No phenotypic signature of acoustic competition in songs of a tropical cricket assemblage

Efficient acoustic communication in multispecies assemblages is challenging due to the presence of heterospecific signals. Masking interference and signal confusion of similarly structured signals can impose fitness costs and, thus, drive evolutionary processes that shape acoustic signals to reduce their overlap in signal space. Although the partitioning of signal space has been frequently studied in frog and bird communities, this topic has received much less attention with reference to insects that communicate acoustically. In this study, we examined the role of acoustic competition in a tropical cricket community and tested the following hypotheses: 1) cosignaling species are expected to exhibit more dissimilar calling songs as compared with species that are spatially and/or temporally separated and 2) species signaling with similar song frequencies, who are thereby subject to energetic masking, are more likely to differ in the temporal domain of their calling songs. Using non-metric multidimensional scaling and pairwise comparison methods to measure acoustic dissimilarity, we found no evidence for either hypothesis: the acoustic signals of species pairs that cosignaled or used similar calling frequencies did not significantly differ from those of species that were spatially/temporally segregated or had large song frequency differences. In conclusion, for the acoustically communicating cricket community investigated, no supportive evidence for the partitioning hypothesis and the widespread belief that acoustic competition has led to divergent selection pressures on signal structure to avoid masking interference was found. Instead, we argue that selection pressures on sensory/neuronal mechanisms seem to more strongly drive reliable communication.