Adults’ and children’s perception of facial expressions is influenced by body postures even for dynamic stimuli

ABSTRACT

A growing literature shows that body postures influence recognition of static facial expressions; a fearful face, for example, is perceived as angry when presented on an angry body posture. In daily life, however, people conveying emotions are moving. Here we provide the first examination of such congruency effects for stimuli with naturalistic movement. Adults and children were asked to label the facial expression in static or dynamic whole-person displays comprising congruent (e.g., sad face on sad body) and incongruent (e.g., sad face on fearful body) expressions. Recognition was impaired on incongruent trials, especially for dynamic stimuli and despite eye-tracking data confirming that both age groups attended to the face, as instructed. Our findings highlight the importance of integrating whole-person and dynamic stimuli into research and theories of emotion perception.

KEYWORDS:

• Congruency effects
• person perception
• emotion perception

Acknowledgements

We thank Bryce Hunt and Gabriela Salgado for their assistance in creating the stimuli for this project, and Tamara Van Der Zant for her help in collecting the data.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 We use the terms “accuracy” and “recognition” to indicate that participants judged the emotion displayed as the emotion predicted by the researcher.

2 Most expressions in the literature are posed, and the poser is not feeling the emotion they are displaying. Thus, we use phrases like “fear face” and “fear posture” to indicate expressions that were judged by naive observers as portraying that emotion, not that the expression displayed is the expression made by individuals feeling that emotion.

3 All ANOVAs performed were fixed effects ANOVAs.

4 Errors involving happy faces or happy responses were rare. Participants rarely responded incorrectly when the face showed a happy expression and rarely called a negatively valenced expression “happy”. Because of the low number of trials with errors, we are only providing descriptive statistics for patterns of errors.

Additional information

Funding

This research was supported by a grant from Social Sciences and Humanities Research Council and Canada Foundation for Innovation awarded to Catherine Mondloch and funding provided by the University of Queensland awarded to Nicole Nelson.