While the processes that drive community assembly continue to be active areas of research in ecology and evolutionary biology, few investigations consider the extent to which the sensory ecology of organisms shape local communities. Previous studies suggest that the interplay between the sensory environment and interspecific competition can facilitate ‘sensory niche partitioning’ in which resources are differentially consumed based on sensory specializations. Here, we examine the interaction network of a community of acoustically specialized frog-biting midges (Corethrellaspp) and their sympatric frog hosts. Female frog-biting midges are micro-predators that find blood meals by eavesdropping on the calls of their frog victims. To investigate the underlying sensory filters within this network, we deployed 244 CDC suction traps in Gamboa, Panamá; each trap baited with a loudspeaker playing mating calls from one of 12 of the most abundant frogs in the community. Network interactions were determined by frog-biting midge species abundances in each trap. We used permutation tests and simulations to assess host specialization and phylogenetic dispersion, and bioacoustics analyses to determine potential drivers of sensory preferences.
Results/Conclusions
We captured over 6000 frog-biting midges from at least 12 morphospecies. Despite equal sampling efforts for the various frog calls, the midge species varied in total abundances and were disproportionately associated with certain frog calls. That is, interactions were influenced by underlying sensory biases. Bioacoustic analyses demonstrate that certain acoustic properties of frog calls, including call frequency, are correlated with host call preferences. The community was phylogenetically overdispersed and while frog-biting midges followed generalist strategies, they were less generalist than expected by chance. Overall, this study highlights how sensory filters can shape community structure and likely play a role modulating interspecific competition and community assembly.