Composition and sequential organization of song repertoires in Cassin’s Vireo (Vireo cassinii)

Abstract

The rules governing bird song sequences vary considerably across the avian phylogeny, and modifications to these rules represent one of the many ways in which bird song varies interspecifically. Cassin's Vireo (Vireo cassinii) is one species that shows a highly structured syntax, with clearly non-random patterns of sequential organization in their songs. Here I present a description of Cassin's Vireo song sequences from the Sierra Nevada Mountains in California and employ network analysis to quantify transition patterns within the songs. Repertoire sizes varied between 44 and 60 phrase types per bird for the 13 birds analyzed here. The repertoire was subdivided into 'themes' containing between two and seven phrase types. The birds sang the phrase types in a given theme for a time before eventually introducing a new theme; in this manner the repertoire was revealed relatively slowly over time. Theme composition within a bird's repertoire did not change within or between singing bouts throughout the season. The tendency to sing in themes was corroborated by network analysis, which revealed small-world structure in the songs. Phrase types were widely shared within the population. I discuss these findings as they compare with the singing styles of other species, both closely and distantly related.

Complexity, Predictability and Time Homogeneity of Syntax in the Songs of Cassin’s Vireo (Vireo cassinii)

“…As the N-gram increased, the L 1 -distances for the first-order model diverged from the values for the training set (one-tailed paired t-tests vs training set, p = 0.28 at N = 2, p<0.001 at N = 3 to N = 7), while the values for the second-order model remained similar to those of the training set until N = 6, only showing a significant difference at N = 7 ( Fig 2 ; one-tailed paired t-tests vs training set, p = 0.27 at N = 2, p = 0.23 at N = 3, p = 0.54 at N = 4, p = 0.20 at N = 5, p = 0.081 at N = 6, p = 0.003 at N = 7). The first- and second-order models both produced a similar recurrence interval distribution to that described by Hedley [ 25 ], biased towards low recurrence intervals. Unlike the distributions derived from the zero-order model, these distributions did not differ significantly from the distributions observed in the training sets (one-tailed paired t-tests vs training set: zero-order, p<0.001; first-order p = 0.66; second-order p = 0.99), suggesting that even models that condition transition probabilities on a short preceding sequence can recover apparent longer-term structure in song sequences ( Fig 2 ).…”

“…Accordingly, each recording was analyzed as a single sequence, regardless of the durations of silence contained therein. A more detailed description of this species’ singing behavior can be found in Hedley [ 25 ].…”

“…Regardless of the exact level of complexity of the songs, it appears that the syntax is governed by relatively simple rules and that transition probabilities are conditioned on the most recent phrase or phrases. Patterns that manifest themselves over longer time-scales, such as the tendency towards low recurrence intervals [ 25 ], appear to be emergent properties of these rules.…”

“…Cassin’s Vireos sing elaborate sequences of phrases that, when recorded over long periods, show extensive evidence of repeated syntactic patterns [ 25 ]. The species also displays high levels of interannual site fidelity, non-overlapping territories, and prolific vocal output, making them an ideal species in which to study the sequential complexity of songs.…”

A robust automatic birdsong phrase classification: A template-based approach

“…For example, Cassin's Vireos (Vireo cassinii; CAVI) have songs made up of phrases, short bursts of notes, <0.7 s in duration, separated by 1 s of silence or more. Each bird will typically have 40-60 phrase types 6 that can be reliably distinguished manually by an observer; doing so automatically is challenging due to within-class variability, limited training data, and noisy environments. This problem shares many features with speech processing in human, while presenting new challenges of its own.…”