Abstract
It has been suggested that racial disparities in pain care might be driven by a perceptual bias. Pain facial expressions are harder to detect on Black than on White faces (Mende-Siedlecki et al., 2019, 2020, 2021). However, the tasks used in those studies did not allow to disentangle whether this bias was perceptual or decisional in nature. Moreover, the impact of face race in pain efficiency discrimination between expressions subtly differing in terms of intensity remains unknown. To address this question, we used a 2-IFC task while 37 White observers were discriminating different intensities of pain facial expressions from same-race (White) and other-race (Black) faces. Four face avatars were created (2 genders x 2 races) and used to create seven intensities of pain between 35% and 65% on the neutral to pain continuum, using parametric morphing. On each trial (240 per participant), a pair of faces were presented to participants who had to identify which one expressed the highest intensity of pain. At any given trial, faces only differed on the expression of pain intensities (i.e., same race and gender). The task included six difficulty levels, determined by the difference of pain intensities between the two stimuli (i.e., 5, 10, 15, 20, 25 or 30% difference). Our data show that observers may have a reduced sensitivity when trying to discriminate between pain intensities of outgroup faces only when differences are very subtle, regardless of the level of pain intensity. These observations support the perceptual hypothesis and pave the way to investigate the roots and consequences of the decoding of pain in other-race faces. Overall, they demonstrate that the fine-grained evaluation of the facial expression of pain is finely tuned to same-race faces.