Interactions are a defining characteristic of every species' "milieu" since no individual organism exists without participating in some sort of ecologically relevant interaction during its lifetime. They also underpin many foundational concepts in ecology, such as keystone species, trophic cascades, the ecological niche, and community stability. Moreover, present-day ecosystems are also subjected to increasing disturbance, such as invasive species or the accumulation of persistent contaminants, the effect of which can readily permeate across entire communities because of the intricate way in which species are linked together.
Because of their central importance, ecology has increasingly adopted a "network" approach to characterizing and studying the structure and dynamics of ecological communities. To date, the bulk of efforts have focused on "uniplex" networks built from a single interaction type (e.g., predator-prey, plant-pollinator, or host-parasite) despite the fact that any given species' milieus of interactions is almost certainly a mixture of antagonistic, mutualistic, amensalistic, and commensalistic interactions. As a result, champions of the multiplex-network approach argue that network ecology must move up to the next step on the complexity ladder if it's to continue shedding novel insights onto questions of fundamental importance.
The primary goal of this talk is to pump the brakes on the multiplex bandwagon without completely pouring cold water on this emerging topic. In particular, I will highlight the extent to which too much of network ecology is still driven more by the pattern rather than the process. Interestingly, this occurs despite the fact that the relative usefulness of having both the network approach---be it uniplex or multiplex---in our collective toolbox depends much more on the questions being asked as opposed to the networks' inherent complexity. Throughout my talk, I will move through various examples of open questions at the uniplex scale that are best tackled without the added complications of shifting to a multiplex approach. To help make my case, I will rely on arguments in favor of uniplex studies that can be found across both the historic and recent literature. Finally, I will conclude by suggesting a few areas in which I suspect the multiplex approach will be particularly advantageous going forward.