Introduction
Seed dispersal is a fundamental process for the survival, reproduction, and spread of plants because it is the only stage in the plant life cycle when many plants may use movement to colonize new habitats, escape competition, or evade attack by pathogens and herbivores (Howe & Smallwood 1982; Howe & Miriti 2000, 2004; Jordano 2000; Schupp et al. 2010; Hirsch et al. 2012; Carlo & Tewksbury 2014). Animal-mediated seed dispersal is among the most common modes of seed dispersal (Howe & Smallwood 1982; Jordano 2000; Herrera & Pellmyr 2002); 64% of gymnosperm and 27% of angiosperm plant species rely on vertebrates for dispersal (Herrera 1989). As a result, understanding patterns in animal-mediated seed dispersal (zoochory) may provide a means to predict spatial patterns in plant recruitment (Beckman & Rogers 2013), how plant species may respond to climate change (Dyer 1995; Cain et al. 2000; Kremer et al. 2012; Corlett & Westcott 2013), and the outcomes of exotic plant invasions (Traveset & Richardson 2014; Baltzinger et al. 2019). However, zoochory is notoriously variable in both the quantity of seeds dispersed and distance of seed movement (Schupp et al. 2010; Côrtes & Uriarte 2013; Rogers et al. 2019). Consequently, developing predictive frameworks that enable more effective classification and prediction of zoochory has been an ongoing challenge to ecologists for several decades (Schupp et al. 2010; Côrtes & Uriarte 2013; Zwolak 2018; Rogerset al. 2019; Zwolak & Sih 2020).
While studies have shown that animal behavior provides a useful means for predicting spatial patterns in plant recruitment (Wang & Smith 2002; Levey et al. 2005; Russo et al. 2006; Kremeret al. 2012; Beckman & Rogers 2013; Sasal & Morales 2013; Herrmann et al. 2016), relationships between specific dispersal agents and spatial patterns of dispersal often remain notably weak (Getzin et al. 2014; Schupp et al. 2017; but see Levey et al. 2005 for an exception) due to high intraspecific variability in the behavior of seed-dispersal agents (Zwolak 2018). Such high intraspecific variation in disperser efficacy may have important pragmatic ramifications because species-level averages of disperser efficacy (i.e., mean number of seeds dispersed by individuals of one species) are often used to predict how changes in animal populations and communities may lead to losses in seed dispersal(e.g., Peres et al. 2016; Culot et al. 2017). For example, realized seed dispersal in fragmented habitats will be much lower than what is predicted by a species-average approach if individual animals that provide disproportionately greater contributions to seed dispersal are also the individuals most likely to be lost or exhibit changes in behavior in fragmented habitats (McConkey & O’Farrill 2016; Zwolak 2018). As such, identifying traits of dispersal agents that explain intraspecific variation in behavior may provide the key to successfully predicting seed dispersal (Côrtes & Uriarte 2013; González-Varo & Traveset 2016; Schupp et al. 2017; Zwolak 2018; Brehm et al. 2019; Zwolak & Sih 2020). Specifically, identifying individual traits that may affect the quantity of seeds handled, the distance that seeds are carried, and the microhabitat in which seeds are dispersed is critical to predicting variation in the effectiveness of individual agents of seed dispersal (Schupp et al. 2010; Brehm et al.2019).
Social status, which may reflect an individual’s age, size, sex, or rank in a dominance hierarchy, may have an important bearing on individual behaviors that directly and indirectly affect seed dispersal by animals. The effectiveness of seed dispersal agents is described as the ‘number of new adults produced by the dispersal activities of a disperser’ (Schupp 1993). Seed dispersal effectiveness is the product of the quantity of seeds dispersed (quantitative component) and the probability of seedling establishment (quality component; Schupp et al. 2010). Since social status can affect individual diet composition, social status may predict intraspecific variation in the quantity of seeds dispersed (Fig. 1). Recent work by Tsuji et al. (2020) provides compelling evidence that social rank in Japanese macaques determines the quantity of seeds dispersed by an individual, and we suggest that this may be applicable for a broad diversity of taxa, including carnivores. We also highlight that social status affects individual space use, which may explain intraspecific variation in both the distance and quality of seed dispersal (Fig. 1). Much of the unresolved variation in seed dispersal could therefore be understood by a novel focus on animal social status.
We present a framework that provides important clarity on the relationships between individual social status, animal behavior, and seed dispersal, allowing for predictions of seed-dispersal patterns across a broad spectrum of systems. This framework may also be used to understand how animal social behavior affects our ability to mitigate several timely ecological challenges, such as biological invasions, plant persistence in urban environments, and plant persistence amidst climate change. We highlight that in many systems, subordinate individuals may be key agents of seed dispersal, illustrating that individuals with minimal demographic contributions may have substantial ecological contributions, i.e., subordinate animals may play a dominant role in seed dispersal. Hence, environmental changes that affect animal social structure may have unappreciated consequences for seed dispersal. This framework is informed by two perspectives, described below, and can be implemented across the wide diversity of plant-frugivore networks that include social frugivores, which may reveal emerging hotspots or catastrophic losses of seed dispersal in contexts where global change disrupts animal social structure. While our framework specifically focuses on zoochory, the general concepts and hypotheses that we present could be applied to other forms of animal-mediated dispersal (e.g., dispersal of fungi or parasites) that are mediated by individual diet and space use.