Guest post by: Eugene Morton, York University
Bird song is one of the most fascinating and complex examples of animal communication, and the quest to understand its evolution and function has fueled the careers of many behavioral ecologists, psychologists and neurophysiologists.
Recently, scientists in the Departments of Biology and Neurobiology at Duke University have made incredible advances in this field. In “Songbirds learn songs least degraded by environmental transmission,” Susan Peters, Elizabeth P. Derryberry, and Stephen Nowicki show how a youngster chooses which songs to learn from the huge number they may be exposed to during their learning period.
Their simple but elegant experiment offered the birds a choice to learn songs that contained echoes versus no echoes and the birds chose to learn only those with no echoes. The rejected songs had been transmitted and re-recorded through 25 meters of habitat, and picked up reverberations and a few other changes along the way, but they were equally loud to the learned versions.
This says a great deal about how birds put to use their extraordinary ability to hear small time differences. What’s so great about hearing echoes?
Compared to our ability, where we hear only echoes from distant large objects, birds can hear echoes from tree trunks and vegetation. They use this ability to learn songs that transmit with the least amount of echoes or, more generally, degradation.
In this way, the birds themselves reject songs less well suited to their environment; cultural selection. As the birds were housed together while learning the songs it is not surprising that they came up with two that were never presented to them; they must also have learned from each other.
Why is it important to understand the criteria birds use to choose songs to learn? I would answer because then we can understand how cultural and natural selection interact. Cultural selection favors birds that learn songs that will propagate for the greatest distance and remain undegraded.
These songs must function better than a random selection of songs would. The function must related to how the listeners of these songs are affected by them: are they more efficiently repelled if they are competitors and attracted if they are potential mates? Natural selection will favor the birds whose songs do this the best.
And it turns out that this interplay is helping birds cope with increasingly human-influenced environments. The traffic noise we generate can favor learning songs that are higher or lower in than the frequency of this noise. This ability is based upon the same cultural choice of songs described here for swamp sparrows.
It is hoped that this excellent study will stimulate others to assess the role of learning in adapting songs, not only to habitats, but to the social functions songs have. Songs function over distance and this study describes how song learning can strengthen this role and the importance of distance in song evolution.
Eugene S. Morton
Hemlock Hill Field Station, Pennsylvania
York University, Ontario, Canada