Local preference encoded by complex signaling: mechanisms of mate preference in the red-eyed treefrog ( Agalychnis callidryas )

The Central American red-eyed treefrog ( Agalychnis callidryas ) is a nocturnal/crepuscular frog that exhibits high population-level phenotypic variation in a suite of traits generally important in anuran communication. Females of this species demonstrate a preference for local over non-local males, although the contribution of different communication modes to mate choice remains untested. We performed eight sets of experiments at La Selva Biological Station (Costa Rica) to test the relative roles of male advertisement call and flank-stripe pattern on female mate choice. For each trial, we used pre-recorded calls for acoustic signals and/or 3D-printed, hand-painted models for visual stimulus. Females selected conspecifics over heterospecifics when presented with either acoustic or visual stimuli only, indicating that both evolve as species-specific signals used for mate discrimination. Bayesian modeling showed that females chose local males over non-locals based on unimodal; responses to multimodal signals varied. Female mate choice in this species is likely nuanced and subject to the constraints of the local environment in which animals are signaling. Our data show that at least two traits that vary among populations are important for mate selection. Significance statement Evolution shapes animal communication in diverse ways to accommodate complicated signaling contexts, including environmental noise, the presence of eavesdroppers, and changing landscapes. Among species that use complex signals or signals in multiple modalities, understanding the role of signaling traits in reproductive behavior and their concomitant shifts between populations can shed light on lineage divergence, speciation, and the forces shaping communication. We used choice tests with Agalychnis callidryas to determine which traits lead to local-male mate preference. We found that both call and stripe are each sufficient to allow this nocturnal frog to choose mates in absence of the other stimulus. Multimodal signals were marginally better than call—but not color—alone. This may be a result of statistical noise or preliminary evidence of asymmetric use of traits across a complex landscape. These results suggest that call and color may encode both redundant information and multiple messages to receivers in this species: either is sufficient to elicit mate choice (redundant), but together, they enhance local mate preference (multiple messages). Further tests on how other communication modalities inform behavior will add to our understanding of complex signaling and lineage divergence among differentiated populations in this species.