The Effect of Wearing a Mask on Facial Attractiveness

Goals/Purpose: Due to the COVID-19 pandemic, people have been encouraged or mandated to wear masks in public spaces and workplaces. The use of face coverings has resulted in perceptual effects, including difficulty in identifying individuals and recognizing facial expression. One might surmise that wearing a mask may affect perceived facial attractiveness, as it can hide facial asymmetries and small deformities, and results in focus on the periorbital area. The present study examines to what extent wearing a mask affects perceived facial attractiveness.

Methods/Technique: Headshot images of 57 subjects (34 female, 23 male) with neutral facial expression were collected from Pexels, an online provider of free stock photos and videos, and put into standardized form. The control group consisted of these facial images in an anteroposterior view, with a plain white background and no facial coverings. These images were then modified using Adobe Photoshop to simulate a uniform face covering, serving as the experimental arm. Utilizing Research Electronic Data Capture (REDCap), a survey was created consisting of the control and experimental subjects in a randomized order. Two hundred seven survey participants were recruited using Amazon Mechanical Turk (Mturk) and asked to rate the control (unmasked) and experimental (masked) subjects on an ordinal scale ranging from 1 (least attractive) to 10 (most attractive). Descriptive statistics and paired sample t-tests were performed to compare the control and experimental arms.

Results/Complications: Ratings were normally distributed. For unmasked subjects who ranked within the first quartile, their average rating increased significantly when wearing a mask (3.81 and 5.83, respectively; P<0.001). For unmasked subjects who ranked within the fourth quartile, their average rating decreased significantly when wearing a mask (9.45 and 7.67, respectively; P<0.001). For unmasked female subjects, their average rating increased significantly when wearing a mask (6.70 and 6.80, respectively; P<0.001). For unmasked female subjects who ranked within the first quartile, their average rating increased significantly when wearing a mask (3.77 and 5.93, respectively; P<0.001). For unmasked female subjects who ranked within the fourth quartile, their average rating decreased significantly when wearing a mask (9.46 and 7.84, respectively; P<0.001). For unmasked male subjects, their average rating decreased significantly when wearing a mask (6.80 and 6.55, respectively; P<0.001). For unmasked male subjects who ranked within the first quartile, their average rating increased significantly when wearing a mask (3.86 and 5.67, respectively; P<0.001). For unmasked male subjects who ranked within the fourth quartile, their average rating decreased significantly when wearing a mask (9.44 and 7.44, respectively; P<0.001). The increase in attractiveness for the lower quartile and decrease for the upper quartile resulted in no significant difference between the average ratings for unmasked and masked subjects as a group (6.74 and 6.70, respectively; P=0.08).

Conclusion: When considering individuals perceived as less attractive at baseline, wearing a facial covering significantly increases perceived attractiveness. For individuals thought to be more attractive at baseline, wearing a facial covering would significantly decrease perceived attractiveness. Although there was a statistical difference when comparing mask and no mask female subjects as well as mask and no mask male subjects, the minimal change in rating challenges the idea that there may be a clinical significance.