Interspecific ecological interactions in nature build up from encounters between individual partners, including variable outcomes as the result of overlying multiple forms of interaction. For example, multiplex ecological networks of plant-animal interactions emerge from the juxtaposition of the different interaction types, e.g., mycorrhizal symbioses, herbivory on leaves, roots, flowers, and fruits, pollination and seed dispersal mutualisms, facilitation, competitive interactions for shared partners. In contrast to other multilayer networks, multiplexed networks link a given set of nodes (i.e., plants in a community) through a series of processes where multiple interactions operate (i.e., the sequential stages of reproduction including layers of pollination, herbivory, seed dispersal, and establishment). The result of multiplexed interactions is, ultimately, final effects on fitness of the interacting partners resulting from the coupled mutualistic/antagonistic forms of interaction that each plant species supports. Yet we are far from understanding how these complex networks are intertwined.
Results/Conclusions
Characterizing and sampling multiplexed networks means assessing the interaction modes that correspond to each interaction type. These can be thought as interaction vectors, each including the outcomes of interaction of each plant species with each partner species, with connections among distinct layers of interaction types in terms of their integration within the plant's life history. I discuss multiple challenges and potential avenues to explore those multi-faceted forms of interaction. Species-level multiplex networks are built on individual-based interaction forms for different processes (e.g., pollination, herbivory, seed dispersal). Some types of these individual-based interaction forms are interaction motifs, overrepresented in a given population relative to a random interaction pattern. Multispecies, multiplexed interaction networks are just the result of assembling the distinct interaction modes represented across individuals and species in a given community. Multiplexed networks thus allow biologically-realistic representation and analysis of the extraordinary diversity of interactions in real-world assemblages. I review case studies illustrating the use of multiplex networks to assess individual-based and multispecific networks of plant-animal mutualisms.