Fig 1.
Sample facial photographs from the database.
Released under a CC BY license, with permission from the face-bearers.
Fig 2.
Histograms for and Correlations between fWHR, BMI, Perceptual Norms, and GM Measurements (DFA = Distance from Average; MF = Maleness/Femaleness) for Males.
Fig 3.
Histograms for and correlations between fWHR, BMI, perceptual norms, and GM measurements (DFA = Distance from Average; MF = Maleness/Femaleness) for Females.
Table 1.
Descriptive statistics for fWHR, BMI, perceptual norms, and GM measurements.
Fig 4.
Thin plate-splines demonstrating the results of multivariate regression of shape coordinates on fWHR and scores of maleness/femaleness.
Deformation grids shows differences in facial shape associated with high and low value of measurements for both men and women compared to an average configuration in the middle.
Fig 5.
Visualization of shape regressions illustrating changes in facial shape associated with perception of dominance, femininity, masculinity, and Turkishness for males.
Each perceived characteristic is shown as thin plate-spline deformations (within observed range and 3x extrapolated) compared to a consensus in the middle. The results for attractiveness and perceived trustworthiness were not statistically significant and cannot be visualized.
Fig 6.
Visualization of shape regressions illustrating changes in facial shape associated with perception of attractiveness, dominance, femininity, masculinity, trustworthiness, and Turkishness for females.
Each perceived characteristic is shown as thin plate-spline deformations (within observed range and 3x extrapolated) compared to an average configuration in the middle.
Fig 7.
Facial composites demonstrating changes in facial shape associated with perception of attractiveness, dominance, femininity, masculinity, trustworthiness, Turkishness, fWHR, and scores of maleness/femaleness (MF) for females (left panel) and perception of dominance, femininity, masculinity, Turkishness, measures of fWHR, and scores of MF for males (right panel).
Each perceived characteristic shows composite consisting of 10 averaged facial textures unwarped to predicted configuration (within observed range) compared to a consensus in the middle. The results for attractiveness and perceived trustworthiness in men were not statistically significant and cannot be visualized.
Table 2.
Summary of results for shape regressions of facial coordinates on eight predictors.