Scholarlink

<p>Sexual dimorphism in birds is frequently expressed through plumage traits. Yet, the extent to which males and females also differ acoustically remains poorly understood in many Neotropical passerines. In antbirds (Thamnophilidae), several species exhibit pronounced plumage dimorphism, but whether these visual differences are accompanied by consistent vocal divergence is still unclear. This gap limits our understanding of how different signaling modalities evolve under sexual and social selection in suboscine birds. Here, we evaluated sexual differences in vocal behavior in four Cerrado antbird species (<i>Thamnophilus doliatus</i>, <i>T. pelzelni</i>, <i>Herpsilochmus atricapillus</i>, and <i>H. longirostris</i>). We employed a comparative framework combining traditional bioacoustic measurements, mel-frequency cepstral coefficients, and deep learning-based BirdNET embeddings. We evaluated sex differences through multivariate analyses, linear mixed effects models, and supervised classification approaches. Across species and analytical methods, male and female vocalizations showed extensive overlap, and classification performance was generally low and only modestly above chance. However, the strength of differentiation varied among species, with <i>T. pelzelni</i> showing the clearest evidence of subtle sex-related divergence. Together, our results indicate that vocal sexual dimorphism in these Cerrado antbirds is generally weak and inconsistent, contrasting with the pronounced plumage dimorphism observed in the same taxa. These findings suggest that sexual differentiation may be expressed unevenly across communication modalities, with stronger divergence in visual than acoustic traits. More broadly, our study highlights the importance of integrating traditional and machine learning-based approaches to investigate subtle variation in animal communication and provides a framework for studying multimodal sexual dimorphism in suboscine birds.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Traditional bioacoustic analyses and machine-learning methods indicate weak vocal dimorphism in four Cerrado antbird species

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月27日

查看全文
  • Enrico L. Breviglieri,
  • Larissa S. M. Sugai,
  • Guilherme Sementili-Cardoso,
  • Reginaldo J. Donatelli,
  • Irina Tolkova

摘要

Sexual dimorphism in birds is frequently expressed through plumage traits. Yet, the extent to which males and females also differ acoustically remains poorly understood in many Neotropical passerines. In antbirds (Thamnophilidae), several species exhibit pronounced plumage dimorphism, but whether these visual differences are accompanied by consistent vocal divergence is still unclear. This gap limits our understanding of how different signaling modalities evolve under sexual and social selection in suboscine birds. Here, we evaluated sexual differences in vocal behavior in four Cerrado antbird species (Thamnophilus doliatus, T. pelzelni, Herpsilochmus atricapillus, and H. longirostris). We employed a comparative framework combining traditional bioacoustic measurements, mel-frequency cepstral coefficients, and deep learning-based BirdNET embeddings. We evaluated sex differences through multivariate analyses, linear mixed effects models, and supervised classification approaches. Across species and analytical methods, male and female vocalizations showed extensive overlap, and classification performance was generally low and only modestly above chance. However, the strength of differentiation varied among species, with T. pelzelni showing the clearest evidence of subtle sex-related divergence. Together, our results indicate that vocal sexual dimorphism in these Cerrado antbirds is generally weak and inconsistent, contrasting with the pronounced plumage dimorphism observed in the same taxa. These findings suggest that sexual differentiation may be expressed unevenly across communication modalities, with stronger divergence in visual than acoustic traits. More broadly, our study highlights the importance of integrating traditional and machine learning-based approaches to investigate subtle variation in animal communication and provides a framework for studying multimodal sexual dimorphism in suboscine birds.