Facial features for age judgments across cultures

Abstract

Until recently, face processing was thought to be culturally invariant. Cross-cultural studies have shown, however, that culture shapes how the brain processes face information to recognize expression (Jack et al., 2009; Jack et al., 2012), ethnicity (Blais et al, 2008), or identity (Tardif et al., 2017; Estéphan et al., 2018). Although it is known that the physical signs of facial aging vary across ethnicities—e.g. Chinese skin starts to wrinkle about 10 years after European skin (Nouveau-Richard et. al., 2005)—and that differences in sociocultural practices can impact on the efficiency of facial-age judgments (Anzures et al., 2010), no study has yet investigated how culture affects the use of facial features for making age judgments. We asked 128 Western-Caucasian women and 132 Chinese women from five age groups (20–30, 31–40, 41–50, 51–60, and 61–70 year-olds) to complete a three-hour age estimation task on 300 high-resolution color photos of Western-Cauca-sian and Chinese women faces from the same five age groups. We applied the Bubbles technique (Gosselin & Schyns, 2001) to reveal directly the facial areas that subtend facial age judgments. Each observer was submitted to 1,600 “bubblized” faces from one ethnicity in a full factorial design. Facial information was sampled through randomly located Gaussian apertures, or “bubbles”, at five spatial frequency scales. The GLM was used to assess effects of culture and ethnicity in use of facial information for making age judgments (Murray, 2012). Our most intriguing result is that within-culture age judgments relied mostly on the eye areas in mid-to-high spatial frequencies whereas between-culture age judgments relied on many facial areas spread across all spatial frequencies.

Acknowledgement: Unilever