native habitats. We also evaluated microhabitat selection across a cheatgrass gradient as an additional assay of perceived risk and behavioral responses for deer mice (Peromyscus maniculatus) at two spatial scales of habitat availability. Finally, we used mark‐recapture analysis to quantify deer mouse apparent survival across a cheatgrass gradient while accounting for detection probability and other habitat features. In the foraging experiment, shrubs were more important as protective cover in cheatgrass‐dominated habitats, suggesting that cheatgrass increased perceived predation risk. Additionally, deer mice avoided cheatgrass and selected shrubs, and marginally avoided native grass, at two spatial scales. Deer mouse apparent survival varied with a cheatgrass–shrub interaction, corresponding with our foraging experiment results, and providing a rare example of a native plant mediating the effects of an invasive plant on wildlife. By synthesizing the results of three individual lines of evidence (foraging behavior, habitat selection, and apparent survival), we provide a rare example of linkage between behavioral responses of animals indicative of perceived predation risk and actual fitness outcomes. Moreover, our results suggest that exotic grass invasions can influence wildlife populations by altering risk landscapes and survival.     

Seed predation by birds and small mammals in semiarid Chile

We studied spatial and temporal patterns in foraging activity among diurnal birds and nocturnal mammals at a semiarid site in northern Chile using artificial foraging trays. Small mammals foraged more extensively under shrubs than in open microhabitats, but birds showed no such selection. Moreover, avian foraging was more extensive than that by small mammals in all seasons and both microhabitats. Avian foraging was highly seasonal, as many birds at our site migrate to the Andean prepuna or to Patagonia during the austral summer. Birds have tended to be overshadowed by small mammals and ants in studies of granivory, but this study suggests that their importance may be underestimated in some systems.     

Seed predation and dispersal in relict Scots pine forests in southern Spain

For two years, the seed rain and magnitude of seed losses due to predation were evaluated in Scots pine forests in southern Spain. The Crossbill was the most important pre-dispersal predator, consuming more than 80% of ripening seeds. In addition, other birds, mainly Tits and Siskin, also consumed seeds just before seed dispersal, reaching values of 16 and 51% losses in 1996 and 1997, respectively. Seed rain was monitored in different microhabitats (under pine canopies, under shrubs and in open areas), and was most intense under the canopy of mother plants both years. Post-dispersal seed predators (rodents and birds) consumed up to 96% of seeds reaching the ground. Both pre- and post-dispersal seed predators preferentially harvested filled seeds. Post-dispersal predation was similarly intense in all microhabitats, so predators did not change the spatial distribution of the seed rain. These high predation rates were constant between years, localities and habitats (woodland and treeline). We hypothesize that this high rate of seed predation is a major factor limiting the regeneration of these relict populations of Scots pine in its southernmost limit.     

Predator Detection is Limited in Microhabitats with High Light Intensity: An Experiment with Brown‐Headed Cowbirds

Variations in ambient light conditions across different microhabitats can modify the detectability of predators and prey. Prey have been shown to be more visible in sunlit than in shaded patches, leading to higher predation risk and more investment in vigilance (predation risk hypothesis). Additionally, prey have been hypothesized to take longer to detect predators in sunlit compared to shaded patches because of the excess of sunlight causing glare effects (disability glare hypothesis). We tested the predictions of these two non‐mutually exclusive hypotheses in a seminatural experiment with brown‐headed cowbirds by measuring vigilance behavior and detection of a ground predator in patches under the shade of vegetation and in the open. Light intensity and achromatic contrast were higher in the sunlit patches, which could enhance glare effects, but chromatic contrast was higher in the shaded patches. Brown‐headed cowbirds took longer to show alert reactions to and flee from a ground predator in sunlit compared to shaded patches. However, the two parameters associated with perceived predation risk (vigilance prior to the predator exposure and time to resume foraging after the attack) did not differ between sunlit and shaded patches. Our findings support to a greater extent the disability glare hypothesis than the predation risk hypothesis. Overall, ambient light conditions can affect two critical components of behavioral predator–prey interactions in terrestrial habitats: detection of and escape from predators. The effects of disability glare are expected to be more pronounced in bird species with wider visual fields or without sun‐shading structures; however, species may compensate through various behaviors (e.g. avoidance of sunlit patches and changes in head orientation).     

Predator occurrence and perceived predation risk determine grouping behavior in guanaco (Lama guanicoe)

Grouping behavior of social ungulates may depend on both predator occurrence and perceived predation risk associated with habitat structure, reproductive state, and density of conspecifics. Over 3 years, we studied grouping behavior of guanaco (Lama guanicoe) families in Chilean Patagonia during the birthing season and determined their response to variation in predator occurrence and perceived predation risk (habitat structure, calf/adult rate, and density of conspecifics). We considered the effect of two predators, puma (Puma concolor) and culpeo fox (Lycalopex culpaeus). We measured two common (family group size and vigilance) and one novel (family group cohesion) behavioral responses of guanaco. Our results show that guanaco family groups adapted their grouping behavior to both predator occurrence and perceived predation risk. Larger family groups were found in open habitats and areas with high puma occurrence, while guanacos stayed in small family groups in areas with high shrub cover or low visibility. Group cohesion increased in areas with higher occurrence of pumas and culpeo foxes, and also increased in smaller family groups and in areas with low guanaco density. Vigilance (number of vigilant adults) was mainly related to group size and visibility, increasing in areas with low visibility, while residual vigilance (vigilance after removing the group‐size effect) did not vary with the explanatory variables examined. Our results suggest that a mix of predator occurrence and perceived predation risk influences guanaco grouping behavior and highlights the importance of evaluating different antipredator responses together and considering all predator species in studies aimed at understanding ungulate behavior.     

Use of Shrubs as Nurse Plants: A New Technique for Reforestation in Mediterranean Mountains

Common techniques currently used for afforestation in the Mediterranean basin consider the pre‐existing vegetation (mainly shrubs) as a source of competition for trees, and consequently it is generally eliminated before planting. Nevertheless, it has been demonstrated that woody plants can facilitate the establishment of understory seedlings in environments that, like the Mediterranean area, are characterized by a pronounced dry season. In this study, we experimentally analyze the usefulness of shrubs as nurse plants for afforestation of two native conifers, Pinus sylvestris L. (Scots pine) and Pinus nigra Arnold (black pine). Two‐year‐old seedlings were planted in four microhabitats: (1) open interspaces without vegetation (which is the usual method used in afforestation programs), (2) under individuals of Salvia lavandulifolia, (3) under the north side of spiny shrubs, and (4) under the south side of spiny shrubs. Pine survival was remarkably higher when planted under individuals of the shrub S. lavandulifolia (54.8% for Scots pine, 81.9% for black pine) compared with open areas (21.5% for Scots pine, 56.8% for black pine; chi square, p < 0.05). The survival of both pines was also higher when planted on the north side of spiny shrubs, although the survival on the south side was similar to that found in open areas. In addition, pine growth was not inhibited when planted in association with shrubs. This pattern appears to result from the combination of abiotic conditions imposed by the presence of a nurse shrub, which leads to improvement in seedling water status and therefore reduced summer mortality by drought. The results show that the use of shrubs as nurse plants is a technique that offers both economic and ecological advantages, in terms of savings in labor and plant material and reduced and even negligible impact on the pre‐existing vegetation.     

Associations between Non‐Lethal Injury,Body Size,and Foraging Ecology in an Amphibian Intraguild Predator

Theoretical treatments of intraguild predation and its effects on behavioral interactions regard the phenomenon as a size‐structured binary response wherein predation among competitors is completely successful or completely unsuccessful. However, intermediate outcomes occur when individuals escape intraguild (IG) interactions with non‐lethal injuries. While the effects of wounds for prey include compromised mobility and increased predation risk, the consequences of similar injuries among top predators are not well understood, despite the implications for species interactions. Using an amphibian IG predator, Ambystoma opacum (Caudata: Ambystomatidae), we examined associations between non‐lethal injuries and predator body size, foraging strategy, microhabitat selection, and intraspecific agonistic interactions. Wounds were common among IG predators, generally increasing in frequency throughout larval ontogeny. Non‐lethal injuries were associated with differences in predator body size and behavior, with injured predators exhibiting smaller body sizes, increased use of benthic microhabitats, reduced agonistic displays, and increased risk of intraspecific aggression. While such effects were not ultimately associated with reduced foraging success, non‐lethal injury could contribute to niche partitioning between injured and healthy predators via habitat selection, but injured predators likely continue to exert predatory pressure on IG and basal prey populations. Our results indicate that studies of top‐down population regulation should incorporate injury‐related modifications to both prey and predator behavior and size structure.     

Antipredatory Response and Food Intake in Wood Mice (Apodemus sylvaticus) under Simulated Predation Risk by Resident and Novel Carnivorous Predators

Chemical signals left by predators are a potential source of information about the risk of predation, and small mammals are known to take them into account when making decisions. We investigated whether wood mice (Apodemus sylvaticus) are more likely to avoid the faeces of resident predators (red fox Vulpes vulpes and common genet Genetta genetta) vs. a novel predator (European pine marten Martes martes). Odour recognition would increase perceived predation risk and reduce food intake by individual mice. Wood mice response to predators was analysed by live‐trapping using two untreated controls (baited/non‐baited) and traps experimentally manipulated with three predator treatments (faeces of red fox, common genet or pine marten). Traps were baited with 4 g of toasted corn, and food intake by wood mice was determined as the amount of bait remaining in each trap. We found that traps treated with faeces of resident predators were the most avoided, and the number of captures in traps treated with pine marten faeces was similar to the control‐baited traps. The variation found in food intake was explained by the interaction between the types of treatment and breeding condition. Food intake was similar in control‐baited traps and in traps with faeces of pine marten, but when predation risk by resident predators (red fox and common genet) was simulated, breeders reduced food intake significantly as compared to non‐breeders. These results indicate that predator recognition and feeding behaviour under predation risk depend on individual factors and the balance of costs‐benefits in each particular predation risk situation at a given place and time.     

Shrub expansion alters forest structure but has little impact on native mammal occurrence

The expansion (or encroachment) of shrubs in forests and woodlands is generally considered a serious threat to biodiversity. The effects of shrub expansion on forest fauna, however, are poorly understood and likely to depend on the availability of key resources in shrub‐encroached forest. Coranderrk Bushland, like many conservation reserves in south‐eastern Australia, is considered threatened by the spread of an indigenous shrub. We investigated the associations between cover of Yarra burgan (Kunzea leptospermoides (Myrtaceae)), vegetation structure and the occurrence of terrestrial native mammals within the reserve, basing our predictions on prior knowledge of burgan growth habits and fauna habitat preferences. We quantified burgan cover and other potentially important habitat attributes using structure surveys, and used motion‐sensing cameras to detect terrestrial mammals. Dense burgan cover was associated with less grass, a sparser understorey, and more cryptogams, dead trees and coarse woody debris. However, there was no evidence that these changes negatively affected native mammals: burgan cover had little influence on the occurrence of any species except swamp wallabies (Wallabia bicolor), which occurred in all areas of the reserve but shifted from sites with high burgan cover during the day to sites with low cover at night. Our findings contrast with those from grassland shrub‐expansion studies, where fauna generally show strong responses to shrub cover. The effects of shrub expansion on forest fauna may be mitigated by the greater pre‐existing structural diversity in forests or the longer time required for structural changes to be fully realized. The large quantities of dead wood in areas with high shrub cover may also provide compensatory resources for small mammals, while the proximity to un‐encroached areas may enable large herbivores to move between dense shelter and forage. Shrub‐encroached forests clearly provide resources for some native fauna, and management strategies need to consider the potential impacts of shrub removal on these taxa.     

Spatial and temporal scaling in habitat utilization by klipspringers (Oreotragus oreotragus) determined using giving‐up densities

An animal's pattern of habitat use can reveal how different parts of its environment vary in quality based on the costs (such as predation risk) and benefits (such as food intake) of using each habitat. We studied klipspringer habitat use in Augrabies Falls National Park, South Africa using giving‐up densities (GUDs; the amount of food remaining in a resource patch following exploitation) in experimental food patches. We tested hypotheses related to how salient habitat variables might influence klipspringers' perceptions of foraging costs. At small spatial scales (3–4 m), klipspringer GUDs did not vary with cover and open microhabitats, or with the four cardinal aspects (shading) around shrubs. Adding water adjacent to food patches did not influence GUDs, showing that water is not a limiting complementary resource to food. Generally, klipspringers do not appear to be physiologically constrained. There was no difference in GUDs between four daily time periods, or between summer and winter; however, a significant interaction effect of time‐of‐day with season resulted from GUDs during the midday time period in winter being significantly higher (perceived value lower) than during the same time period in summer. At moderate spatial scales (10–60 m), klipspringer GUDs increased with distance from rocks because of increased predation risk. Based on GUDs collected at the largest scale (two 4.41‐ha grids), klipspringers preferred foraging at greater distances from drainage lines and on pebble and cobble substrates. Overall, this study has shown the efficacy of measuring GUDs to determine klipspringers' habitat utilization while foraging.     

Microhabitat Preference of Egernia napoleonis in Undisturbed Jarrah Forest,and Availability and Introduction of Microhabitats to Encourage Colonization of Restored Forest

An animal's microhabitat requirements can impact its ability to colonize restored areas, particularly species requiring slow developing microhabitats, such as logs and woody debris piles. Introduction of these microhabitats may be required to facilitate colonization by some species. Restored bauxite mine‐pits in the Jarrah (Eucalyptus marginata) forest of south‐western Australia contain introduced log piles at densities of 1 ha^?1. However, these have not facilitated colonization by Napoleon's skink (Egernia napoleonis), which rely on logs for habitat and are largely absent from restored sites. We radio‐tracked 12 skinks in unmined forest to determine their microhabitat preferences and examined differences in vegetation structure, and microhabitat and food availability, between restored and unmined forests to identify reasons for their absence. Restored and unmined forests differed in canopy, mid‐ and understory cover and ground substrates, which were all potential barriers to colonization. Food availability was similar between restored and unmined forest, thus not a barrier to colonization. Skinks primarily utilized long logs, large woody debris piles, and large trees; microhabitats that were scarce or absent in restored sites and, therefore, potential barriers to colonization. Using this information, we introduced small woody debris piles into restored sites in close proximity to unmined areas containing skinks to facilitate skink colonization. This showed early signs of success and suggested that the lack of logs and woody debris were barriers to colonization. However, further monitoring is required to accurately determine the long‐term value of woody debris piles in facilitating skink colonization.     

The role of seed predation in the maintenance of the Cross Timbers ecotone of Oklahoma,USA

I investigated how seed predation differed among tree species and among microhabitats across the Cross Timbers and what that variation may tell us about how this ecotone is maintained. The ecotone is located in Oklahoma, USA, between the eastern deciduous forest and tallgrass prairie where seeds of eight common tree species were placed in three microhabitats (oak forest, tallgrass prairie, and sumac shrub/small-tree/grass mix). After nine days in the field, percent seeds remaining were scored for each of the 120 (8 species×3 microhabitats×5 replicates) dishes. I found for both wind-dispersed tree species, (ash, elm) there was significantly more predation in the prairie microsite, with similar small predation levels in the shrub and forest. For two of the three bird-dispersed species (dogwood, hackberry), there was significantly more predation in the prairie and shrub microsites compared to the forest. Red cedar, however, was not taken by predators very much anywhere. Finally, all three mammal-dispersed tree species (two oaks, pecan) showed significantly more predation in the shrub and forest microsites compared to the prairie. Whereas wind- and bird-dispersed species suffered less predation as microsites became more woody and dark, the dominant oaks showed the opposite trend. Consequently, seed predators are not preventing oaks from advancing across this ecotone, but yearly fluctuations in predator population density, especially in the shrub transitional zone, could be helping to maintain it.     

Forest fragmentation and rhinocryptid nest predation in central Chile

Fragmentation of forest landscapes can raise the intensity of nest predation by increasing the abundance and richness of generalist or introduced predators. Understory foraging birds, such as rhinocryptids, can be highly vulnerable to nest predation in fragmented landscapes because they often place their nests on the ground. Temperate deciduous forests in Chile have been intensively fragmented in the last centuries, causing changes in nest predator densities. We tested if predation of artificial nests, mimicking those of rhinocryptids, placed on and above ground was higher in the remnant fragments of central Chile due to an increase in predator abundance. The rate of nest predation in forest remnants was larger than in native continuous forest. Small mammals were the main nest predators. Despite high predation rates, the abundance of rhinocryptids is higher in forest remnants, suggesting that fragments might constitute ecological traps.     

Woody Debris in the Mangrove Forests of South Florida1

Woody debris is abundant in hurricane‐impacted forests. With a major hurricane affecting South Florida mangroves approximately every 20 yr, carbon storage and nutrient retention may be influenced greatly by woody debris dynamics. In addition, woody debris can influence seedling regeneration in mangrove swamps by trapping propagules and enhancing seedling growth potential. Here, we report on line‐intercept woody debris surveys conducted in mangrove wetlands of South Florida 9–10 yr after the passage of Hurricane Andrew. The total volume of woody debris for all sites combined was estimated at 67 m³/ha and varied from 13 to 181 m³/ha depending upon differences in forest height, proximity to the storm, and maximum estimated wind velocities. Large volumes of woody debris were found in the eyewall region of the hurricane, with a volume of 132 m³/ha and a projected woody debris biomass of approximately 36 t/ha. Approximately half of the woody debris biomass averaged across all sites was associated as small twigs and branches (fine woody debris), since coarse woody debris >7.5 cm felled during Hurricane Andrew was fairly well decomposed. Much of the small debris is likely to be associated with post‐hurricane forest dynamics. Hurricanes are responsible for large amounts of damage to mangrove ecosystems, and components of associated downed wood may provide a relative index of disturbance for mangrove forests. Here, we suggest that a fine:coarse woody debris ratio ≤0.5 is suggestive of a recent disturbance in mangrove wetlands, although additional research is needed to corroborate such findings.     

Ant colonization and coarse woody debris decomposition in temperate forests

Ants are ubiquitous, abundant and have widespread impacts on ecological communities and ecosystem processes. However, ant effects on coarse woody debris decomposition are unexplored. Several ant species colonize coarse woody debris for nesting, and this puts them in contact with fauna and microbes that utilize coarse woody debris as habitat and food, potentially influencing nutrient cycling and, ultimately, forest productivity. We report results from a field experiment employing 138 artificial ant nests (routed pine blocks) across five locations in southeastern US deciduous forests. We examine the correspondence between ant, termite and wood-eating fungi colonization and variation in coarse woody debris decomposition. After 1 year, nests colonized by ants had 5% more mass than those not colonized. Ant colonization corresponded with significantly less termite- and fungal-mediated decomposition of the nests. Without ants, termites removed 11.5% and fungi removed 4% more wood biomass. Ants, termites and wood-eating fungi all colonized pine nests where temperatures were highest, and ants also preferred higher soil moisture whereas termites and fungi responded negatively to high soil moisture when temperatures were higher. Ants reduce termite colonies through predation, and may inhibit fungi through the secretion of antimicrobial compounds. Our results indicate that interactions between forest understory ants, termites and fungi may influence the rate of coarse woody debris decomposition—biotic interactions that potentially influence forest structure and function.     

Different nest predator faunas and nest predation risk on ground and shrub nests at forest ecotones: an experiment and a review

This study examined predator faunas of artificial ground and shrub nests and whether nest predation risk was influenced by nest site, proximity to forest edge, and habitat structure in 38 grassland plots in south-central Sweden. There was a clear separation of predator faunas between shrub and ground nests as identified from marks in plasticine eggs. Corvids accounted for almost all predation on shrub nests whereas mammals mainly depredated ground nests. Nest predation risk was significantly greater for shrub than for ground nests at all distances (i.e. 0, 15 and 30 m) from the forest edge. However, nest predation risk was not significantly related to distance to forest edge, but significantly increased with decreasing distance to the nearest tree. Different corvid species robbed nests at different distances from the forest edge, with jays robbing nests closest to edges. We conclude that the relationship between the predation risk of grassland bird nests and distance to the forest edge mainly depends on the relative importance of different nest predator species and on the structure of the forest edge zone. A review of published articles on artificial shrub and ground nest predation in the temperate zone corroborated the results of our own study, namely that shrub nests experienced higher rates of depredation in open habitats close to the forest edge and that avian predators predominantly robbed shrub nests. Furthermore, the review results showed that predation rates on nests in general are highest <50 m inside the forest and lower in open as well as forest interior habitats (≥50 m from the edge). Received: 16 March 1998 / Accepted: 30 July 1998     

Regulation of forest defoliating insects through small mammal predation: reconsidering the mechanisms

Population densities of forest defoliating insects may be regulated by small mammal predation on the pupae. When outbreaks do occur, they often coincide with warm, dry weather and at barren forest sites. A proposed reason for this is that weather and habitat affect small mammal population density (numerical response) and hence pupal predation. We propose an alternative explanation: weather and habitat affect small mammal feeding behaviour (functional response) and hence the outbreak risks of forest pest insects. We report results from laboratory and field-enclosure experiments estimating rates of pupal predation by bank voles (Myodes glareolus) on an outbreak insect, the European pine sawfly (Neodiprion sertifer), at different temperatures (15 and 20 °C), in different microhabitats (sheltered and non-sheltered), and with or without access to alternative food (sunflower seeds). We found that the probability of a single pupa being eaten at 20 °C was lower than at 15 °C (0.49 and 0.72, respectively). Pupal predation was higher in the sheltered microhabitat than in the open one, and the behaviour of the voles differed between microhabitats. More pupae were eaten in situ in the sheltered microhabitat whereas in the open area more pupae were removed and eaten elsewhere. Access to alternative food did not affect pupal predation. The results suggest that predation rates on pine sawfly pupae by voles are influenced by temperature- and habitat-induced variation in the physiology and behaviour of the predator, and not necessarily solely through effects on predator densities as previously proposed.     

Effects of Structural Refuge and Density on Foraging Behaviour and Mortality of Hungry Tadpoles Subject to Predation Risk

Theoretical models of prey behaviour predict that food‐limited prey engage in risk‐prone foraging and thereby succumb to increased mortality from predation. However, predation risk also may be influenced by factors including prey density and structural cover, such that the presumed role of prey hunger on predation risk may be obfuscated in many complex predator–prey systems. Using a tadpole (prey) – dragonfly larva (predator) system, we determined relative risk posed to hungry vs. sated prey when both density and structural cover were varied experimentally. Overall, prey response to perceived predation risk was primarily restricted to increased cover use, and hungry prey did not exhibit risk‐prone foraging. Surprisingly, hungry prey showed lower activity than sated prey when exposed to predation risk, perhaps indicating increased effort in search of refuge or spatial avoidance of predator cues among sated animals. An interaction between hunger level and predation risk treatments indicated that prey state affected sensitivity to perceived risk. We also examined the lethal implications of prey hunger by allowing predators to select directly between hungry and sated prey. Although predators qualitatively favoured hungry prey when density was elevated and structural cover was sparse, the overall low observed variation in mortality risk between hunger treatments suggests that preferential selection of hungry prey was weak. This implies that hunger effects on prey mortality risk may not be readily observed in complex landscapes with additional factors influencing risk. Thus, current starvation‐predation trade‐off theory may need to be broadened to account for other mechanisms through which undernourished prey may cope with predation risk.