October 2020
Volume 20, Issue 11
Open Access
Vision Sciences Society Annual Meeting Abstract  |   October 2020
Linking language descriptions and social trait perception of three-dimensional body shapes
Author Affiliations
  • Ying Hu
    The University of Texas at Dallas, USA
  • Victoria Huang
    The University of Texas at Dallas, USA
  • Matthew Q. Hill
    The University of Texas at Dallas, USA
  • Alice J. O’Toole
    The University of Texas at Dallas, USA
Journal of Vision October 2020, Vol.20, 581. doi:https://doi.org/10.1167/jov.20.11.581
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Ying Hu, Victoria Huang, Matthew Q. Hill, Alice J. O’Toole; Linking language descriptions and social trait perception of three-dimensional body shapes. Journal of Vision 2020;20(11):581. doi: https://doi.org/10.1167/jov.20.11.581.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Three-dimensional body shapes can be reconstructed using physical attribute descriptions (e.g., skinny, wide shoulders; Hill et al., 2016; Streuber et al., 2016). Body shapes can also predict diverse social trait judgments (e.g., extraverted, lazy; Hu et al., 2018). Here we explored the relationships between body shapes, physical descriptions, and social judgments, bridging the gap between Hill et al. (2016) and Hu et al. (2018). Bodies (N=140; 70 female, 70 male; Hu et al., 2018) were synthesized using normally distributed random values on 10 coefficients in a PCA space from the Skinned Multi-Person Linear Model (Loper et al., 2015), a model derived from laser scans of over 1700 people. Each body was rated (N=38 raters) using 30 physical attribute descriptions. First, we predicted these descriptions from body shape coefficients using multiple linear regression. Body shapes predicted physical descriptions (average cosine similarity between predicted and rated physical descriptions was 0.78 for female and 0.75 for male bodies) more accurately than body shapes predicted social judgments (0.36 for female and 0.39 for male bodies, Hu et al., 2018). Second, we explored the correlations between physical descriptions and trait judgments. Social judgments that indicated high extraversion, conscientiousness, openness, and low neuroticism were positively correlated with skinny, fit and attractive. These judgments were negatively correlated with heavy, round apple, curvy, and broad shoulders. Third, we predicted social trait judgments from physical attribute descriptions. The best-predicted traits were from the extraversion and conscientiousness domains. Physical attribute descriptions predicted trait judgments (average cosine similarity between predicted and rated trait judgments was 0.48 for female and 0.42 for male bodies) more accurately than body shapes predicted trait judgments (0.36 for female, 0.39 for male bodies, Hu et al., 2018). This study provides insights useful for integrating 3D body modeling, verbal descriptions, and social perception of bodies.

×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×