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

Abstract

Many species of animals deliver vocalizations in sequences presumed to be governed by internal rules, though the nature and complexity of these syntactical rules have been investigated in relatively few species. Here I present an investigation into the song syntax of fourteen male Cassin’s Vireos (Vireo cassinii), a species whose song sequences are highly temporally structured. I compare their song sequences to three candidate models of varying levels of complexity–zero-order, first-order and second-order Markov models–and employ novel methods to interpolate between these three models. A variety of analyses, including sequence simulations, Fisher’s exact tests, and model likelihood analyses, showed that the songs of this species are too complex to be described by a zero-order or first-order Markov model. The model that best fit the data was intermediate in complexity between a first- and second-order model, though I also present evidence that some transition probabilities are conditioned on up to three preceding phrases. In addition, sequences were shown to be predictable with more than 54% accuracy overall, and predictability was positively correlated with the rate of song delivery. An assessment of the time homogeneity of syntax showed that transition probabilities between phrase types are largely stable over time, but that there was some evidence for modest changes in syntax within and between breeding seasons, a finding that I interpret to represent changes in breeding stage and social context rather than irreversible, secular shifts in syntax over time. These findings constitute a valuable addition to our understanding of bird song syntax in free-living birds, and will contribute to future attempts to understand the evolutionary importance of bird song syntax in avian communication.

Finding, visualizing, and quantifying latent structure across diverse animal vocal repertoires

“…To compare and visualize the structure captured by both PAF and spectrogram representations of vocalizations, we used a subset of the 20 most frequent syllable-types from a dataset of Cassin’s vireo song recorded in the Sierra Nevada Mountains [ 7 , 41 ]. We computed both spectrographic representations of syllables as well as a set of 18 temporal, spectral, and fundamental characteristics ( S2 Table ) over each syllable using the BioSound python package [ 24 ].…”

“…If clusters in latent space are highly similar to experimenter-labeled element categories, unsupervised latent clustering could provide an automated and less time-intensive alternative to hand-labeling elements of vocalizations. To examine this, we compared how well clusters in latent space correspond to experimenter-labeled categories in three human-labeled datasets: two separate Bengalese finch datasets [ 63 , 64 ], and one Cassin’s vireo dataset [ 7 ]. We compared four different labeling techniques: a hierarchical density-based clustering algorithm (HDBSCAN; Fig 9 ; [ 65 , 66 ]) applied to UMAP projections of spectrograms, HDBSCAN applied to PCA projections of spectrograms (HDBSCAN is applied to 100-dimensional PCA projections rather than spectrograms directly because HDBSCAN does not perform well in high-dimensional spaces [ 66 ]), k-means [ 67 ] clustering applied over UMAP, and k-means clustering applied over spectrograms ( Fig 10 ; Table 1 ).…”

“…The California thrasher dataset is from [ 6 ] and was acquired from BirdDB [ 41 ]. The Cassin’s vireo dataset is from [ 7 ] and was also acquired from BirdDB. The giant otter dataset was acquired from [ 98 ].…”

Five Ways in Which Computational Modeling Can Help Advance Cognitive Science: Lessons From Artificial Grammar Learning

“…Neural network projections of birdsong vocalizations into a 2D latent space. (A) A scatter plot where each point in 2D space represents a syllable sung by a Cassin's vireo (library acquired from Hedley, ). Colors denote hand labeled syllable categories, which tend to cluster in the low‐dimensional space.…”

TweetyNet: A neural network that enables high-throughput, automated annotation of birdsong

“…Annotation makes several types of analyses possible. For example, annotation is required to build statistical models of syntax [11][12][13][14], to fit computational models of motor learning that precisely quantify how single syllables change over the course of an experiment [15,16], and to relate behavior to neural activity [17][18][19]. Annotating song greatly increases our ability to leverage songbirds as a model system when answering questions about how the brain produces syntax observed in sequenced motor skills, and how the brain learns to adaptively control muscles.…”