Signatures of personality on dense 3D facial images

Abstract

It has long been speculated that there exist cues on human face that allow observersto make reliable judgments of otherspersonality traits. However, direct evidences ofassociation between facial shapes and personality are missing. This study assessed thepersonality attributes for 834 Han Chinese volunteers (405 males and 429 females) utilizing the five-factor personality model (the Big Five model), and collected their neutral 3D facial images. Dense anatomical correspondence was established across the 3D facial images to allow high-dimensional quantitative analyses on the face phenotypes. Two different approaches, Composite Partial Least Square Component(CPLSC) and principle component analysis (PCA) were used to test the associations between the self-testedpersonality scores and the dense 3D face image data. Among the fivepersonality factors, Agreeableness and Conscientiousness in male, and Extraversion in female were significantly associated to specific facial patterns. The personality-related facial patterns were extracted and their effects were extrapolated on simulated 3Dfacial models.