Abstract:There is a growing body of evidence showing that humans make automatic and reliable personality inferences from facial appearance. Interestingly, it has been robustly shown that the recognition of other-race faces is impaired compared to same-race faces (the so-called other-race effect), with categorization of gender and age also achieved inefficiently. However, the extent to which the ability of making reliable inferences from faces generalizes across cultures and faces from different races is poorly understood. This issue is even timelier considering recent studies that have challenged the universality of face processing. For instance, we recently showed that Westerners predominantly fixate the eyes during face recognition, whereas Easterners fixate the central region of the face (i.e., nose) (Blais et al., 2008). Culture also modulates the strategy observers use to gather visual information during facial expression categorization (Jack et al., 2009). Therefore, asking whether social judgments from faces generalize across cultures is a natural question to be addressed. To this aim, we tested Western Caucasian and East Asian observers performing visual and social judgments from all the combinations of face pairs sampled from 40 Western Caucasian and 40 East Asian unfamiliar faces. Observers from both cultures first evaluated the physical and the social similarity of each face pair. Subsequently, observers performed binary social evaluations on the same face pairs for attractiveness, competence, trustworthiness and warmth. All binary decisions were paired with a measure of confidence. Finally, we represented the face space for each of the judgments in matrices of dissimilarity weighted by confidence levels for each observer and culture. Mantel correlations performed on the matrices of dissimilarity indicated a fairly robust agreement across cultures for all judgments, both visual and social. Our data show that humans rely on universal rules to perform trait inferences from facial appearance.

Diverse Face Images (DFI): Validated for racial representation and eye gaze

Diversity in Faces

Social Judgments from Faces Are Universal

The Chicago face database: A free stimulus set of faces and norming data

Hawai'i face database: a racially and ethnically diverse set of facial stimuli

SDFD: Building a Versatile Synthetic Face Image Dataset with Diverse Attributes

FairFace: Face Attribute Dataset for Balanced Race, Gender, and Age

Presentation and validation of the Radboud Faces Database

TCDiff: Triple Condition Diffusion Model with 3D Constraints for Stylizing Synthetic Faces

About Face: A Survey of Facial Recognition Evaluation

Explaining Bias in Deep Face Recognition via Image Characteristics

Multi-dim: A Multi-Dimensional Face Database Towards the Application of 3D Technology in Real-World Scenarios.

DifFIQA: Face Image Quality Assessment Using Denoising Diffusion Probabilistic Models

DIG-FACE: De-biased Learning for Generalized Facial Expression Category Discovery

Towards Measuring Fairness in AI: the Casual Conversations Dataset

Validation of dynamic virtual faces for facial affect recognition

Face Image Quality Assessment for Model and Human Perception

D2SP: Dynamic Dual-Stage Purification Framework for Dual Noise Mitigation in Vision-based Affective Recognition

Ig3D: Integrating 3D Face Representations in Facial Expression Inference

InFER: A Multi-Ethnic Indian Facial Expression Recognition Dataset

IIITM Face: A Database for Facial Attribute Detection in Constrained and Simulated Unconstrained Environments