Scholarlink

<p>Nonverbal behaviour (NVB) is a critical component of human communication, particularly in high-stakes, naturalistic settings such as sport. Despite advances in computational analysis, the extent to which objectively computable NVB features reflect subjective human judgments remains unclear. In this study, we present the inaugural investigation into the correspondence between computable NVB indicators and observer perceptions in the context of elite football. The utilisation of 3D pose data from a FIFA World Cup match enabled the extraction of volumetric, angular and kinematic features from players’ body movements during match events. Parallelly, human participants rated short video segments of these events on multiple NVB dimensions. Statistical analyses revealed significant differences in objective computable NVB indicators across play contexts (scoring, conceding, regular play), with players on the scoring team exhibiting more expansive and dynamic body language. Subjective ratings closely mirrored these differences, and repeated measures correlation analyses demonstrated strong associations between objectively computable NVB features and human judgments. These findings may provide a scientific basis for the use of 3D pose-based metrics for scalable, automated NVB analysis and may highlight their ecological validity in capturing expressive behaviours meaningful to human observers. The study has the potential to contribute to the advancement of methodological approaches for NVB research.</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

Nonverbal Behaviour in Football: A 3D Pose-Based Analysis of Player Expressions and Observer Perceptions

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年3月11日

查看全文
  • Aki Uehata,
  • Philip Furley,
  • Daniel Memmert

摘要

Nonverbal behaviour (NVB) is a critical component of human communication, particularly in high-stakes, naturalistic settings such as sport. Despite advances in computational analysis, the extent to which objectively computable NVB features reflect subjective human judgments remains unclear. In this study, we present the inaugural investigation into the correspondence between computable NVB indicators and observer perceptions in the context of elite football. The utilisation of 3D pose data from a FIFA World Cup match enabled the extraction of volumetric, angular and kinematic features from players’ body movements during match events. Parallelly, human participants rated short video segments of these events on multiple NVB dimensions. Statistical analyses revealed significant differences in objective computable NVB indicators across play contexts (scoring, conceding, regular play), with players on the scoring team exhibiting more expansive and dynamic body language. Subjective ratings closely mirrored these differences, and repeated measures correlation analyses demonstrated strong associations between objectively computable NVB features and human judgments. These findings may provide a scientific basis for the use of 3D pose-based metrics for scalable, automated NVB analysis and may highlight their ecological validity in capturing expressive behaviours meaningful to human observers. The study has the potential to contribute to the advancement of methodological approaches for NVB research.