Collaborative Research: Developmental and Receiver-Dependent Costs of Avian Signals

Duke University, Durham NC

Investigators

Abstract

Collaborative research: Developmental and receiver-dependent costs of avian signals. Stephen Nowicki & William A. Searcy A central question in the study of animal communication is why signals provide reliable information, especially in cases where the signaler has different interests from the receiver. Theory suggests that signals may be reliable if they are in some way costly. The proposed research applies this theory to a well studied signaling system, birdsong. Female songbirds use male song to judge the quality of prospective mates. The "nutritional stress hypothesis" proposes that the costs of song development produce a reliable link between a male's song and his quality as a mate. Because song development is costly, any male experiencing an early stress such as food shortage will produce both poor song and a poor overall phenotype, making song quality a reliable measure of overall quality. Previous work has shown that early nutritional stress has a negative impact on the development of the brain areas that control song, on the ability of males to learn songs, and on other aspects of adult phenotype such as body size. The proposed research will extend this work in three ways. First, it will test whether early nutritional stress affects a male's vocal performance, that is his ability to sing songs that are physically challenging to produce. Second, the research will test whether early stress affects the quality of the immune system of adult birds, a trait that should be of special interest to females. Third, the research will ask whether early stress also affects the development of song preferences in females. Song also serves as an aggressive signal between competing males. In this context, signal reliability may be maintained by costs imposed by the aggressive response of receivers. Signals can reliably predict escalation if those signals also elicit the most aggressive response by receivers. Previous work shows that "soft songs" (songs produced very quietly) reliably predict aggression on the part of the signaler. The proposed research will test whether soft song also is likely to provoke an aggressive response from receivers, as theory suggests. Second, research will test whether soft song reduces the ability of third party individuals to gain information from an interaction. Third, the proposed research will investigate how patterns of signal matching contribute to signaling in aggressive interactions. The proposed research integrates theory and data concerning animal signaling systems and mate choice with testable hypotheses about the development of brain and behavior. This area has proven to be an effective platform for recruiting and training women and minority scientists, and for disseminating the process of how science is done to the general public. An interactive website will be developed to allow students in grades 6-12 to simulate playback and learning experiments with songbirds.