Jeffrey Cynx

Experimental determination of a unit of song production in the zebra finch (Taeniopygia guttata)

Zebra finch (Taeniopygia guttata) song is composed of syllables delivered in a set order. Little is known about the program that controls this temporal delivery. A decision to sing or not to sing may or may not affect the entire song. Song, once commenced, may continue or may halt. If song is halted, stops may occur only at certain points. Seven zebra finches were presented with short bursts of strobe light while engaged in song. The variables of interest were whether the birds stopped and where they stopped. The results can be summarized as follows: Ongoing zebra finch song can be interrupted, interruptions occur at discrete locations in song, and the locations almost always fall between song syllables. These results reveal a functional representation of song production and place constraints on possible neural mechanisms that underlie song production in zebra finches and probably other oscine species. The results also raise hypotheses about the elements of song perception and memory.

Timbre Discrimination in Zebra Finch (Taeniopygia guttata) Song Syllables

Zebra finch (Taeniopygia guttata) songs include syllables of a fundamental frequency and harmonics. Individual harmonics in 1 syllable can be more or less emphasized. The functional role of this variability is unknown. These experiments provide evidence of how the phenomenon is perceived. We trained 12 male and female zebra finches on a go-no-go operant procedure to discriminate between 2 song syllables that varied only in the absence of the 2nd or 5th harmonic. Training involved many thousands of trials. Both sexes used the presence or absence of the 2nd harmonic as the sole discriminative cue. Females had more difficulty learning to perform the task when the presence of the 2nd harmonic was the go stimulus, which indicates that their use of the information was biased by stimulus-response contingencies. The results are discussed in terms of a broad strategy to understand how animals perceive sounds used in communication.

Auditory frequency generalization and a failure to find octave generalization in a songbird, the European starling (Sturnus vulgaris).

Many species of songbirds have absolute pitch perception. They memorize tones when discriminating between sequences and fail to generalize to novel frequency ranges. I examined frequency generalization in starlings (Sturnus vulgaris) that were trained to discriminate between individual frequencies. Starlings were tested for frequency generalization after training on 2- and 3-tone discriminations. There was no evidence for octave generalization, which is a hallmark of human absolute pitch perception. This suggests that avian absolute pitch perception must not be interpreted as identical with that in humans. A control experiment with 1-tone discriminations indicated that the presence of lit response keys affects the shape of the generalization gradients. Lit response keys are a common feature in avian auditory perception experiments, and this control experiment cautions that results may be affected by this seemingly minor procedural change.