Experimental analysis of acoustic adaptation in bird song

The acoustic adaptation hypothesis (AAH) predicts that vocalizations intended for long range unambiguous communication should possess amplitude modulation (AM) characteristics such that the temporal patterning of amplitude degrades less than alternative patterns during transmission through native type habitat. The AAH predicts that closed habitat signals should be structured as low rate AM whistles, because such a structure is expected to best combat reverberation, the main form of degradation typical of closed habitats, which tends to blur the distinction between closely placed elements (i.e. rapid AM trills). The predicted signal structure for open habitat signals is rapid AM trills because such a structure is expected to best combat irregular amplitude fluctuations (IAFs), the main form of degradation typical of open habitats, which is expected to mask low rate AM signal patterns (i.e. whistles). Three experimental approaches were taken to investigate both the predictions of the AAH and their underlying rationale.;Results from a computer simulation, the transmission of synthetic signals through natural habitats, and the transmission of natural bird song through natural habitats, all clearly support the general predictions of the AAH. However, the results reveal that the rationale which underlies those predictions is incomplete because signal performance assessment must not only consider which structure minimizes average degradation levels incurred, but also a balance of 2 other factors: minimizing the variability in transmission quality, and maximizing degradation thresholds.;Results from the 3 experimental approaches revealed that in closed habitats, whistled signals incur less degradation on average and are not more variable in transmission quality than trilled signals, while also having a higher maximum degradation threshold. In open habitats, although trills incur more degradation, on average, than whistles, they clearly transmit with a greater consistency in quality compared to highly variable whistles. Degradation thresholds are not a factor in open habitats.;Therefore, in open and closed habitats, the only types of signals that degrade both increasingly and predictably with distance are trilled and whistled signals, respectively. These results bear significantly on the hypothesized song functions of ranging and the transmission of signaler identity.