Social Perception of Male Faces. Individuals, Clusters, Dimensions

In an online experiment in a German and an English version, portrait photographs of 59 men from the age range late 20s to late 50s served as stimulus material. 1618 participants each rated a randomly selected photo on 36 personality-descriptive rating scales and estimated the age, body height and weight. In supplementary assessments, the degree of smile, degree of hair loss, hair color, facial hair, glasses, formal dress, head rotation, head tilt and brightness of the image background were determined. All variables show sufficient, mostly high to very high reliability. The data analysis was carried out at the level of the stimulus persons, on the one hand from a dimensional and on the other hand from a typological perspective. A principal component analysis of the personality-descriptive traits yielded a five-dimensional space with orthogonal factors. These can be interpreted straightforwardly as Social Agreeableness, Attractiveness, Masculinity, Status/Intelligence and Negative Affects. Using multiple regression, the age estimates and the degree of smiling can be fitted into the psychological space in an excellent way. Wearing glasses and height estimates can also be localized well in this space. In addition to the customary dimensional approach, a typological approach was adopted, which is rarely found in this field of research. In a hierarchical cluster analysis of the five personality dimensions and the external characteristics, groups of stimulus persons were identified who are similar to each other in a bundle of characteristics and who differ markedly from other groups. These clusters were mapped into the five-dimensional face perception space. In this way, it is demonstrated that not only does each approach provide interesting insights in its own right, but that both approaches mutually enrich each other by combining them. A main emphasis of this paper is the comparison with the two-dimensional Valence-Dominance model, which has been the most influential paradigm for more than a decade. It is not disputed that evaluation on the Badness – Goodness dimension is of paramount importance, nor is the importance of Dominance or Power called into question. Nevertheless, it is shown that a two-dimensional model cannot do justice to the complexity of face perception. Face perception is much more differentiated and – this is the crucial point – it largely coincides with the personality structure known from differential psychology, which is based on self-assessments and assessments of acquaintances.