Abstract
Evaluating the pain experienced by someone else is a skill of high social and biological importance. Interestingly, underestimation bias in pain judgments are often observed. The present study aims at investigating the way an observer has encoded the appearance of facial expressions of pain in visual memory as one potential perceptual source for this bias. The mental representation of pain facial expressions was extracted in 49 participants using Reverse Correlation (Mangini & Biederman, 2004). On each trial, a base face embedded in white sinusoidal noise was presented, and participants were asked to judge, on a scale from 0 to ten, the degree to which it expressed pain. Participants were then presented with videos of individuals experiencing different levels of pain, after which they were asked to evaluate their pain. A region-of-interest analysis was then conducted to measure the salience with which three core facial features associated with pain expressions were coded in the mental representations (i.e. eyes narrowing, brow lowering, nose wrinkling/upper-lip raising). A correlation between the saliency of these three features and the underestimation bias of each participant was then calculated. The results confirm the presence of an underestimation bias in our sample (t(48)=-8.5, p<.001) and replicate previous findings showing that brow lowering and nose wrinkling/upper-lip raising are given more weight than eye narrowing in the average mental representation (Blais et al., 2019). The underestimation bias was also significantly correlated with the saliency of the brow lowering (r=0.32, p=.03) and the nose wrinkling/upper-lip raising (r=-0.44, p=.002) features, but not with the saliency of eye narrowing (r=-0.10, p=.48). Overall, these results indicate that perceptual factors may underlie the underestimation bias. Individuals that encode pain expressions by giving more importance to nose wrinkling/upper-lip raising than brow lowering show a higher tendency to underestimate the pain experienced by others.