Eco-evolutionary experience and responses to biological control agents
Recent research on prey responses to exotic enemies emphasizes the importance of prey’s eco-evolutionary experience (EEE) with enemies (Blumstein 2006; Cox & Lima 2006; Sih et al. 2010; Saul & Jeschke 2015; Trimmer et al. 2017; Carthey & Blumstein 2018; Ehlman et al. 2019). By ‘eco-evolutionary experience’ we mean either an evolutionary history, or an earlier ecological (developmental) history that allowed prey to either evolve or learn to cope with a predator. Naïve prey that lack previous experience with a novel predator often respond insufficiently, suffering heavy predation (high CEs). Examples include the devastating impacts of novel predators (humans, other mammals, brown tree snakes) on naïve island prey, or of novel predatory fish on naïve prey in previously fishless lakes (Cox & Lima 2006). For classical biological control, the expectation is that if the target pest has had an extensive evolutionary history with the imported enemy, it will likely exhibit adaptive responses (and thus NCEs) that reduce CEs. In contrast, non-target prey that have not had previous EEE with the biocontrol agent might exhibit much weaker, if any, anti-predator response. If the predator can attack these non-target prey, then the biocontrol agent might prefer and exert strong CEs on non-target prey and less consumptive impact on the targeted pest.
            Some naïve prey, however, exhibit appropriate responses to novel predators. One key factor is the prey’s past history not with the specific novel predator, but with predation pressure in general. Prey that have experienced little predation pressure of any sort tend to be bolder and thus exhibit weaker response to novel predators, as compared to those that have evolved with moderate to heavy predation pressure (Ferrari et al. 2015). Therefore, non-target prey should be particularly vulnerable to novel biocontrol agents if those prey species have evolved with little predation (demonstrated by novel invasive social insects in Hawaii; Wilson et al. 2009; Krushelnycky et al. 2017). Recent work adds that if prey have experienced persistent high predation risk, then they should also be bold, not cautious. If predators are persistently present, prey cannot hide indefinitely, and should only respond strongly to cues that indicate particularly high impending risk (Trimmer et al. 2017; Ehlman et al. 2019). Additionally, though we focus on arthropod pests, work on invasive plants suggests that invasive species facing no top-down pressure may evolve to devote fewer resources to anti-enemy responses and more to competitive ability (Blossey & Notzold 1995). This process may be rapidly reversed upon the reintroduction of natural enemies through biological control programs, with invasive species rapidly developing anti-enemy responses that could drastically change the initial CE-NCE ratio (Stastny & Sargent 2017).
            Another key factor in predicting prey response to an introduced biocontrol agent is its similarity to familiar, native predators. Even if non-target prey have never experienced the particular novel predator, the ‘cue similarity’ hypothesis posits that if the introduced predator resembles familiar predators, ‘naïve’ prey are likely to respond (Sih et al. 2010; Saul & Jeschke 2015). Understanding the sensory/cognitive ecology of how target versus non-target prey perceive risk from biocontrol agents is then key (see Box 3). Even if prey correctly perceive the risk and respond, they can still suffer heavy predation if they show an inappropriate response (e.g., freeze when they should flee) or if their response is ineffective (e.g., they flee but the predator is too fast; Sih et al. 2010; Carthey & Blumstein 2018). Sih et al. (2010) suggested that the effectiveness of naïve prey responses to novel predators should depend on the functional ‘attack mode’ similarity of novel and familiar predators, and on whether prey rely on generalized responses (that work well against a broad range of predators) or specialized ones (that work very well, but only with specific predators). If the novel predator exhibits cue similarity but attack mode dissimilarity to familiar predators, it might induce both strong but ineffective responses that result in high CEs and high NCEs. This scenario could be ideal for suppressing target prey, but disastrous if it applies to nontarget prey.
            A community-level prediction is that prey should be more likely to respond well to a novel predator if the prey have eco-evolutionary experience (EEE) with a greater diversity of predator archetypes (Blumstein 2006; Cox & Lima 2006; Ehlman et al. 2019). If prey have EEE with only one main type of predator, they might exhibit predator-specific defenses. In contrast, if either target or nontarget prey have EEE with a broad range of predators, they should be more likely to exhibit a diversity of specialized and generalized defenses that could be effective against novel biocontrol agents.
            Finally, it is possible that contemporary evolution could occur during a long-term biocontrol relationship. While there are examples of evolved resistance to parasitism through enhanced immune responses (Berberet et al. 2003), we know of no cases where arthropod pests evolve anti-enemy responses to biocontrol agents. Hufbauer & Roderick (2005) thoroughly review microevolution in biocontrol, which may provide insights along with those gleaned from evolution of prey responses to invasive predators. Studying this directly in biocontrol systems would require measuring enemy-risk effects over long timescales, which could become a routine part of long-term efficacy studies.