A Computational Tool for Automated Large-Scale Analysis and Measurement of Bird-Song Syntax

Arij Daou, Frank Johnson, Wei Wu, Richard Bertram

We present computer software for automated, high throughput, quantitative syllable-level analysis of bird song syntax. The primary advantage of our tool is the ease and effectiveness it provides in quantifying syllable sequence and performing a comparison of syllable sequence from one day of singing with one or more other days of singing. The software utilizes the output of the Feature Batch module in Sound Analysis Pro (Tchernichovski et al., 2000) that can be used to measure the temporal and spectral features of each syllable produced during a day of singing. We use these measurements to identify individual syllables based on their temporal and spectral properties and then identify transition probabilities among syllables to determine changes in syntax. This quantifies the ordering of syllables in songs and the frequency with which subsequences appear. Moreover, the software computes the linearity, consistency, and stereotypy scores for every bout presented as well as descriptive statistics for each of these measures for each day of singing. We also report statistical measures that the software utilizes (the Kullback-Leibler distance and the sequence entropy) to quantify the degree of dissimilarity between sequences of syllable transitions. Our tool is useful for comparing the syntactic structure of songs produced by a bird prior to and after a manipulation such as ablation of part of the vocal motor pathway or infusion of pharmacological agents, or for assessing the degree of individual variation in syntactic structure across populations of birds.