Nelson, B.S., G.J.L. Beckers, and R.A. Suthers. 2005. Vocal tract filtering and sound radiation in a songbird. The Journal of Experimental Biology. 208:297-308.

Researchers studied the impact of beak gape on vocalizations in the eastern towhee exploring two possible scenarios. The first is based upon the fact that trachea length determines resonant frequencies in vocalizations; thus, it is predicted that beak gape is used to alter the effective vocal tract length in order to emphasize fundamental frequencies over higher frequency harmonics. The other theory is that beak gape alters the amplitude of sounds above 4 kHz through filtering.

Researchers removed the vocal syringes of four towhees and attached small speakers to the trachea of these birds in order to study how sound leaving the birds’ bodies changed with modification of the physical orientation of the head, body, and beak gape of mounted birds within a chamber.

The researchers determined that the resonant frequencies of the vocal tracts appeared to amplify sound at frequencies of 2 and 5.5 kHz which are near the resonant frequencies of 1.88 and 5.67 kHz that would be predicted for the average towhee trachea of length ~45 mm diameter of ~1 mm. The researchers found that changes in beak gape did not change the resonant frequencies of the vocal tract but instead modulated the amplitude of sounds between 4 and 7.5 kHz (and to a lesser degree between 7.5 and 10 kHz) while not substantially changing lower frequencies. The researchers also found that at 6.5 kHz the amplitude of sound produced (dB) increased linearly with increasing beak gape (mm). The researchers found that directionality increased as both vocalization frequency and beak gape increased. Researchers also used audio and video recordings of Florida and Indiana towhees to further explore the impact of beak gape on sound production, finding that smaller gapes were associated with frequencies of 3.5-4 kHz and that larger gapes were associated with production of frequencies greater than 4 kHz.

The researchers theorized that the resonances of 2 and 5.5 kHz produced by towhee tracheas are designed to increase the distance over which vocalizations can be detected by concentrating the sound in these two ‘ frequency channels’. The researchers believe that the towhees decrease beak gape in order to filter out the resonance peak at 5.5 kHz as a way to modulate song. Reasons for this might include increasing vocalization control by removing higher frequency harmonics that could potentially degrade the complex temporal pattern of vocalization.

This study is similar to work discussed in class where instrumentation was used to detect independence of sound production from each of the two syrinxes of songbirds by placing a sensor to measure activity on each one. This experiment is a continuation of the questions surrounding how birds produce such complicated sounds, with this study moving beyond the syrinx to filtering further up the vocal tract. This also relates to the fact discussed in class that in humans many sounds (including vowels in English) are produced through filtering after generation of tones by the larynx. As we have discussed with examples in class (as well as with our continuing work with Raven) these researchers use software to conduct Fourier transforms ultimately creating spectrograms that can then be analyzed, in this case to determine how frequency sweeps generated by a speaker at the base of a trachea resonate and are filtered by a bird’s vocal tract.

This article addresses a number of issues discussed in class, including the issue of bandwidth and tone damping in signal production. The researchers believe that the trachea creates two concentrated ‘frequency channels’ with relative amplitudes modulated by beak gape. It appears that these eastern towhees are able to overcome the problem of low damping of a narrow bandwidth sound through the use of filtering of harmonic frequencies and are able to preserving temporal patterns in vocalizations with narrow bandwidths. Thus, this article also addresses the issue of pattern loss from reverberation which can be a serious problem for a time-based signal and which the researchers believe is important issue for the eastern towhee. Whereas the examples of reverberation that we discussed in class and the book appear to come from environmental factors the reverberations here are generated by the bird’s own anatomy and therefore can be dealt with by the use of filtering.