Selective predation
Predators select their prey along a number of axes that can influence parasite loads and prevalences in the prey population. The healthy herds hypothesis was initially conceived on the basis that preferential predation on infected prey would lead to a decrease in parasitism at the population level (Hudson et al. 1992; Packer et al. 2003). It is reasonable to expect that infected individuals will often be more easily captured and eaten by predators (Hudson et al. 1992; Johnson et al. 2006). However, predators sometimes preferentiallyavoid consuming infected prey (Flick et al. 2016) and in their original formation of the healthy herd hypothesis, Packer et al. (2003) noted that this type of selective predation should cause predators to increase parasitism in their prey.
Predators also select prey in a variety of other ways, leading to effects of consumption on disease that are more indirect but still potentially important. For example, many predators prey preferentially on particular sizes or ages of prey (Price 1975; Nilsson & Brönmark 2000; King 2002), and prey sizes/ages commonly differ in their levels of infection (Dobson 1989). These intersecting patterns can result in functional predator-spreading via changes in prey population demography. For example, large intertidal snails (Littorina littorea ) are far more likely to carry trematode infections than their smaller (and younger) conspecifics, but these larger snails are also far less likely to be preyed on by shell-breaking predators such as crabs (Byerset al. 2015). As a result, predator preference for snail traits not directly related to infection resulted in much higher levels of parasitism in areas subjected to higher predation pressure (Byerset al. 2015), similar to the red curve in Fig. 3a.
This type of selective predation predator-spreading requires a clear predation preference (e.g., due to innate or learned behavioral preferences and/or biomechanics) that is negatively correlated with the infection rates in prey (as in the red curves in Fig. 3). Size/age may be the most straightforward version of this pattern, as in the case of the crab-snail-trematode system discussed above (Byers et al.2015) and systems where prey are attacked by a gape-limited predator and a parasite that is more likely to infect larger individuals (as in the case of the Chaoborus -Daphnia -fungal parasite system (Cáceres et al. 2009)). However, many other possibilities for traits where there might be a negative correlation between infection rates and predation risk exist, such as sex (Gwynne 1987; Acharya 1995; McCurdy et al. 1998; Reimchen & Nosil 2001; Lodé et al.2004; Krasnov et al. 2005; Harrison et al. 2010), reproductive status (Tait et al. 2008; VanderWaal & Ezenwa 2016), prey food choice (Kester & Barbosa 1994; Geervliet et al.1996; Garvey et al. 2020a, b), species assemblage (de Rijket al. 2013), and personality (Kortet et al. 2010). In any system where a trait increases the risk of predation but decreases the risk of parasitism, we expect predator spreading; conversely if there is a positive relationship between the trait and both infection risk and predation risk (as in the blue curves in Fig. 3), we expect predation to have a healthy herds effect.
Much of this logic assumes that the prey classes that are selectively removed by predators simply disappear without any other effects on prey population dynamics. However, the removal of large classes of a population can have immense effects on the growth, development, and reproduction rates of other groups which can, in turn, influence infection. For example, the loss of large bodied prey may make resources available for smaller or younger individuals, increasing their growth, maturation, and reproduction rates (Abrams & Rowe 1996; Relyea 2007), potentially influencing population level parasitism rates in unexpected ways, especially since susceptibility to infection commonly changes with age, as reviewed in Ben-Ami (2019). Moreover, in addition to the potential for these ecological dynamics to impact disease, rapid evolution in response to a predator may also promote parasitism, particularly if there is a tradeoff between resistance to predation and parasitism (Buss & Hua 2018).
As selective predators and unevenly distributed parasites are both ubiquitous, much additional experimental work is required to understand the interaction between these two processes. We encourage both predator-prey ecologists and disease ecologists to look at their study system from the alternative perspective and identify any potentially interacting patterns of parasite and predator selectivity. There are also likely extensive datasets on selective predation by human hunters with which these questions could be addressed (e.g. Chronic Wasting Disease in white tailed deer (Rivera et al. 2019)). This area would also benefit greatly from the combined usage of dynamical population modeling to make predictions for the outcomes of selective predation in a system and manipulative experiments to test these predictions (Box 3).