The metabolic cost of birdsong production

Citation
K. Oberweger et F. Goller, The metabolic cost of birdsong production, J EXP BIOL, 204(19), 2001, pp. 3379-3388
Citations number
40
Categorie Soggetti
Biology,"Experimental Biology
Journal title
JOURNAL OF EXPERIMENTAL BIOLOGY
ISSN journal
00220949 → ACNP
Volume
204
Issue
19
Year of publication
2001
Pages
3379 - 3388
Database
ISI
SICI code
0022-0949(200110)204:19<3379:TMCOBP>2.0.ZU;2-I
Abstract
The metabolic cost of birdsong production has not been studied in detail bu t is of importance in our understanding of how selective pressures shape so ng behavior. We measured rates of oxygen consumption during song in three s ongbird species, zebra finches (Taeniopygia guttata), Waterslager canaries (Serinus canaria) and European starlings (Sturnus vulgaris). These species sing songs with different acoustic and temporal characteristics: short ster eotyped song (zebra finch), long song with high temporal complexity (canary ) and long song with high acoustic, but low temporal, complexity (starling) . In all three species, song slightly increased the rate of oxygen consumptio n over pre-song levels (1.02-1.36-fold). In zebra finches, the metabolic co st per song motif averaged 1.2 mul g(-1). This cost per motif did not chang e over the range of song duration measured for the four individuals. Surpri singly, the metabolic cost of song production in the species with the tempo rally most complex song, the canary, was no greater than in the other two s pecies. In starlings, a 16dB increase in sound intensity was accompanied by a 1.16-fold increase in the rate of oxygen consumption. These data indicat e that the metabolic cost of song production in the songbird species studie d is no higher than that for other types of vocal behavior in various bird groups. Our analysis shows that the metabolic cost of singing is also simil ar to that of calling in frogs and of human speech production. However, dif ficulties with measurements on freely behaving birds in a small respirometr y chamber limit the depth of analysis that is possible.