Field Experiments on the Perception of Song Types by Birds

Abstract

The songs of passerine birds are enormously variable, but usually the many variations on the species-specific song can be divided into a set, or repertoire, of discrete, stereotyped songs known as song types (Hartshorne 1973; Dobson and Lemon 1975). Song types have been shown to have various functions in territory defence and mate attraction (reviews in Krebs and Kroodsma 1980; Catchpole 1982; Searcy and Andersson 1988). If song types are to serve any function, birds must be able to tell them apart; in fact this may be the main reason why song types are so different from one another (Kroodsma 1982). However, only recently have researchers used playbacks to ask how different songs have to be for birds to treat them as different types (Horn and Falls 1988a; Falls et al. 1988; Weary et al. 1990). At first glance, one might think that such perceptual questions are best answered in the laboratory. However, this is a question as much about how birds evaluate and respond to differences between songs as about whether they are capable of sensing the differences. Birds might treat the same structural contrast between songs as negligible in one situation and crucial in another. Provided enough is known about natural singing behaviour, this possibility can be tested through playback experiments that mimic these different situations.