My recent research has focused on the evolution of signal reliability, using bird song as a model system. Signal reliability poses a problem whenever there are conflicts of interest between signalers and receivers. Female birds, for example, use song in assessing male quality during mate choice, and this imposes selection on males to exaggerate the song attributes of interest. Similarly, song is used in male-male competition to signal aggressive intentions and fighting ability, and in this context selection again may often favor signal exaggeration. If exaggeration is not curbed, the information content of the signal is lost, and along with it any reason for receiver response. Thus it becomes interesting to ask whether bird song contains reliable information on a male’s quality as a mate, on his aggressive intentions, and on his fighting ability, and if so to examine the mechanisms acting to counter exaggeration and ensure reliability.
We have tested the developmental stress hypothesis as a possible explanation for the reliability of song in male-female signaling. This hypothesis explains reliability as an outcome of the developmental costs of song. In songbirds, the brain structures that underlie song learning and production develop during a limited period after hatching, at a time when young birds are particularly subject to nutritional and other stresses. During the same period, birds undergo much of the growth and development that produce their overall phenotype. Only individuals that experience reduced stress, or that have genotypes particularly buffered against stress, are able to invest adequately in both song development and general phenotypic development. Learned features of song thus become reliably correlated with male phenotypic quality.
In one test of developmental stress hypothesis, we manipulated early nutrition in swamp sparrows, and found effects on the sizes of two of the song system nuclei in the brain, the HVc and RA (Nowicki et al. 2002a). Nutrition also affected the accuracy of song learning, as measured by cross-correlation scores between model and copied notes (Nowicki et al. 2002a). In a separate experiment with song sparrows, we showed that the accuracy with which song was learned affected female response to song, with higher response to well learned than to poorly learned songs (Nowicki et al. 2002b). In a more recent experiment, again with swamp sparrows, we found that adult females give stronger courtship responses to songs of males with good early growth rates than to songs of males with poor growth rates (Searcy, Peters, Kipper and Nowicki unpublished). Taken together these results support the hypothesis that female songbirds use properties of male song to judge the developmental quality of potential mates.
We have tested whether singing conveys aggressive intent in song sparrows and in swamp sparrows. In experiments with both species, we elicited aggressive signaling from a male subject using one minute of playback of song on his territory, and recorded his aggressive signaling behavior for five minutes. We then exposed a taxidermic mount of a conspecific coupled with a further two-minute playback and continued to record the subject’s behavior while giving him the opportunity to attack the mount. We measured a number of signaling behaviors that have hypothesized to be aggressive signals, including song type switching, song matching, overall song rate, and the production of low amplitude soft song. In song sparrows, none of the display measures from the initial five-minute recording period predicted attack. For the one minute preceding attack, soft song was the one display that differed significantly between attackers and non-attackers (Searcy et al. 2006). In swamp sparrows, production of soft song was again the best predictor of attack; in this case, production of soft song in both the initial recording period and the one-minute preceding attack differed significantly between attackers and non-attackers (Ballentine et al. 2008). Soft song may be a “conventional” signal, that is one whose honesty cannot be explained by developmental or production costs tied to its acoustic structure, but which instead is honest because of costs imposed by receiver response.
We have also been interested in whether male songbirds use performance signals to communicate fighting ability. In simple trilled songs, such as those sung by swamp sparrows, vocal performance can be measured by deviation from an upper bound regression of frequency bandwidth on trill rate. We found that male swamp sparrows increase this aspect of vocal performance when in aggressive contexts by increasing both trill rate and frequency bandwidth of individual song types (DuBois et al. 2009); this response supports the hypothesis that vocal performance is used as a signal in male-male interactions. In addition, we have tested male swamp sparrows for aggressive response towards songs of high and low vocal performance (DuBois et al. unpublished). When differences in vocal performance were on the level shown between males, territory owners responded more aggressively towards high vocal performance than towards low. When, however, differences in vocal performance were on the level shown within males when modulating vocal performance of particular songs, subjects showed no discrimination. These results are in accord with vocal performance being an index signal, with honesty maintained by physical constraints that prevent a functionally important level of exaggeration.
Recently we have initiated a new project to examine the interplay between cognition, fitness, and communication. Song learning is a complex cognitive task, requiring precise learning from external models and long term memory, and therefore is expected to be positively associated with other aspects of cognitive ability. We propose to assess the association of cognition and song by recording the song repertoires of male song sparrows in the field and then measuring their learning ability on foraging tasks in the laboratory. We also plan to measure associations between cognitive ability, inbreeding, and fitness measures in the field, again using song sparrows. This project is being done in collaboration with Steve Nowicki, Rindy Anderson, Neeltje Boogert, Susan Peters, Peter Arcese, and others.
Ballentine, B., W. A. Searcy, and S. Nowicki. 2008. Reliable aggressive signalling in swamp sparrows. Animal Behaviour. 75:693-703.
DuBois, A. L, S. Nowicki, and W. A. Searcy. 2009. Swamp sparrows modulate vocal performance in aggressive contexts. Biology Letters 5:163-165.
Nowicki, S., W. A. Searcy, and S. Peters. 2002a. Brain development, song learning, and mate choice in birds: a review and experimental test of the "nutritional stress hypothesis." J. Comp. Phys. A 188:1003-1014.
Nowicki, S., W. A. Searcy, and S. Peters. 2002. Quality of song learning affects female response to male bird song. Proceedings of the Royal Society (London) B 269:1949-
Searcy, W. A., R. C. Anderson, and S. Nowicki. 2006. Bird song as a signal of aggressive intent. Behavioral Ecology and Sociobiology 60:234-241.