Figure 1.
Measurements for symmetry and sexual dimorphism.
Symmetry was calculated by taking left and right deviation from the midline, calculated from inter-pupillary distance, for points D1-D6 and then summing the absolute value of individual scores. Sexual dimorphism was measured by measuring distance between specific points and calculating four ratios based on these distances: Cheekbone Prominence (ChP, D3/D6), Jaw Height/Lower Face Height (JH/LFH, D9/D8), Lower Face Height/Face Height (LFH/FH, D8/D7), and Face Width/Lower Face Height (FW/LFH, D3/D8). All images were normalised on inter-pupillary distance.
Figure 2.
Asymmetry (+/− 1SE of mean) of faces classified as male or female in the discriminant analysis by sex of face.
A significant interaction was found between sex of face and classification (F1,835 = 4.07 , p = .044) indicating that those correctly classified to their own sex were more symmetric than those misclassified to the opposite-sex.
Figure 3.
High and low symmetry composite faces for macaques, Hadza, and Europeans.
All images are normalised on inter-pupillary distance to control relative image size, have been made perfectly symmetric, and each high/low pair possesses the average colour information of both. Perceptual differences are then dependent on shape differences between high and low symmetry faces that are independent of symmetry.
Figure 4.
Proportion of individuals choosing high and low symmetry composite faces for macaques, Hadza, and Europeans as most sex-typical (i.e. masculine for males, feminine for females).