Perception of geographical variation in song by male Puget Sound white-crowned sparrows, Zonotrichia leucophrys pugetensis

Abstract

The songs of male Puget Sound white-crowned sparrows form about 13 dialects along the Pacific Northwest coast of North America. In his original description of dialects in this subspecies, Baptista (1977, Condor, 79, 356–370) defined dialects by the terminal trill portion of the song because most males at a given location sing the same trill. Complex syllables in the introductory part of the song vary between males at a sample location and occur over a wider geographical range than trills. Baptista also recognized two superdialects or ‘themes’ based on differences in the phrase order within songs. We performed three playback experiments to territorial male white-crowned sparrows using both natural and computer-modified stimuli to test which acoustic features of song these birds attend to. Two of the experiments were conducted in nine dialect areas spanning nearly the entire geographical range of the subspecies. The results of the first such experiment indicate that across dialect areas, males give stronger responses (increased song rate and a shorter approach distance) to natural songs from the local dialect than they do to songs from a neighbouring dialect, a different theme, or another subspecies of white-crowned sparrow. Two experiments using modified stimuli conducted either across nine dialects or in one dialect led to the same conclusion: substitution of the trill from a neighbouring dialect into a song elicited weaker responses relative to the local control than did substitution of a neighbouring dialect's introduction. We conclude that the response of male Puget Sound white-crowned sparrows to playback is influenced most by variation in the song's trill. We found no evidence that the birds recognize ‘superdialects’ as do humans.