Dan Oh (DongWon) | 오동원

“Don’t judge a book by its cover,” the saying goes, and, being rational people, we like to believe we follow it. But in reality, we make snap judgments (quite literally, face-value ones) all the time. A single glance at someone’s face can affect whether we trust them, approach them, or keep our distance. These rapid first impressions influence everyday decisions and social interactions (Olivola et al., 2014). One mechanism behind this is known as "emotion overgeneralization:" people unconsciously extend subtle emotional cues into stable personality judgments. For example, faces that structurally resemble happy expressions are seen as more trustworthy, even when no emotion is explicitly displayed (Zebrowitz, 2017). In fact, among various facial cues, this resemblance to happiness is a particularly strong predictor of perceived trustworthiness (Jaeger & Jones, 2022). But how universal is this effect? Most of the existing evidence comes from Western samples, where emotional expression is frequent, encouraged, and socially informative. Emotional norms, however, differ across cultures. While a smile may signal friendliness or intelligence in some countries, in others, it can be interpreted as foolishness or misplaced informality (Krys et al., 2014). In East Asian societies such as China and Japan, emotional restraint is much more common than emotional expression (Heine et al., 1999; Rychlowska et al., 2015). What if emotion overgeneralization depends on how frequently and consistently emotional cues co-occur with relevant social outcomes? If so, can its strength vary systematically across cultures? This is the question Yuqing Shi (@syuuqing), a Ph.D. student at the National University of Singapore working with Dr. DongWon Oh (@dongwon_oh), is addressing in her dissertation. Growing up in China, Yuqing observed how differently smiles are interpreted depending on context. In Chinese culture, there is a saying, “喜怒不形于色”— do not display joy or anger on one’s face. These cultural scripts emphasizing emotional restraint contrast sharply with contemporary U.S. norms that equate smiling with warmth and trustworthiness. Yuqing combines large-scale surveys with computational facial analysis. Participants from multiple countries rate the trustworthiness of neutral faces, while deep-learning algorithms quantify how closely those faces resemble happiness-related expressions. She then examines how individual and societal norms around emotional expressivity relate to the tendency to overgeneralize emotion cues into trust judgments. If first impressions are indeed culturally learned rather than universal, this has broad implications — not only for everyday social interaction, but for global collaboration, hiring, diplomacy, and even AI systems trained on culturally biased facial data. Yuqing is a Besample Dissertation Grant finalist, and we are excited to support research that asks whether something as automatic as face perception is universal or culturally constructed. #PhDResearch #PhDGrant #CrossCulturalResearch #GlobalResearch #BesampleDissertationGrant #Besample