Arianne Richer¹ (), Camille Saumure², Daniel Fiset¹, Zoé Glardon², Caroline Blais¹; ¹University of Quebec in Outaouais, ²University of Fribourg

Facial expressions are crucial for assessing others' affective states. However, pain facial expressions (PFE) are poorly recognized, often confused with other negative affective states (Kappesser & Williams, 2002) and less easily perceived in women's faces (Riva et al., 2011). Studies have revealed various configurations of PFE (Kunz & Lautenbacher, 2014). Yet, it is unclear whether some of these configurations are easier to recognize than others–such differences may in part explain the disparities in perceived pain as a function of face gender. This study explored potential gender differences in the configurations of PFE (encoding) as well as their perception by external observers (decoding). We used the Delaware Pain Database (DPD; Mende-Siedlecki et al., 2020), containing 225 pictures of White individuals posing PFE. To investigate potential differences in PFE encoding between men and women, we used OpenFace to measure the activation levels of 17 action units (AUs) in those pictures. A principal component analysis indicated five main groups of AUs with correlated activations. Most importantly, the first component included AUs typically associated with PFE, and was more prominent in PFE of men than women. To verify if PFE are decoded differently as a function of face gender, we used ratings openly available within the DPD. Each picture in the DPD has indeed been rated on the perceived intensity of six basic emotions and pain. A mixed ANOVA 7 (affective states) by 2 (genders) indicated significant main effects of affective states and gender, as well as an interaction between both factors. T-tests indicated that fear and sadness were perceived as significantly higher in women's PFE, while pain was perceived as significantly higher in men's. These findings emphasize gender-specific disparities in PFE, their potential overlap with other affective states, and underscore the potential contribution of both encoding and decoding in observed gender differences.

Acknowledgements: The present study is supported by the Canada Research Chair in cognitive and social vision to Caroline Blais (# 950-232282)