Synthesis
Woven together, the ‘hunting mode-habitat domain’ and ‘evasion landscape’ concepts produce a general framework for predicting the nature and strength of NCEs on prey behavior during phase two. This framework predicts that four different patterns of anti-predator behavior can emerge depending on the degree of overlap between the habitat domains of the interacting predator and prey species and spatial variability in the efficacy of the prey’s evasion strategy (Fig. 4 ). NCEs manifest in all four scenarios and are expected to be especially strong in three of them.
When predator and prey exhibit narrow, overlapping habitat domains, and consequently encounters between them should be common, prey individuals are predicted to chronically invest in anti-predator countermeasures (Fig. 4a ). The nature of the investment, however, should depend on the prey’s evasion landscape. If the prey’s evasion landscape is spatially heterogeneous, enabling modification of the probability of surviving an encounter situation, then it should be both chronically vigilant and use space in a way that promotes the efficacy of its evasion strategy (e.g., by seeking backgrounds against which it is more camouflaged or, if the landscape lacks physical structure, grouping with conspecifics) (scenario one). If its evasion landscape is homogeneous, whereby the effectiveness of its evasion tactic is independent of location, then the prey individual should be chronically vigilant but only engage in evasion behavior such as fleeing when perceived risk is elevated (i.e., from an encounter situation up to an attack; scenario two). Risk effects and cascading indirect NCEs (phase three) under both of these scenarios are expected to be strong given the opportunity and energetic costs (Creel & Christianson 2008) and persistent changes to prey foraging and distribution resulting from chronic defensive investment.
When facing a predator with a narrow domain, prey individuals with broad domains should seek predator-free space via spatial shifts (scenario three; Fig. 4b ). These shifts should be chronic, given the high potential for encounters associated with use of the predator’s domain, and independent of the prey individual’s evasion landscape because avoidance of predators in space obviates the need for escape behaviors. This scenario should give rise to substantial risk effects and cascading indirect NCEs because of marked increases in intra-specific competition (e.g., from crowding in predator-free space) and changes to prey distribution accompanying chronic predator avoidance.
When a narrow prey domain falls within a broader predator domain, the predator should converge on the prey species, leading to high encounter rates (Fig. 4c ). Under these circumstances, prey individuals whose evasion landscape is heterogeneous should invest chronically in vigilance and use space in a way that facilitates their evasion strategy (scenario one), whereas those with homogeneous evasion landscapes should exhibit chronic vigilance and engage in evasion behavior only when perceived risk is heightened (scenario two). Risk effects and indirect NCEs under these conditions are probable. However, the degree to which predators converge on prey should depend on the relative importance of the prey in question to the energy budget of the predator. Hence, relatively simple predator-prey systems or situations in which the prey species is highly profitable to the predator should produce the strongest NCEs.
When predator and prey share broad, overlapping domains, encounters should be infrequent (Fig. 4d ). Given that anti-predator investment is not expected if the likelihood of predator-induced mortality in the absence of countermeasures is low (Peacor et al . 2013), joint investment in vigilance and evasive behavior is predicted under these circumstances (scenario four) only when the immediacy of perceived risk is elevated (i.e., a predator has been encountered) irrespective of the evasion landscape. Accordingly, CEs should predominate under this scenario, with ephemeral NCEs emerging as the result of a temporally dynamic landscape of fear for the prey.
In all scenarios, individual prey responses will be contingent on their defensive repertoire and state. For example, prey individuals relying exclusively on resistance, or that are constitutively defended, should invest minimally in behavioral countermeasures no matter how immediate the perceived risk cue is, save perhaps during an attack. Similarly, prey individuals that are naïve to predators or in compromised nutritional condition lack the experience, capacity, or incentive to respond behaviorally to perceived risk. Thus, well-defended populations or those with constrained opportunities for anti-predator investment (e.g., by low food supply; Bolnick & Preisser 2005) should be subject primarily to CEs.