If you play the audio file below, you will hear the extraordinary song of a skylark – eulogised in poetry and a quintessential sound of spring and summer across the northern hemisphere. Even the ghastliness of the first world war failed to silence the larks flying high above the trenches. Writing shortly before his death at the battle of the Somme, John William Streets pays homage to their “lyrics wild and free”.


But what is the message of their song? Why do these birds soar from the ground and sing all spring and summer long? This behaviour is male-specific and coincides with the establishment of a territory, the attraction of a mate and the maintenance of both to provide a rearing ground for their young. Evolutionary and behavioural biologists have long believed that the songs are a way of exhibiting fitness to females, and a show of strength to nearby males that might otherwise encroach. Putting this theory to the test has been a challenge, however one that is successfully tackled in a recent BMC Biology study from Thierry Aubin and Nicole Geberzahn from the Centre for Neurosciences Paris-Sud, France, which finds a neat new way of rating the performance difficulty of complex birdsong.

Skylark. Image source: Flick, Nottsexminer

Previous studies assessing whether birdsong can function effectively as an honest signal of competitive quality have focused on simple trilled song, for which there is an observed trade-off between the frequency shift between the two notes of the trill, and how quickly it is repeated. This performance limit is thought to be primarily due to the need for the vocal tract to adjust to each new fundamental frequency produced in the syrinx, in order to suppress energy in higher harmonics and emit a pure note. Such tuning of the vocal tract is constrained by the speed of the motor systems involved and could thus reflect the overall motor skills of the performer.

The hypothesis that trilling prowess can serve as a meaningful competitive signal is supported by studies on species with trilled song structure. Using a single metric, termed vocal deviation, that measures the distance from the performance limit established for the species as a whole, it has been found that for individual birds, trill rate with respect to the frequency bandwidth comes closer to being the best possible in the context of attracting females or competing with other males, while assessing the personal best of many different individuals has documented a relationship between high vocal performance and male quality, condition, age or aggressive motivation.

Birds who simply trill are in a small minority however and as Goncalo Cardoso from the University of Porto, Portugal, emphasises in a commentary linked to the new research, it is a significant achievement to have found a comparably simple way of assessing complex birdsong. Scoring for performance across all the varied acoustic features of larksong (which can use any of about 300 different syllables in varied arrangements) had always seemed a daunting if not impossible task. Furthermore, any success in doing so would be unlikely to be applicable to other species. Yet Geberzahn and Aubin establish a simple and widely applicable metric that they term vocal gap deviation, which like vocal deviation, measures distance from a performance limit, and which narrows when birds are singing in a competitive context.

Spectrograms of syllables produced by a male skylark in response to a territorial playback.Red stars – peak frequency of the end of the first syllable and the start of the subsequent syllable. Horizontal bar indicates gap duration. Image source: Geberzahn and Aubin BMC Biology 2014 12:58

The authors achieved this by focusing not on the baroque complexity of the audible song, but on quantitative features of the silent gaps between syllables. As shown in this BBC video, although the in-flight song of skylarks sounds unbroken to our ears, when an acoustic recording is replayed more slowly, the gaps between syllables become apparent. Geberzahn and Aubin recorded a large number of individual larks singing in fields outside Paris, and for each of these recordings measured the frequency of the note just before the gap, and that at the start of the new syllable, plus the time interval of each gap, repeating this for all the gaps in their recordings. They found that the shortest gaps were seen for smaller frequency shifts, and established a performance limit that reflected this trade-off. The distance from this performance limit for an individual song trace, gives a measure of performance that they have named ‘vocal gap deviation’ in recognition of the similarities with the more narrowly applicable vocal deviation metric for trilled birdsong. As for vocal deviation, the new metric gives a measure of how close the song is to the limit imposed by the physical constraints of adjusting to sing a different note.

A significant advantage of the new metric is that it is applicable, in principle to all complex birdsong. It does not, of course, capture the beauty and complexity of the song itself, and the females for whom these songs are sung are likely to be rating song quality in a more comprehensive way. Nevertheless, this neat new metric will be a boon to birdsong biologists. Geberzahn and Aubin’s data already shows that measuring performance by this metric supports the hypothesis of a higher vocal performance in response to the presence of other males, and it will now be possible to explore further, in many different bird species, the competitive signals conveyed through song.


Written by Penelope Austin, Associate Editor for BMC Biology.


Research article

Open Access


Open Access

Related posts