Animal sociality emerges from individual decisions that balance the benefits of associations against the costs of proximity or interactions with neighbours (Tanner and Jackson, 2012; Webber and Vander Wal, 2018; Webber et al., 2023; Gil et al., 2018). Such associations can yield useful social information whether inadvertently or deliberately transmitted about resource availability (Danchin et al., 2004; Dall et al., 2005; Gil et al., 2018), but they also provide opportunities for the transmission of parasites and infectious pathogens among associating individuals (Weinstein et al., 2018; Romano et al., 2020; Albery et al., 2021a; Cantor et al., 2021; Romano et al., 2022). Wildlife pathogen outbreaks affect most animal taxa, including mammals (Blehert et al., 2009; Fereidouni et al., 2019; Chandler et al., 2021; Kuchipudi et al., 2022), birds (Wille and Barr, 2022), amphibians (Scheele et al., 2020), and social insects (Goulson et al., 2015). Weighing the potential risk of infection from social interactions against the benefits of social movements where to move in relation to other individuals’ positions is thus a common behavioural context shared by many animal species. Movement decisions incorporating social information the presence and status of neighbours can facilitate or reduce spatial associations and help animals balance the costs and benefits of sociality (Albery et al., 2021a; Gil et al., 2018; Webber and Vander Wal, 2018; Webber et al., 2023). Animals’ social movements link landscape spatial structure, individual distributions, and the emergent structure of animal societies (Gil et al., 2018; Webber et al., 2023; Kurvers et al., 2014). Together, they influence the dynamics of disease outbreaks in animal populations (White et al., 2018b; Romano et al., 2020; Romano et al., 2022; Keeling et al., 2001), and such outbreaks may in turn have cascading effects on landscape structure and community ecology (Monk et al., 2022).

Over relatively brief ecological timescales of a few months or years, animal pathogen outbreaks can reduce social interactions among individuals due to a combination of factors. For instance, mortality from the disease may induce decreases in population density (e.g. Fereidouni et al., 2019; Monk et al., 2022), leading to fewer associations. Furthermore, adaptive behavioural responses by which animals identify infected individuals (and indeed, whether they are themselves infected) can trigger quarantining or self-isolation behaviours that reduce encounters between infected and healthy individuals overall (Stroeymeyt et al., 2018; Pusceddu et al., 2021; Stockmaier et al., 2021; Weinstein et al., 2018). When pathogens are first introduced into a population, such as during novel cross-species spillover (Kuchipudi et al., 2022; Chandler et al., 2021), fine-tuned avoidance responses are less likely, as individuals may have no prior experience of cues that indicate infection (Weinstein et al., 2018; Stockmaier et al., 2021; although general cues of infection may still play a role; see Townsend et al., 2020). A novel pathogen spreading through host–host contacts and imposing costs upon infected individuals could thus confer an evolutionary advantage upon less social individuals if these are also less frequently infected. Therefore, it is a common expectation that pathogen introduction broadly selects against host social behaviour, and hence against social connectivity itself (Altizer et al., 2003; Cantor et al., 2021; Romano et al., 2022; Poulin and Filion, 2021; Ashby and Farine, 2022).

Important aspects of animal ecology, including the transmission of foraging tactics (Klump et al., 2021) and migration routes (Jesmer et al., 2018; Guttal and Couzin, 2010), depend on social interactions. This makes it important to understand the long-term, evolutionary consequences of pathogen introductions for animal sociality. Climate change is only expected to make novel pathogen introductions more common (Sanderson and Alexander, 2020; Carlson et al., 2022a), making such studies more urgent. Despite this salience, novel pathogen introductions are primarily studied for their immediate demographic (Fey et al., 2015), and potential medical (Wille and Barr, 2022; Chandler et al., 2021; Kuchipudi et al., 2022; Levi et al., 2012) and economic implications (Keeling et al., 2001; Goulson et al., 2015; Jolles et al., 2021). Indeed, most introductions of novel pathogens into wildlife only come to light when they result in mass mortality events (Fey et al., 2015; Wille and Barr, 2022). Host evolutionary dynamics (and especially changes in sociality) are mostly ignored, and this is presumably because the evolution of pathogen host traits, and moreover complex behavioural traits such as sociality, is expected to be slow and not immediately relevant for management.

Theory suggests that animal sociality evolves to balance the value of social associations against the risk of pathogen transmission (Bonds et al., 2005; Prado et al., 2009; Ashby and Farine, 2022). However, analytical models often reduce animal sociality to single parameters, while it actually emerges from individual decisions conditioned on multiple internal and external cues. Social decision-making and movement often also vary among individuals (Tanner and Jackson, 2012; Wolf and Weissing, 2012; Spiegel et al., 2017; Gartland et al., 2022), but analytical models are unable to include individual differences in sociability. Epidemiological models based on contact networks can incorporate individual variation in social behaviour by linking these differences to positions in a social network (White et al., 2017; Albery et al., 2021a; Albery et al., 2021b). Yet network models often cannot capture fine-scale feedbacks between individuals’ social and spatial positions (Albery et al., 2021a; Albery et al., 2021b), nor spatial variation in infection risk (Albery et al., 2022). Networks constructed from relatively low-resolution spatial relocation data (such as infrequent direct observations; see e.g. Albery et al., 2021b), may be sensitive to the network formation process when seeking to understand the rapid spread of diseases, especially if transmission has a non-linear relationship with association strength (Farine, 2017; White et al., 2017). While high-resolution animal tracking could help construct more detailed networks on which to run disease outbreak models (Wilber et al., 2022; Nathan et al., 2022), such networks could also be biased by individual variation in social traits (Gartland et al., 2022), such as when sociality is correlated with capture probability (see e.g. Carter et al., 2012). Consequently, adding an explicit spatial setting to movement-disease models can be valuable in gaining a more general understanding of the interplay between social decisions, movement, and pathogen transmission (He et al., 2021; Scherer et al., 2020; White et al., 2018c; White et al., 2017).

Mechanistic, individual-based simulation models (IBMs) suggest themselves as a natural solution. IBMs can incorporate substantial ecological detail, including explicit spatial settings (DeAngelis and Diaz, 2019), and detailed disease transmission dynamics (White et al., 2018a; Scherer et al., 2020; Lunn et al., 2021; White et al., 2018c). Most importantly, IBMs can include individual decision-making, allowing ecological and epidemiological outcomes to emerge from individuals’ movement choices. Individual-based models hitherto have focused on immediate epidemiological outcomes, such as infection persistence, and do not have an evolutionary component examining long-term consequences for either pathogens or their hosts (White et al., 2018a; Scherer et al., 2020; Lunn et al., 2021). Incorporating an evolutionary component to movement-disease IBMs could allow predictions on important feedbacks between the proximate ecological outcomes of infectious disease and the ultimate consequences for the evolution of host behaviour (Cantor et al., 2021). This could include the emergence of individual differences in the trade-offs between the costs and benefits of sociability (Gartland et al., 2022), with cascading effects for landscape ecology and the structure of animal societies (Monk et al., 2022; Spiegel et al., 2017; Tanner and Jackson, 2012; Webber et al., 2023). The range of animal taxa at risk from a wide array of pathogens and parasites (Carlson et al., 2022a; Sanderson and Alexander, 2020) makes it important to conceive, as a starting point, of models that can capture the key features of diverse host–pathogen dynamics and offer broad conceptual insights (White et al., 2018c; White et al., 2018a).

We built a model that seeks to capture the essential elements of animal movement decisions in the context of foraging on patchily distributed resources, under the risk of pathogen (or parasite) transmission. Our model adopts a step-selection framework in an explicit spatial setting (Fortin et al., 2005), allowing individuals to choose their movement directions a key component of animal movement ecology (Nathan et al., 2008) based on their perception of local environmental cues. These are the presence of resources (personal information) and the presence of other individuals (social information). Our model also adds an evolutionary component by allowing individuals’ ecological performance (energy) over their lifetime to influence the mixture of movement strategies in their offspring generation. We examined the ecological and evolutionary consequences of the introduction of a pathogen into a novel host population (such as during cross-species spillover: Blehert et al., 2009; Bastos et al., 2000; Wille and Barr, 2022; Fereidouni et al., 2019; Scheele et al., 2020; Sanderson and Alexander, 2020; Carlson et al., 2022a; Kuchipudi et al., 2022; Monk et al., 2022). We modelled two scenarios of the introduction of an infectious pathogen to populations that had already evolved foraging movement strategies in its absence. Our model scenarios could be conceived as abstract representations of, among others, cross-species introductions of foot-and-mouth disease from buffalo to impala (Bastos et al., 2000; Vosloo et al., 2009), or of sarcoptic mange from llamas to vicuas (Monk et al., 2022), the current and historic spread of avian influenza among bird species (and more recently, spillovers into certain mammal species; Global Consortium for H5N8 and Related Influenza Viruses, 2016; Wille and Barr, 2022), of the spread of borrelliosis in novel populations of its wildlife hosts (Levi et al., 2012), or of SARS-CoV-2 from humans to deer (Chandler et al., 2021; Kuchipudi et al., 2022).

In scenario 1, we repeatedly introduced an infectious pathogen to a small proportion of individuals in each generation, allowing it to spread with a low probability among proximate individuals thereafter. This scenario parallels conditions that we expect are common but poorly known: that animal populations suffer pathogen introductions regularly from external sources such as individuals from an infected subpopulation of a metapopulation or sympatric heterospecifics such as those sharing breeding or wintering grounds both of these appear to be plausible events in the spread of diseases such as highly pathogenic avian influenza (Wille and Barr, 2022; Global Consortium for H5N8 and Related Influenza Viruses, 2016). We classified individuals across the evolutionary timescale of our simulation based on their inherited preferences (or selection coefficients) for environmental cues into movement strategies (similar to; ; Bastille-Rousseau and Wittemyer, 2019; see ‘Methods’). We compared how social information was used in movement strategies evolved before and after pathogen introductions began, and the ecological outcomes for individual movement and associations with other foragers. In a further scenario 2, we modelled only a single introduction event, but allowed the pathogen to be transmitted from parents to their offspring at the end of each generation (vertical transmission in a general sense), in addition to spreading among proximate individuals within each generation. Empirical examples of such parent-to-offspring transmission are less well known, but are implicated in the maintenance of foot-and-mouth disease in African buffalo (Jolles et al., 2021) and of mange among wolves (Almberg et al., 2015). We examined how these simulated outbreaks persisted across generations, the resulting evolutionary change in social movement strategies, and the consequences for individual behavioural outcomes. Using network epidemiological models (Wilber et al., 2022; Stroeymeyt et al., 2018; White et al., 2017; Bailey, 1975), we examined whether the spread of infections was reduced in pathogen-risk-adapted populations compared to their pathogen-risk-naive ancestors. We also investigated the effect of landscape productivity and the cost of infection, which are both expected to influence the selection imposed by pathogen transmission (Ezenwa et al., 2016; Almberg et al., 2015; Hutchings et al., 2000). Overall, we provide a theoretical framework applicable to a broad range of novel host–pathogen introduction scenarios and demonstrate the importance of including evolutionary dynamics in movement-disease models.

In our model, individuals move and forage on a landscape with patchily distributed food items, and select where next to move in their vicinity, based on inherited preferences for environmental cues food items, and other individuals (Figure 1). Food items, once consumed, regenerate at a rate $R$, and pathogen infection imposes a per-timestep cost $\delta E$. We classified individuals’ social movement strategies in our model using a simplified behavioural hypervolume approach (Bastille-Rousseau and Wittemyer, 2019) based on the sign of their preferences for successful foragers handling a food item (handlers, preference $s_H$) and for unsuccessful foragers still searching for food (non-handlers, preference $s_N$).

Figure 1

Download asset Open asset

In our models’ default implementation of scenario 1, $R$ = 2, food regenerates twice per generation, and $\delta E$ = 0.25, that is, consuming one food item offsets four timesteps of infection. Over the 500 generations before the introduction of the pathogen, populations reached an eco-evolutionary equilibrium where the most common social movement strategy was to prefer moving towards handlers while avoiding non-handlers (handler-tracking; ${\displaystyle s_H>0,s_N<0}$) (Figure 2A). This is consistent with observations from a different simulation model which shares many mechanisms with this one (Gupte et al., 2023). A small proportion of individuals prefer to move towards both handlers and non-handlers, and are thus indiscriminately social (agent-tracking; ${\displaystyle s_H,s_N>0}$).

Figure 2

Download asset Open asset

Movement-intake-sociality trade-offs and the coexistence of social movement strategies

At eco-evolutionary equilibrium in our default implementation of scenario 1 (3000 ≤ G ≤ 3500), the three main social movement strategies coexist, allowing a comparison of ecological and behavioural outcomes that illustrates the trade-offs between sociality, movement, and infection, which are otherwise masked by a population-level analysis. For example, the population-level increase in movement after pathogen introduction is shown to be due to the increase in frequency of the agent-avoiding strategy as these individuals move more than handler-tracking or agent-tracking foragers (Figure 3A). Simultaneously, agent-avoiding individuals have a lower intake than either handler-tracking or agent-tracking individuals which have similar intakes (Figure 3B). Surprisingly, the more social strategies appear to increase their intake slightly following pathogen introduction this could be because exploitation competition may be reduced as agent-avoiding foragers also avoid resource clusters and have less intake than the pre-introduction average. Despite this, all three strategies have comparable if not identical net energy and hence equivalent fitness this is to be expected given their coexistence (Figure 3C).

Figure 3

Download asset Open asset

The energy equivalence of the three strategies despite different per-capita intake can be explained by differing infection rates. These are in turn likely influenced by the non-linear relationship between movement and the mean number of per-capita associations of each strategy. The shape of the movement–association curve is broadly a quadratic one (Figure 3D). Across strategies, individuals that move more have more associations until a threshold, with associations declining from their peak as individual movement increases further; the peak of the curve is different for each strategy. For example, agent-tracking individuals that move 50 units have around 600 associations with other foragers, while handler trackers have approximately 300 associations, and agent-avoiding individuals have about 150 associations. At the extremes of movement behaviour individuals that move throughout their lifetime (movement >75) and which do not move at all (movement <15), all three strategies have similar numbers of per-capita associations; individuals that move constantly (movement = 100) have almost no associations at all. These differences likely explain why agent-avoiding and handler-tracking individuals have differing mean infection rates, at ∼25 and ∼33%, respectively (Figure 3E). Individuals of the agent-tracking strategy, on the other hand, have a wide range of infection rates (Figure 3E), potentially because they are rare these likely represent mutants that do not give rise to persistent lineages.

Changes to spatial-social structure and emergent superspreading

Following pathogen introduction, the mixture of individual-level movement strategies elicits a change in the emergent spatial and social structure at the population level. Pre-introduction populations are spatially clustered near food item patches (Figure 4A) due to movement strategies that favour grouping with successful foragers. Pathogen-risk-adapted populations are more dispersed over the landscape, with many individuals found far from food item clusters (Figure 4B). This reflects the increased prevalence of the agent-avoiding strategy which leads to a sort of dynamic social distancing. The change in the mixture of population social movement strategies is reflected in the left-skewed degree distributions of pathogen-risk-adapted populations compared to pathogen-risk-naive ones (Figure 4C).

Figure 4

Download asset Open asset

We examined the distribution of individual reproductive numbers ($\nu$) from two separate intervals in the simulation: just after pathogen introductions begin (3000 ≤ G ≤ 3100), and 500 generations after introductions begin (3500 ≤ G ≤ 3600). Individual reproductive number distributions from both intervals are strongly left-skewed but have long right-hand tails (up to 12 just after introductions begin; Figure 4D). While most infected individuals do not infect any of their neighbours, a small number of these are responsible for a disproportionately large number of further infections, even after the population has adapted to moving under the risk of transmission (Figure 4D); this is consistent with the phenomenological definition of superspreading (Lloyd-Smith et al., 2005). Our model thus shows how, even in a population with identical individuals that differ only in their movement decision-making rules, there can be substantial variation in individuals’ contribution to the spread of an infectious pathogen.

Pathogen-risk-adapted movement strategies and the spread of infection

A large majority of individuals in the generations just after pathogen introduction are infected (≈75%; Figure 5A). However, tracking the evolutionary change in movement strategies, the number of infected individuals falls to below 50% within 100 generations (Figure 5A), remaining low for the rest of the simulation. To examine potential pathogen spread in pre-introduction populations, we ran a simple epidemiological model on the social networks emerging from individuals’ movements before and after pathogen introduction (pre-introduction: G = 500; post-introduction: G = 700). We modelled two infections, (1) first, an infection requiring one encounter, and (2) second, an infection requiring 10 encounters between individuals for a potential transmission event (transmission rate $\beta$ = 5.0, recovery rate $\gamma$ = 1.0).

Figure 5

Download asset Open asset

Both the single-encounter and multiple-encounter diseases would infect >75% of individuals overall when spreading through the networks of pre-introduction populations (Figure 5B). Pathogen-risk-adapted populations’ social networks are however more resilient to the multiple-encounter infection compared to their pre-introduction, pathogen-risk-naive ancestors as these social networks are sparser and individuals are more weakly connected (Figure 5B). While nearly all individuals in post-introduction populations would be finally infected by the single-encounter infection the same as their pre-introduction, pathogen-risk-naive ancestors, the spread of the multiple-encounter infection would be substantially reduced in comparison (ever infected: ≈ 50%).

Effect of landscape productivity and infection cost

For our scenario 1, we further explored the effect of two ecological parameters, landscape productivity (${\displaystyle R\in 1,2,5}$) and infection cost per timestep ($\delta E\in$ 0.1, 0.25, 0.5), on simulation outcomes. Before pathogen introduction, the same social movement strategies evolve on landscapes of all productivity levels (Figure 6).

Figure 6

Download asset Open asset

Pathogen persistence after a single introduction with vertical transmission

In our scenario 2, we introduced the pathogen only once to 4% (N = 20) individuals in generation 500, and this more closely simulates the sort of introduction that would be expected in a novel, cross-species spillover. Focusing on our default parameter combination ($R$ = 2, $\delta E$ = 0.25, p_v = 0.2), we observed that prior to pathogen introduction the population followed the same ecological and evolutionary principles we laid out for scenario 1, and all replicates were similar (Figure 7A). The pathogen is successfully transmitted from parents to offspring in the initial generations following the introduction event and among individuals of the same generational cohort. This produces ecological patterns very similar to scenario 1, with large numbers of infections (Figure 7A).

Figure 7

Download asset Open asset

Infection cost as a reduction in foraging efficiency

We considered an alternative implementation in which the infectious pathogen is considered to reduce an animal’s ability to process intake, or to require a portion of daily intake to resist, such that an individual with lifetime intake $N$, has a net energetic gain $E=N{(1-\delta E)}^t$ after being infected by a pathogen for $t$ timesteps. In this implementation, there is a rapid evolutionary shift in movement strategies after pathogen introduction, similar to that in our default implementation, but only when the costs of infection are relatively high ($\delta E$ = 7.5%), and the usefulness of social information is limited by the abundance of food items ($R$ = 5). Under these conditions, the agent-avoiding strategy becomes the most common strategy. Under conditions of median landscape productivity and intermediate to high pathogen costs ($\delta E\in$ 5.0% and 7.5%, $R$ = 2), the agent-avoiding strategy also emerges, but forms only a low proportion of the population. Under all other conditions, the handler-tracking strategy continues as the most common strategy (Figure 8).

Figure 8

Download asset Open asset

Large-scale natal dispersal of individuals

Our model implements small-scale or local natal dispersal and individuals are initialised close to their parents’ last position a defensible choice as many organisms do not disperse very far from their ancestors. An alternative implementation is to initialise individuals in each new generation at random locations on the landscape (see e.g.; Gupte et al., 2023); this can be called large-scale or global natal dispersal. This may be a reasonable choice when modelling animals during a specific stage of their life cycle, such as after arriving on a wintering or breeding site after migration. When animals do not disperse very far, they may adapt their movement strategies to the local conditions which they inherit from their parents (ecological inheritance) (Badyaev and Uller, 2009). By forcing animals in each new generation to encounter ecological circumstances potentially different from those of their parents, implementing global dispersal can help investigate whether animals’ evolved movement strategies are truly optimal at the global scale (Gupte et al., 2023). We implemented global dispersal by running 10 replicates of each parameter combination (three combinations of $\delta E$ = 0.25 and $R\in$ 1, 2, 5; 30 simulations in all), with dispersal set to 10. This means that individuals’ initial positions are drawn from a normal distribution with standard deviation = 10, centred on the location of their parent.

We found that our model is broadly robust to implementing large-scale natal dispersal, with the evolutionary outcomes very similar to those seen in our default implementation with small-scale natal dispersal (Figure 9A). Most individuals are handler-tracking before the introduction of the novel pathogen, which makes them to gain the benefits of social information on the location of a resource patch (of which handlers are an indirect cue), while avoiding potential competitors, as well as potentially moving away from areas without many food items. After pathogen introduction, there is a rapid evolutionary shift in social movement strategies, with an increase in agent avoidance, similar to the shift seen in our default implementation of local dispersal. The effect of landscape productivity on the mix of proportions of the pre- and post-pathogen introduction strategies does not appear to be significant.

Figure 9

Download asset Open asset

Spatial structure of the resource landscape

Since ours is a spatial model, and the explicit consideration of space and movement is key to its outcomes, we very briefly examined the effect of landscape spatial structure on the evolutionary outcomes of our scenario 1 (Figure 10). We considered two alternative food item distributions: (1) food items distributed uniformly across the landscape, and (2) food items more patchily distributed than the default, with only 10 food item clusters (default = 60). We compared the outcomes on these landscapes with those from our default scenario, with all parameters expect spatial structure kept the same ($R$=2, $\delta E$ = 0.25, $N$ food items = 1800; Figure 10B).

Figure 10

Download asset Open asset

Landscape spatial structure influences the mixture of social movement strategies evolved before pathogen introductions (Figure 10A). On the uniform landscape, handler-tracking was the most common strategy before pathogen introduction, with nearly all individuals of this strategy. In contrast, on the more patchy landscape, the indiscriminately social agent-tracking strategy was the most common before pathogen introductions. Both of these are in contrast with our default scenario, in which most individuals were handler-tracking, but with a substantial proportion of agent-tracking individuals.

This overall pattern is likely due to the increasing benefit of social information and the increasing costs of movement between profitable areas of the landscape. As landscapes become more clustered, direct food item cues become more difficult to find, as food items are found in smaller and denser patches. This increases the value of sociality as individuals are likely to found near food item clusters. Furthermore, the indirect costs of movement also increase on patchy resource landscapes as individuals have to pay an increased cost in time (which could have been spent foraging) in moving between food item clusters. In an implementation not formally shown here, the same effect can be achieved by adding a small cost to each movement step; this leads to the evolution of indiscriminate sociality in the form of agent-tracking on the default landscape as well. Overall, both the increasing local density of food items and the costs of movement lead to an increase in agent-tracking as individuals prefer to trade movement costs for the costs of increased local competition for food items.

Following pathogen introduction, populations on both landscapes undergo a rapid evolutionary transition to a mixture of handler-tracking and agent-avoiding strategies, which is similar to the change observed in our default scenario (Figure 10A). However, the landscapes differ in the proportions of the two strategies, with agent avoidance more common on the uniform landscape than on the patchy landscape. Interestingly, both of these extremes of landscape structure have more agent-avoiding individuals than our default landscape of 60 food item clusters. On the uniform landscape, this is likely because food items are readily found with the need for indirect social cues, and so most individuals avoid each other. It is less clear why this is the case on the more patchy landscape; it is possible that the denser food item patches lead to more associations and more rapid pathogen spread, with handler-tracking individuals infected for longer periods than agent-avoiding ones, leading to a stronger intake-infection trade-off. Overall, this scenario demonstrates how spatial structure can play an important role in the evolution of social movement strategies, but also how the risk of infection can lead to landscapes with very different spatial structures eventually populated by similar social movement strategies.

Sporadic introduction of infectious pathogens

Finally, we implemented a variant of our main model, in which the infectious pathogen is introduced only sporadically after the first introduction event (at G = 3000). Specifically, we modelled probabilistic introduction of the pathogen in each generation following the initial introduction. We call the per-generation probability of a novel pathogen introduction event the spillover rate, and we ran this model variant for three values of the spillover rate: 0.05, 0.1, and 0.25. Instead of examining the joint effect of landscape productivity and cost of infection as well, we only examined the effect of infection cost, implementing three different variants with an infection cost $\delta E$ of 0.1, 0.25, and 0.5. We kept all other model parameters similar to the default scenario of our main model, and importantly, considered only a landscape productivity $R$ of 2.

Following pathogen introduction, we found that there was little to no change in the population-level mixture of movement strategies in this model variant (Figure 11). This is regardless of the probability of a novel pathogen introduction, and the cost of infection by a pathogen. Across the simulation, the most common social movement strategy remains handler-tracking, that is, preferring locations with multiple individuals regardless of their foraging status. Since there is little to no change in social movement strategies, we did not expect nor find changes in ecological outcomes.

Figure 11

Download asset Open asset

Our general model captures important features of infectious pathogen or parasite transmission among host animals in a foraging context that is relevant to many species. Adding an explicit spatial setting has allowed us to more finely probe the effects of individual behavioural variation, pathogen characteristics, and landscape properties on the emergence of animal sociality and the spread of disease. The mechanistic combination of ecological, evolutionary, and epidemiological dynamics in a spatial setting is unprecedented for host movement-disease models (White et al., 2018b; Manlove et al., 2022). The key feature of our approach is to let the ecological outcomes (intake, time infected) of individual social movement decisions in one generation affect the mixture of social movement strategies of the next generation. Our approach shows how host evolutionary dynamics can be incorporated into mechanistic movement-disease models (Manlove et al., 2022) and how this approach extends current understanding of the evolutionary causes and consequences of animal spatial and social behaviours (Albery et al., 2021a; Webber and Vander Wal, 2018; Webber et al., 2023; Romano et al., 2020; Romano et al., 2022; Kurvers et al., 2014). To aid in the uptake of our modelling approach, we provide both a written description of the model (see ‘Methods’) as well as the full, documented source code (see ‘Data availability’).

Presently, most movement-disease models are non-evolutionary (White et al., 2018a; White et al., 2017; Scherer et al., 2020; Lunn et al., 2021; Manlove et al., 2022), presumably because evolution is expected to be too slow to impact epidemiological–ecological outcomes. We demonstrate the pitfalls of this assumption: evolutionary transitions in sociality occur within only a few generations, comparable to the time required for the development of key social aspects of animal ecology, such as migration routes (Jesmer et al., 2018; Cantor et al., 2021). We also demonstrate the tension inherent to sociality under the risk of an infectious pathogen in an explicitly spatial context. We show how populations, initially evolved to find patchily distributed food using social information, rapidly evolve to become more sensitive to potential infection risk and eschew social encounters when an infectious pathogen is introduced. Our results suggest how qualitatively and quantitatively different social movement strategies making different trade-offs between social information and infection risk can coexist in a single population (Wolf and Weissing, 2012; Webber and Vander Wal, 2018; Gartland et al., 2022; Webber et al., 2023; Wolf et al., 2008). Furthermore, our model shows how these trade-offs are outcomes of movement decisions, an aspect which would be difficult to study in a non-spatial model.

Following pathogen introduction, the evolutionary shift in social movement strategies is much more rapid than the timescales usually associated with the evolution of complex traits such as sociality (about 100 generations). Avoiding potentially infectious individuals is a key component of navigating the landscape of disgust (Weinstein et al., 2018). Our results show that sensitivity to cues of high pathogen transmission risk can rapidly evolve following the introduction of a novel pathogen. The emergence of qualitative individual variation in social movement strategies, and especially the trade-off between movement, associations, and infection risk, also demonstrates the evolution of sociability as a personality trait (Gartland et al., 2022). We also find substantial individual variation in the quantitative importance of social cues overall, which is a key component of the evolution of large-scale collective behaviours, such as migration (Guttal and Couzin, 2010). Our work suggests how, by leading to the necessary diversity in social movement strategies, a novel pathogen may actually lay the groundwork for the evolution of more complex collective behaviour. Nonetheless, the rapid decreases in social interactions should primarily prompt concern that the evolutionary consequences of pathogen introduction could slow the transmission of, and erode, animal culture (Cantor et al., 2021) including foraging (Klump et al., 2021) and migration behaviours (Jesmer et al., 2018; Guttal and Couzin, 2010).

Pathogens themselves typically have shorter generation times than their hosts and may also evolve rapidly in response to changes in host sociality (Prado et al., 2009; Ashby and Farine, 2022; Bonds et al., 2005). Our aim was to investigate how host behaviour evolved according to a predetermined (but varied) suite of pathogen characteristics across different simulation runs. Furthermore, we wanted to examine the effects of introduction events which are expected to become more common (Carlson et al., 2022a), but which need not necessarily lead to the pathogen becoming endemic in a population. Holding the pathogen traits steady and unable to evolve in the course of a simulation is thus a necessary choice in order to gain these first tangible insights from our model. Allowing simultaneous antagonistic coevolution between trophic levels, such as hosts and pathogens or predators and prey, could exponentially complicate the findings of a given eco-evolutionary model, such as by producing generationally staggered outcomes or cyclical Red Queen patterns (Prado et al., 2009; Netz et al., 2022), and can require much longer runs to attain stationary results or to identify optimal strategies. However, pathogen evolution in response to host behaviour is something that we would be excited to investigate in the future using this modelling framework. Indeed, a mixture of host social strategies could allow for the maintenance of a corresponding diversity in pathogen strategies as well (Ashby and Farine, 2022; Prado et al., 2009) as is also seen in predator–prey co-evolution (Netz et al., 2022). One conceptual impediment is modelling pathogen traits in a mechanistic way. For example, it is widely held that there is a trade-off between infection cost and transmissibility with a quadratic relationship between them (Ashby and Farine, 2022; Bonds et al., 2005; Prado et al., 2009), but this is a pattern reported from empirical studies and not a process per se. A tractable starting point might be to adapt our scenario 2 with vertical transmission to examine the evolution of pathogen traits that influence both transmissibility and virulence with an unchanging host (such as an adaptation of Lion and Boots, 2010).

In our model, landscape productivity ($R$) is a proxy for the usefulness of sociality overall as social information is less useful when direct resource cues are abundant (high $R$; see also) (Gupte et al., 2023). Social information benefits in disease models often have no mechanistic relationship with the subject of the information (e.g. food or predators; Ashby and Farine, 2022). In contrast, social information benefits in our model are emergent outcomes of animal movement and foraging behaviour which is only possible due to the explicit spatial nature of our model. It is surprising then that landscape productivity does not strongly influence the evolution of social movement strategies, but this may yet be an important factor in enabling high-movement, low-infection strategies when movement is inherently costly. In our model, movement has an indirect time cost; moving away from food items leaves less time in which to make up fitness differences with other individuals through foraging. This is essentially why we find that landscape spatial structure strongly influences the mixture of social strategies evolved before pathogen introduction. However, we found that across a spectrum of spatial structures, pathogen introduction resulted in a convergence in social movement strategies; this evolutionary component may be an important consideration in studies of how spatial structure can influence the spread of infection (He et al., 2021; Scherer et al., 2020; White et al., 2018c; White et al., 2017). Furthermore, movement can be an energetically demanding process that could influence whether dynamic social distancing to avoid infection risk, as evolved in our model, would be a viable movement strategy. Future extensions of our model could add a small cost to movement in order to explore the interplay of landscape productivity and spatial structure in determining direct indirect movement costs and the consequences for social movement strategies.

Infection costs do affect which social movement strategies evolve in our model and may help explain intra- and inter-specific diversity in social systems across gradients of infection costs (Altizer et al., 2003; Sah et al., 2018). Studies tracking social movements and the potential for pathogen spread could form initial tests of our basic predictions (Wilber et al., 2022). Our model suggests that animal populations may be able to adapt relatively quickly to the spillover and eventual persistence of infectious pathogens, even when they cannot specifically detect and avoid infected individuals (Altizer et al., 2003; Stroeymeyt et al., 2018; Stockmaier et al., 2021; Pusceddu et al., 2021). While the most noticeable effect of pathogen outbreaks is mass mortality (Fey et al., 2015), even quite serious pathogens such as sarcoptic mange (Almberg et al., 2015), foot-and-mouth disease (Jolles et al., 2021; Bastos et al., 2000; Vosloo et al., 2009), SARS-CoV-2 (Chandler et al., 2021; Kuchipudi et al., 2022), and avian influenza (Global Consortium for H5N8 and Related Influenza Viruses, 2016; Wille and Barr, 2022), among others, appear to spread at sublethal levels for many years between lethal outbreaks. Our model shows how population-level behavioural changes could occur even without mortality effects due to evolutionary shifts in sociality alone. The pathogen-risk-adapted population in our model are unable to escape infection entirely and have significantly worse net energy per-capita (just over zero), which could leave them vulnerable to extreme ecological conditions. Our work suggests that decreased sociality resulting from adaptation to a novel pathogen could slow the transmission of future novel pathogens. While decreased sociality could also reduce the prevalence of previously endemic pathogens adapted to a more social host, it may also degrade social immunity through reduced sharing of beneficial commensal microbes, or of low, immunising doses of pathogens (Ezenwa et al., 2016; Almberg et al., 2015).

The results of our scenario 1 are contingent upon sustained introduction of the pathogen (or its novel strains) to host populations. More sporadic introductions (once every few generations) apparently do not cause evolutionary shifts in social movement. Our scenario 2, which includes transmission from parents to offspring, suggests a mechanism by which such sporadic events, or even a single cross-species spillover event, could have far-reaching evolutionary consequences. Such vertical transmission is believed responsible for the circulation of foot-and-mouth disease in African buffalo (Jolles et al., 2021) and of mange among wolves (Almberg et al., 2015). Pathogen persistence across a broad swath of parameter combinations for scenario 2 suggests that even single introduction events can lead to a population rapidly becoming a novel source of transmission (loosely speaking, a reservoir) for other, overlapping species. Such dynamics would likely be increased should vertical transmission be coupled with multiple, sporadic pathogen or parasite introductions, which appear to be common in nature (Levi et al., 2012; Jolles et al., 2021; Vosloo et al., 2009; Bastos et al., 2000; Scherer et al., 2020; Global Consortium for H5N8 and Related Influenza Viruses, 2016; Wille and Barr, 2022). By demonstrating the multiple ways in which pathogens can affect an animal population, our model suggests how disease is a powerful selective force in favour of detecting and avoiding infection risk cues (Weinstein et al., 2018), among which are social cues.

We note that the pathogen characteristics (infection cost) as well as the probability of vertical transmission affect the evolutionary dynamics in scenario 2. In the context of our model, the latter could be interpreted as a factor influencing the association between parents and offspring, such as the length of parental care. This suggests that a directly transmitted novel pathogen should become established readily in species with greater social associations between generations, such as parental care of young Chakarov et al., 2015; this may, however, be counteracted by suites of infection-risk-reducing behaviours on the part of adults (Stroeymeyt et al., 2018; Ratz et al., 2021). Positively, we also find that when the pathogen is eliminated from the population, there is a near instantaneous shift towards (or recovery in) animal sociality. This suggests that if pathogens are extirpated from parts of their former ranges (due to a range of mechanisms, with climatic change as an influence) (Carlson et al., 2022b), some animal populations may show a hitherto unexpected increase in sociality, and potentially, novel social behaviours and structures or other aspects of animal culture. Our findings thus suggest an additional consideration when thinking about implementing campaigns that seek to reduce wildlife disease burdens, such as through wildlife immunisation (Barnett and Civitello, 2020; Ezenwa and Jolles, 2015).

In order to be widely applicable to diverse novel host–pathogen introduction scenarios, our model necessarily makes quite general assumptions. For example, our individuals use both personal and inadvertent social information whenever it is available; even though animals’ use of information sources does depend on their behavioural context, this could be examined more thoroughly in future implementations. A wide diversity of pathogens and their dynamics remains to be accurately represented in individual-based models (White et al., 2017; White et al., 2018a; Scherer et al., 2020; Lunn et al., 2021). Our framework could be expanded and specifically tailored to real-world situations in which animal populations are exposed to novel pathogens (or strains) that transmit between individuals (Scherer et al., 2020; Wille and Barr, 2022; Bastos et al., 2000; Jolles et al., 2021; Chandler et al., 2021; Kuchipudi et al., 2022). Such detailed implementations could include aspects of the pathogen life cycle (White et al., 2018b; White et al., 2017), account for sociality as a counter to infection costs (Ezenwa et al., 2016; Almberg et al., 2015), or model host–pathogen sociality-virulence co-evolution (Ashby and Farine, 2022; Prado et al., 2009; Bonds et al., 2005). Our work could serve as a good base for future models that focus on the importance of other factors, especially more nuanced implementations of reproduction and demography on the evolution of spatial-social strategies under infection risk. For instance, allowing sexual reproduction and considering the effects of infection status on mate choice or limiting pairing to nearby individuals could help explore how individual movement decisions can scale up to speciation and community assembly (Getz et al., 2016; Getz et al., 2015). Future empirical extensions of our work would ideally combine wildlife monitoring and movement tracking across gradients of pathogen prevalence to detect novel cross-species spillovers (Chandler et al., 2021; Kuchipudi et al., 2022) and study the spatial and epidemiological consequences of animal movement strategies (Bastille-Rousseau and Wittemyer, 2019; Wilber et al., 2022; Monk et al., 2022). Our model shows why it is important to consider evolutionary responses in movement-disease studies and provides a general framework to further the integration of evolutionary approaches in wildlife spatial epidemiology.

We implemented an individual-based simulation model to represent foraging animals (foragers) making movement decisions in an explicit spatial context. Individuals seek out discrete, immobile, depletable food items from which they gain energy that can be devoted to reproduction (similar to capital breeding; see Figure 1; Spiegel et al., 2017; Gupte et al., 2023). Food items are distributed over a two-dimensional, continuous-space resource landscape with wrapped boundaries (a torus). Our model, similar to earlier IBMs with both ecological and evolutionary dynamics (Getz et al., 2015; Netz et al., 2022; Gupte et al., 2023), has two distinct timescales: (1) an ecological timescale comprising of T timesteps that make up one generation ($T$=100 by default), and (2) an evolutionary timescale consisting of 5000 generations (G). At the ecological timescale, individuals perceive cues from their local environment: the presence and numbers of food items and other individuals. Individuals make movement decisions according to their inherited movement strategies (see below), and when chancing upon food items, consume them. At the same timescale, individuals that carry an infectious, fitness-reducing pathogen may, when in close proximity with uninfected individuals, pass on the pathogen with a small probability (see ’Pathogen introduction, transmission, and infection cost). At the evolutionary timescale, individuals reproduce and transmit their inherited cue preferences, and hence their movement strategies (see ‘starting location and inheritance of movement rules’) to their offspring. The number of offspring is linked to individuals’ success in finding and consuming food items, and to the duration that they were infected by the pathogen at the ecological timescale; this is in line with the replicator equation (Hofbauer and Sigmund, 1988). The model was implemented in R and C++ using Rcpp (R Development Core Team, 2020; Eddelbuettel, 2013) and the Boost.Geometry library for spatial computations (https://www.boost.org/); see the ‘Data availability’ statement for the code archive and development repository.

This manuscript presents the results of a simulation model study, and no real data were generated. The Pathomove simulation model code (v.1.2.0) is available on Zenodo at https://doi.org/10.5281/zenodo.7789072 (Gupte, 2023) and on Github at https://github.com/pratikunterwegs/pathomove. Code to run the simulations and analyse the output is on Zenodo at https://doi.org/10.5281/zenodo.7789079 and on Github at: https://github.com/pratikunterwegs/patho-move-evol (v.1.1.0). The data presented in this manuscript are also archived on Zenodo with the DOI https://doi.org/10.5281/zenodo.7789060.

1. 1. Gupte PR

   (2023) Zenodo

   Reference data from the Pathomove simulation, for the manuscript "Novel pathogen introduction triggers rapid evolution in animal social movement strategies".

   https://doi.org/10.5281/zenodo.7789060
2. 1. Gupte PR
   2. Albery GF
   3. Gismann J
   4. Sweeny AR
   5. Weissing FJ

   (2023) Zenodo

   Source Code and Supplementary Material for "Novel pathogen introduction triggers rapid evolution in animal social movement strategies".

   https://doi.org/10.5281/zenodo.7789079

1. 1. Albery GF
   2. Becker DJ
   3. Brierley L
   4. Brook CE
   5. Christofferson RC
   6. Cohen LE
   7. Dallas TA
   8. Eskew EA
   9. Fagre A
   10. Farrell MJ
   11. Glennon E
   12. Guth S
   13. Joseph MB
   14. Mollentze N
   15. Neely BA
   16. Poisot T
   17. Rasmussen AL
   18. Ryan SJ
   19. Seifert S
   20. Sjodin AR
   21. Sorrell EM
   22. Carlson CJ

   (2021a) The science of the host–virus network

   Nature Microbiology 6:1483–1492.

   https://doi.org/10.1038/s41564-021-00999-5

   • Google Scholar
2. 1. Albery GF
   2. Kirkpatrick L
   3. Firth JA
   4. Bansal S

   (2021b) Unifying spatial and social network analysis in disease ecology

   The Journal of Animal Ecology 90:45–61.

   https://doi.org/10.1111/1365-2656.13356

   • PubMed
   • Google Scholar
3. 1. Albery GF
   2. Sweeny AR
   3. Becker DJ
   4. Bansal S

   (2022) Fine‐scale spatial patterns of wildlife disease are common and understudied

   Functional Ecology 36:214–225.

   https://doi.org/10.1111/1365-2435.13942

   • Google Scholar
4. 1. Almberg ES
   2. Cross PC
   3. Dobson AP
   4. Smith DW
   5. Metz MC
   6. Stahler DR
   7. Hudson PJ

   (2015) Social living mitigates the costs of a chronic illness in a cooperative carnivore

   Ecology Letters 18:660–667.

   https://doi.org/10.1111/ele.12444

   • PubMed
   • Google Scholar
5. 1. Altizer S
   2. Nunn CL
   3. Thrall PH
   4. Gittleman JL
   5. Antonovics J
   6. Cunningham AA
   7. Dobson AP
   8. Ezenwa V
   9. Jones KE
   10. Pedersen AB
   11. Poss M
   12. Pulliam JRC

   (2003) Social organization and parasite risk in mammals: Integrating theory and empirical studies

   Annual Review of Ecology, Evolution, and Systematics 34:517–547.

   https://doi.org/10.1146/annurev.ecolsys.34.030102.151725

   • Google Scholar
6. 1. Ashby B
   2. Farine DR

   (2022) Social information use shapes the coevolution of sociality and virulence

   Evolution; International Journal of Organic Evolution 76:1153–1169.

   https://doi.org/10.1111/evo.14491

   • PubMed
   • Google Scholar
7. 1. Badyaev AV
   2. Uller T

   (2009) Parental effects in ecology and evolution: mechanisms, processes and implications

   Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 364:1169–1177.

   https://doi.org/10.1098/rstb.2008.0302

   • PubMed
   • Google Scholar
8. Book

   1. Bailey NTJ

   (1975)

   The Mathematical Theory of Infectious Diseases and Its Applications

   London: Griffin.

   • Google Scholar
9. 1. Barnett KM
   2. Civitello DJ

   (2020) Ecological and evolutionary challenges for wildlife vaccination

   Trends in Parasitology 36:970–978.

   https://doi.org/10.1016/j.pt.2020.08.006

   • PubMed
   • Google Scholar
10. 1. Bastille-Rousseau G
    2. Wittemyer G

    (2019) Leveraging multidimensional heterogeneity in resource selection to define movement tactics of animals

    Ecology Letters 22:1417–1427.

    https://doi.org/10.1111/ele.13327

    • PubMed
    • Google Scholar
11. 1. Bastos AD
    2. Boshoff CI
    3. Keet DF
    4. Bengis RG
    5. Thomson GR

    (2000) Natural transmission of foot-and-mouth disease virus between African buffalo (Syncerus caffer) and impala (Aepyceros melampus) in the Kruger National Park, South Africa

    Epidemiology and Infection 124:591–598.

    https://doi.org/10.1017/s0950268899004008

    • PubMed
    • Google Scholar
12. 1. Blehert DS
    2. Hicks AC
    3. Behr M
    4. Meteyer CU
    5. Berlowski-Zier BM
    6. Buckles EL
    7. Coleman JTH
    8. Darling SR
    9. Gargas A
    10. Niver R
    11. Okoniewski JC
    12. Rudd RJ
    13. Stone WB

    (2009) Bat white-nose syndrome: an emerging fungal pathogen?

    Science 323:227.

    https://doi.org/10.1126/science.1163874

    • PubMed
    • Google Scholar
13. 1. Bonds MH
    2. Keenan DC
    3. Leidner AJ
    4. Rohani P

    (2005) Higher disease prevalence can induce greater sociality: a game theoretic coevolutionary model

    Evolution; International Journal of Organic Evolution 59:1859–1866.

    https://doi.org/10.1111/j.0014-3820.2005.tb01056.x

    • PubMed
    • Google Scholar
14. 1. Cantor M
    2. Maldonado-Chaparro AA
    3. Beck KB
    4. Brandl HB
    5. Carter GG
    6. He P
    7. Hillemann F
    8. Klarevas-Irby JA
    9. Ogino M
    10. Papageorgiou D
    11. Prox L
    12. Farine DR

    (2021) The importance of individual-to-society feedbacks in animal ecology and evolution

    The Journal of Animal Ecology 90:27–44.

    https://doi.org/10.1111/1365-2656.13336

    • PubMed
    • Google Scholar
15. 1. Carlson CJ
    2. Albery GF
    3. Merow C
    4. Trisos CH
    5. Zipfel CM
    6. Eskew EA
    7. Olival KJ
    8. Ross N
    9. Bansal S

    (2022a) Climate change increases cross-species viral transmission risk

    Nature 607:555–562.

    https://doi.org/10.1038/s41586-022-04788-w

    • PubMed
    • Google Scholar
16. 1. Carlson CJ
    2. Gibb RJ
    3. Albery GF
    4. Brierley L
    5. Connor RP
    6. Dallas TA
    7. Eskew EA
    8. Fagre AC
    9. Farrell MJ
    10. Frank HK
    11. Muylaert RL
    12. Poisot T
    13. Rasmussen AL
    14. Ryan SJ
    15. Seifert SN

    (2022b) The global virome in one network (VIRION): an Atlas of vertebrate-virus associations

    mBio 13:e0298521.

    https://doi.org/10.1128/mbio.02985-21

    • PubMed
    • Google Scholar
17. 1. Carter AJ
    2. Heinsohn R
    3. Goldizen AW
    4. Biro PA

    (2012) Boldness, trappability and sampling bias in wild lizards

    Animal Behaviour 83:1051–1058.

    https://doi.org/10.1016/j.anbehav.2012.01.033

    • Google Scholar
18. 1. Chakarov N
    2. Linke B
    3. Boerner M
    4. Goesmann A
    5. Krüger O
    6. Hoffman JI

    (2015) Apparent vector-mediated parent-to-offspring transmission in an avian malaria-like parasite

    Molecular Ecology 24:1355–1363.

    https://doi.org/10.1111/mec.13115

    • PubMed
    • Google Scholar
19. 1. Chandler JC
    2. Bevins SN
    3. Ellis JW
    4. Linder TJ
    5. Tell RM
    6. Jenkins-Moore M
    7. Root JJ
    8. Lenoch JB
    9. Robbe-Austerman S
    10. DeLiberto TJ
    11. Gidlewski T
    12. Kim Torchetti M
    13. Shriner SA

    (2021) SARS-CoV-2 exposure in wild white-tailed deer (Odocoileus virginianus)

    PNAS 118:e2114828118.

    https://doi.org/10.1073/pnas.2114828118

    • PubMed
    • Google Scholar
20. 1. Dall SRX
    2. Giraldeau L-A
    3. Olsson O
    4. McNamara JM
    5. Stephens DW

    (2005) Information and its use by animals in evolutionary ecology

    Trends in Ecology & Evolution 20:187–193.

    https://doi.org/10.1016/j.tree.2005.01.010

    • PubMed
    • Google Scholar
21. 1. Danchin E
    2. Giraldeau LA
    3. Valone TJ
    4. Wagner RH

    (2004) Public information: from nosy neighbors to cultural evolution

    Science 305:487–491.

    https://doi.org/10.1126/science.1098254

    • PubMed
    • Google Scholar
22. 1. DeAngelis DL
    2. Diaz SG

    (2019) Decision-making in agent-based modeling: A current review and future prospectus

    Frontiers in Ecology and Evolution 6:237.

    https://doi.org/10.3389/fevo.2018.00237

    • Google Scholar
23. Book

    1. Eddelbuettel D

    (2013) Seamless R and C++ Integration with Rcpp

    New York, NY: Springer-Verlag.

    https://doi.org/10.1007/978-1-4614-6868-4

    • Google Scholar
24. 1. Ezenwa VO
    2. Jolles AE

    (2015) Epidemiology. Opposite effects of anthelmintic treatment on microbial infection at individual versus population scales

    Science 347:175–177.

    https://doi.org/10.1126/science.1261714

    • PubMed
    • Google Scholar
25. 1. Ezenwa VO
    2. Ghai RR
    3. McKay AF
    4. Williams AE

    (2016) Group living and pathogen infection revisited

    Current Opinion in Behavioral Sciences 12:66–72.

    https://doi.org/10.1016/j.cobeha.2016.09.006

    • Google Scholar
26. 1. Farine DR

    (2017) A guide to null models for animal social network analysis

    Methods in Ecology and Evolution 8:1309–1320.

    https://doi.org/10.1111/2041-210X.12772

    • PubMed
    • Google Scholar
27. 1. Fereidouni S
    2. Freimanis GL
    3. Orynbayev M
    4. Ribeca P
    5. Flannery J
    6. King DP
    7. Zuther S
    8. Beer M
    9. Höper D
    10. Kydyrmanov A
    11. Karamendin K
    12. Kock R

    (2019) Mass die-off of saiga antelopes, kazakhstan, 2015

    Emerging Infectious Diseases 25:1169–1176.

    https://doi.org/10.3201/eid2506.180990

    • Google Scholar
28. 1. Fey SB
    2. Siepielski AM
    3. Nusslé S
    4. Cervantes-Yoshida K
    5. Hwan JL
    6. Huber ER
    7. Fey MJ
    8. Catenazzi A
    9. Carlson SM

    (2015) Recent shifts in the occurrence, cause, and magnitude of animal mass mortality events

    PNAS 112:1083–1088.

    https://doi.org/10.1073/pnas.1414894112

    • PubMed
    • Google Scholar
29. 1. Fieberg J
    2. Signer J
    3. Smith B
    4. Avgar T

    (2021) A “How to” guide for interpreting parameters in habitat-selection analyses

    The Journal of Animal Ecology 90:1027–1043.

    https://doi.org/10.1111/1365-2656.13441

    • PubMed
    • Google Scholar
30. 1. Fortin D
    2. Beyer HL
    3. Boyce MS
    4. Smith DW
    5. Duchesne T
    6. Mao JS

    (2005) Wolves influence elk movements: Behavior shapes a trophic cascade in yellowstone national park

    Ecology 86:1320–1330.

    https://doi.org/10.1890/04-0953

    • Google Scholar
31. 1. Gartland LA
    2. Firth JA
    3. Laskowski KL
    4. Jeanson R
    5. Ioannou CC

    (2022) Sociability as a personality trait in animals: methods, causes and consequences

    Biological Reviews of the Cambridge Philosophical Society 97:802–816.

    https://doi.org/10.1111/brv.12823

    • PubMed
    • Google Scholar
32. 1. Getz WM
    2. Salter R
    3. Lyons AJ
    4. Sippl-Swezey N

    (2015) Panmictic and clonal evolution on a single patchy resource produces polymorphic foraging guilds

    PLOS ONE 10:e0133732.

    https://doi.org/10.1371/journal.pone.0133732

    • PubMed
    • Google Scholar
33. 1. Getz WM
    2. Salter R
    3. Seidel DP
    4. van Hooft P

    (2016) Sympatric speciation in structureless environments

    BMC Evolutionary Biology 16:50.

    https://doi.org/10.1186/s12862-016-0617-0

    • PubMed
    • Google Scholar
34. 1. Gil MA
    2. Hein AM
    3. Spiegel O
    4. Baskett ML
    5. Sih A

    (2018) Social information links individual behavior to population and community dynamics

    Trends in Ecology & Evolution 33:535–548.

    https://doi.org/10.1016/j.tree.2018.04.010

    • PubMed
    • Google Scholar
35. 1. Global Consortium for H5N8 and Related Influenza Viruses

    (2016) Role for migratory wild birds in the global spread of avian influenza H5N8

    Science 354:213–217.

    https://doi.org/10.1126/science.aaf8852

    • PubMed
    • Google Scholar
36. 1. Goulson D
    2. Nicholls E
    3. Botías C
    4. Rotheray EL

    (2015) Bee declines driven by combined stress from parasites, pesticides, and lack of flowers

    Science 347:1255957.

    https://doi.org/10.1126/science.1255957

    • PubMed
    • Google Scholar
37. Software

    1. Gupte PR

    (2023) Pathomove: an individual-based model for the evolution of animal movement strategies under the risk of pathogen transmission, version v1.2.0

    Zenodo.

    https://doi.org/10.5281/zenodo.7789072
38. 1. Gupte PR
    2. Netz C
    3. Weissing FJ

    (2023) The joint evolution of animal movement and competition strategies

    The American Naturalist 202:E65–E82.

    https://doi.org/10.1086/725394

    • PubMed
    • Google Scholar
39. 1. Guttal V
    2. Couzin ID

    (2010) Social interactions, information use, and the evolution of collective migration

    PNAS 107:16172–16177.

    https://doi.org/10.1073/pnas.1006874107

    • PubMed
    • Google Scholar
40. 1. Hamblin S
    2. Hansen T

    (2013) On the practical usage of genetic algorithms in ecology and evolution

    Methods in Ecology and Evolution 4:184–194.

    https://doi.org/10.1111/2041-210X.12000

    • Google Scholar
41. 1. He P
    2. Montiglio PO
    3. Somveille M
    4. Cantor M
    5. Farine DR

    (2021) The role of habitat configuration in shaping animal population processes: A framework to generate quantitative predictions

    Oecologia 196:649–665.

    https://doi.org/10.1007/s00442-021-04967-y

    • PubMed
    • Google Scholar
42. Book

    1. Hofbauer J
    2. Sigmund K

    (1988)

    The Theory of Evolution and Dynamical Systems

    Cambridge University Press.

    • Google Scholar
43. 1. Hutchings MR
    2. Kyriazakis I
    3. Papachristou TG
    4. Gordon IJ
    5. Jackson F

    (2000) The herbivores’ dilemma: trade-offs between nutrition and parasitism in foraging decisions

    Oecologia 124:242–251.

    https://doi.org/10.1007/s004420000367

    • PubMed
    • Google Scholar
44. 1. Jesmer BR
    2. Merkle JA
    3. Goheen JR
    4. Aikens EO
    5. Beck JL
    6. Courtemanch AB
    7. Hurley MA
    8. McWhirter DE
    9. Miyasaki HM
    10. Monteith KL
    11. Kauffman MJ

    (2018) Is ungulate migration culturally transmitted? Evidence of social learning from translocated animals

    Science 361:1023–1025.

    https://doi.org/10.1126/science.aat0985

    • PubMed
    • Google Scholar
45. 1. Jolles A
    2. Gorsich E
    3. Gubbins S
    4. Beechler B
    5. Buss P
    6. Juleff N
    7. de Klerk-Lorist LM
    8. Maree F
    9. Perez-Martin E
    10. van Schalkwyk OL
    11. Scott K
    12. Zhang F
    13. Medlock J
    14. Charleston B

    (2021) Endemic persistence of a highly contagious pathogen: Foot-and-mouth disease in its wildlife host

    Science 374:104–109.

    https://doi.org/10.1126/science.abd2475

    • PubMed
    • Google Scholar
46. 1. Keeling MJ
    2. Woolhouse MEJ
    3. Shaw DJ
    4. Matthews L
    5. Chase-Topping M
    6. Haydon DT
    7. Cornell SJ
    8. Kappey J
    9. Wilesmith J
    10. Grenfell BT

    (2001) Dynamics of the 2001 UK foot and mouth epidemic: stochastic dispersal in a heterogeneous landscape

    Science 294:813–817.

    https://doi.org/10.1126/science.1065973

    • PubMed
    • Google Scholar
47. 1. Klump BC
    2. Martin JM
    3. Wild S
    4. Hörsch JK
    5. Major RE
    6. Aplin LM

    (2021) Innovation and geographic spread of a complex foraging culture in an urban parrot

    Science 373:456–460.

    https://doi.org/10.1126/science.abe7808

    • PubMed
    • Google Scholar
48. 1. Kuchipudi SV
    2. Surendran-Nair M
    3. Ruden RM
    4. Yon M
    5. Nissly RH
    6. Vandegrift KJ
    7. Nelli RK
    8. Li L
    9. Jayarao BM
    10. Maranas CD
    11. Levine N
    12. Willgert K
    13. Conlan AJK
    14. Olsen RJ
    15. Davis JJ
    16. Musser JM
    17. Hudson PJ
    18. Kapur V

    (2022) Multiple spillovers from humans and onward transmission of SARS-CoV-2 in white-tailed deer

    PNAS 119:e2121644119.

    https://doi.org/10.1073/pnas.2121644119

    • PubMed
    • Google Scholar
49. 1. Kurvers R
    2. Krause J
    3. Croft DP
    4. Wilson ADM
    5. Wolf M

    (2014) The evolutionary and ecological consequences of animal social networks: emerging issues

    Trends in Ecology & Evolution 29:326–335.

    https://doi.org/10.1016/j.tree.2014.04.002

    • PubMed
    • Google Scholar
50. 1. Levi T
    2. Kilpatrick AM
    3. Mangel M
    4. Wilmers CC

    (2012) Deer, predators, and the emergence of Lyme disease

    PNAS 109:10942–10947.

    https://doi.org/10.1073/pnas.1204536109

    • PubMed
    • Google Scholar
51. 1. Lion S
    2. Boots M

    (2010) Are parasites "prudent" in space?

    Ecology Letters 13:1245–1255.

    https://doi.org/10.1111/j.1461-0248.2010.01516.x

    • PubMed
    • Google Scholar
52. 1. Lloyd-Smith JO
    2. Schreiber SJ
    3. Kopp PE
    4. Getz WM

    (2005) Superspreading and the effect of individual variation on disease emergence

    Nature 438:355–359.

    https://doi.org/10.1038/nature04153

    • PubMed
    • Google Scholar
53. 1. Lunn TJ
    2. Peel AJ
    3. McCallum H

    (2021) Spatial dynamics of pathogen transmission in communally roosting species

    Journal of Animal Ecology 90:2609–2622.

    https://doi.org/10.1111/1365-2656.13566

    • Google Scholar
54. 1. Manlove K
    2. Wilber M
    3. White L
    4. Bastille-Rousseau G
    5. Yang A
    6. Gilbertson MLJ
    7. Craft ME
    8. Cross PC
    9. Wittemyer G
    10. Pepin KM

    (2022) Defining an epidemiological landscape that connects movement ecology to pathogen transmission and pace-of-life

    Ecology Letters 25:1760–1782.

    https://doi.org/10.1111/ele.14032

    • PubMed
    • Google Scholar
55. 1. Monk JD
    2. Smith JA
    3. Donadío E
    4. Perrig PL
    5. Crego RD
    6. Fileni M
    7. Bidder O
    8. Lambertucci SA
    9. Pauli JN
    10. Schmitz OJ
    11. Middleton AD

    (2022) Cascading effects of a disease outbreak in a remote protected area

    Ecology Letters 25:1152–1163.

    https://doi.org/10.1111/ele.13983

    • PubMed
    • Google Scholar
56. 1. Nathan R
    2. Getz WM
    3. Revilla E
    4. Holyoak M
    5. Kadmon R
    6. Saltz D
    7. Smouse PE

    (2008) A movement ecology paradigm for unifying organismal movement research

    PNAS 105:19052–19059.

    https://doi.org/10.1073/pnas.0800375105

    • PubMed
    • Google Scholar
57. 1. Nathan R
    2. Monk CT
    3. Arlinghaus R
    4. Adam T
    5. Alós J
    6. Assaf M
    7. Baktoft H
    8. Beardsworth CE
    9. Bertram MG
    10. Bijleveld AI
    11. Brodin T
    12. Brooks JL
    13. Campos-Candela A
    14. Cooke SJ
    15. Gjelland KØ
    16. Gupte PR
    17. Harel R
    18. Hellström G
    19. Jeltsch F
    20. Killen SS
    21. Klefoth T
    22. Langrock R
    23. Lennox RJ
    24. Lourie E
    25. Madden JR
    26. Orchan Y
    27. Pauwels IS
    28. Říha M
    29. Roeleke M
    30. Schlägel UE
    31. Shohami D
    32. Signer J
    33. Toledo S
    34. Vilk O
    35. Westrelin S
    36. Whiteside MA
    37. Jarić I

    (2022) Big-data approaches lead to an increased understanding of the ecology of animal movement

    Science 375:eabg1780.

    https://doi.org/10.1126/science.abg1780

    • PubMed
    • Google Scholar
58. 1. Netz CFG
    2. Hildenbrandt H
    3. Weissing FJ

    (2022) Complex eco-evolutionary dynamics induced by the coevolution of predator–prey movement strategies

    Evolutionary Ecology 36:1–17.

    https://doi.org/10.1007/s10682-021-10140-x

    • Google Scholar
59. 1. Poulin R
    2. Filion A

    (2021) Evolution of social behaviour in an infectious world: comparative analysis of social network structure versus parasite richness

    Behavioral Ecology and Sociobiology 75:105.

    https://doi.org/10.1007/s00265-021-03039-8

    • Google Scholar
60. 1. Prado F
    2. Sheih A
    3. West JD
    4. Kerr B

    (2009) Coevolutionary cycling of host sociality and pathogen virulence in contact networks

    Journal of Theoretical Biology 261:561–569.

    https://doi.org/10.1016/j.jtbi.2009.08.022

    • PubMed
    • Google Scholar
61. 1. Pusceddu M
    2. Cini A
    3. Alberti S
    4. Salaris E
    5. Theodorou P
    6. Floris I
    7. Satta A

    (2021) Honey bees increase social distancing when facing the ectoparasite Varroa destructor

    Science Advances 7:eabj1398.

    https://doi.org/10.1126/sciadv.abj1398

    • PubMed
    • Google Scholar
62. 1. Ratz T
    2. Monteith KM
    3. Vale PF
    4. Smiseth PT

    (2021) Carry on caring: infected females maintain their parental care despite high mortality

    Behavioral Ecology 32:738–746.

    https://doi.org/10.1093/beheco/arab028

    • PubMed
    • Google Scholar
63. Software

    1. R Development Core Team

    (2020) R: A language and environment for statistical computing

    R Foundation for Statistical Computing, Vienna, Austria.

    https://www.r-project.org/index.html
64. 1. Romano V
    2. MacIntosh AJJ
    3. Sueur C

    (2020) Stemming the flow: Information, infection, and social evolution

    Trends in Ecology & Evolution 35:849–853.

    https://doi.org/10.1016/j.tree.2020.07.004

    • PubMed
    • Google Scholar
65. 1. Romano V
    2. Sueur C
    3. MacIntosh AJJ

    (2022) The tradeoff between information and pathogen transmission in animal societies

    Oikos 2022:.

    https://doi.org/10.1111/oik.08290

    • Google Scholar
66. 1. Ruxton GD
    2. Gurney WSC
    3. de Roos AM

    (1992) Interference and generation cycles

    Theoretical Population Biology 42:235–253.

    https://doi.org/10.1016/0040-5809(92)90014-K

    • Google Scholar
67. 1. Sah P
    2. Mann J
    3. Bansal S

    (2018) Disease implications of animal social network structure: A synthesis across social systems

    The Journal of Animal Ecology 87:546–558.

    https://doi.org/10.1111/1365-2656.12786

    • PubMed
    • Google Scholar
68. 1. Sánchez CA
    2. Li H
    3. Phelps KL
    4. Zambrana-Torrelio C
    5. Wang L-F
    6. Zhou P
    7. Shi Z-L
    8. Olival KJ
    9. Daszak P

    (2022) A strategy to assess spillover risk of bat SARS-related coronaviruses in Southeast Asia

    Nature Communications 13:4380.

    https://doi.org/10.1038/s41467-022-31860-w

    • PubMed
    • Google Scholar
69. 1. Sanderson CE
    2. Alexander KA

    (2020) Unchartered waters: Climate change likely to intensify infectious disease outbreaks causing mass mortality events in marine mammals

    Global Change Biology 26:4284–4301.

    https://doi.org/10.1111/gcb.15163

    • PubMed
    • Google Scholar
70. 1. Scheele BC
    2. Pasmans F
    3. Skerratt LF
    4. Berger L
    5. Martel A
    6. Beukema W
    7. Acevedo AA
    8. Burrowes PA
    9. Carvalho T
    10. Catenazzi A
    11. De la Riva I
    12. Fisher MC
    13. Flechas SV
    14. Foster CN
    15. Frías-Álvarez P
    16. Garner TWJ
    17. Gratwicke B
    18. Guayasamin JM
    19. Hirschfeld M
    20. Kolby JE
    21. Kosch TA
    22. La Marca E
    23. Lindenmayer DB
    24. Lips KR
    25. Longo AV
    26. Maneyro R
    27. McDonald CA
    28. Mendelson J
    29. Palacios-Rodriguez P
    30. Parra-Olea G
    31. Richards-Zawacki CL
    32. Rödel M-O
    33. Rovito SM
    34. Soto-Azat C
    35. Toledo LF
    36. Voyles J
    37. Weldon C
    38. Whitfield SM
    39. Wilkinson M
    40. Zamudio KR
    41. Canessa S

    (2020) Response to Comment on “Amphibian fungal panzootic causes catastrophic and ongoing loss of biodiversity.”

    Science 367:eaay2905.

    https://doi.org/10.1126/science.aay2905

    • PubMed
    • Google Scholar
71. 1. Scherer C
    2. Radchuk V
    3. Franz M
    4. Thulke H
    5. Lange M
    6. Grimm V
    7. Kramer‐Schadt S

    (2020) Moving infections: individual movement decisions drive disease persistence in spatially structured landscapes

    Oikos 129:651–667.

    https://doi.org/10.1111/oik.07002

    • Google Scholar
72. 1. Spiegel O
    2. Leu ST
    3. Bull CM
    4. Sih A

    (2017) What’s your move? Movement as a link between personality and spatial dynamics in animal populations

    Ecology Letters 20:3–18.

    https://doi.org/10.1111/ele.12708

    • PubMed
    • Google Scholar
73. 1. Stockmaier S
    2. Stroeymeyt N
    3. Shattuck EC
    4. Hawley DM
    5. Meyers LA
    6. Bolnick DI

    (2021) Infectious diseases and social distancing in nature

    Science 371:eabc8881.

    https://doi.org/10.1126/science.abc8881

    • PubMed
    • Google Scholar
74. 1. Stroeymeyt N
    2. Grasse AV
    3. Crespi A
    4. Mersch DP
    5. Cremer S
    6. Keller L

    (2018) Social network plasticity decreases disease transmission in a eusocial insect

    Science 362:941–945.

    https://doi.org/10.1126/science.aat4793

    • PubMed
    • Google Scholar
75. 1. Tanner CJ
    2. Jackson AL

    (2012) Social structure emerges via the interaction between local ecology and individual behaviour

    The Journal of Animal Ecology 81:260–267.

    https://doi.org/10.1111/j.1365-2656.2011.01879.x

    • PubMed
    • Google Scholar
76. 1. Townsend AK
    2. Hawley DM
    3. Stephenson JF
    4. Williams KEG

    (2020) Emerging infectious disease and the challenges of social distancing in human and non-human animals

    Proceedings. Biological Sciences 287:20201039.

    https://doi.org/10.1098/rspb.2020.1039

    • PubMed
    • Google Scholar
77. 1. Vosloo W
    2. Thompson PN
    3. Botha B
    4. Bengis RG
    5. Thomson GR

    (2009) Longitudinal study to investigate the role of impala (Aepyceros melampus) in foot-and-mouth disease maintenance in the Kruger National Park, South Africa

    Transboundary and Emerging Diseases 56:18–30.

    https://doi.org/10.1111/j.1865-1682.2008.01059.x

    • PubMed
    • Google Scholar
78. 1. Webber QMR
    2. Vander Wal E

    (2018) An evolutionary framework outlining the integration of individual social and spatial ecology

    The Journal of Animal Ecology 87:113–127.

    https://doi.org/10.1111/1365-2656.12773

    • PubMed
    • Google Scholar
79. 1. Webber QMR
    2. Albery GF
    3. Farine DR
    4. Pinter-Wollman N
    5. Sharma N
    6. Spiegel O
    7. Vander Wal E
    8. Manlove K

    (2023) Behavioural ecology at the spatial-social interface

    Biological Reviews of the Cambridge Philosophical Society 98:868–886.

    https://doi.org/10.1111/brv.12934

    • PubMed
    • Google Scholar
80. 1. Weinstein SB
    2. Buck JC
    3. Young HS

    (2018) A landscape of disgust

    Science 359:1213–1214.

    https://doi.org/10.1126/science.aas8694

    • PubMed
    • Google Scholar
81. 1. White LA
    2. Forester JD
    3. Craft ME

    (2017) Using contact networks to explore mechanisms of parasite transmission in wildlife

    Biological Reviews of the Cambridge Philosophical Society 92:389–409.

    https://doi.org/10.1111/brv.12236

    • PubMed
    • Google Scholar
82. 1. White LA
    2. Forester JD
    3. Craft ME

    (2018a) Disease outbreak thresholds emerge from interactions between movement behavior, landscape structure, and epidemiology

    PNAS 115:7374–7379.

    https://doi.org/10.1073/pnas.1801383115

    • PubMed
    • Google Scholar
83. 1. White LA
    2. Forester JD
    3. Craft ME

    (2018b) Dynamic, spatial models of parasite transmission in wildlife: Their structure, applications and remaining challenges

    The Journal of Animal Ecology 87:559–580.

    https://doi.org/10.1111/1365-2656.12761

    • PubMed
    • Google Scholar
84. 1. White LA
    2. Forester JD
    3. Craft ME

    (2018c) Covariation between the physiological and behavioral components of pathogen transmission: host heterogeneity determines epidemic outcomes

    Oikos 127:538–552.

    https://doi.org/10.1111/oik.04527

    • Google Scholar
85. Book

    1. Whitehead H

    (2008) Analyzing Animal Societies: Quantitative Methods for Vertebrate Social Analysis

    University of Chicago Press.

    https://doi.org/10.7208/chicago/9780226895246.001.0001

    • Google Scholar
86. 1. Wilber MQ
    2. Yang A
    3. Boughton R
    4. Manlove KR
    5. Miller RS
    6. Pepin KM
    7. Wittemyer G

    (2022) A model for leveraging animal movement to understand spatio-temporal disease dynamics

    Ecology Letters 25:1290–1304.

    https://doi.org/10.1111/ele.13986

    • PubMed
    • Google Scholar
87. 1. Wille M
    2. Barr IG

    (2022) Resurgence of avian influenza virus

    Science 376:59–60.

    https://doi.org/10.1126/science.abo1232

    • PubMed
    • Google Scholar
88. 1. Wolf M
    2. van Doorn GS
    3. Weissing FJ

    (2008) Evolutionary emergence of responsive and unresponsive personalities

    PNAS 105:15825–15830.

    https://doi.org/10.1073/pnas.0805473105

    • PubMed
    • Google Scholar
89. 1. Wolf M
    2. Weissing FJ

    (2012) Animal personalities: Consequences for ecology and evolution

    Trends in Ecology & Evolution 27:452461.

    https://doi.org/10.1016/j.tree.2012.05.001

    • PubMed
    • Google Scholar

Author details

1. Pratik Rajan Gupte

   Groningen Institute for Evolutionary Life Sciences, University of Groningen, Groningen, Netherlands

   Present address

   Centre for Mathematical Modelling of Infectious Diseases, London School of Hygiene and Tropical Medicine, London, United Kingdom

   Contribution

   Conceptualization, Data curation, Software, Formal analysis, Investigation, Visualization, Methodology, Writing – original draft, Writing – review and editing

   For correspondence

   pratikgupte16@gmail.com

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-5294-7819

2. Gregory F Albery

   1. Georgetown University, Washington, United States
   2. Wissenschaftskolleg zu Berlin, Berlin, Germany

   Contribution

   Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

3. Jakob Gismann

   Groningen Institute for Evolutionary Life Sciences, University of Groningen, Groningen, Netherlands

   Contribution

   Conceptualization, Investigation, Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-2570-590X

4. Amy Sweeny

   Institute of Evolutionary Biology, University of Edinburgh, Edinburgh, United Kingdom

   Contribution

   Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

5. Franz J Weissing

   Groningen Institute for Evolutionary Life Sciences, University of Groningen, Groningen, Netherlands

   Contribution

   Conceptualization, Resources, Supervision, Funding acquisition, Investigation, Methodology, Writing – review and editing

   For correspondence

   f.j.weissing@rug.nl

   Competing interests

   No competing interests declared

• 495

  views

• 92

  downloads

• 2

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.