Feature proportion accuracy of hand-drawn facial approximation.

Several studies have investigated the predictive accuracy of facial approximation methods for individual facial features, but few have investigated the ability of these methods to accurately predict the relative dimensions of these features. Photographs of 10 skulls from the William M. Bass Donated Skeletal Collection were used to create hand-drawn facial approximations following the guidelines presented in Taylor (2001). Measurements of the eyes, nose, and mouth were made and converted to ratios for comparison to their corresponding antemortem photographs. Potential error introduced by angulation in the antemortem photographs was assessed by finding similar ratio values for volunteers who were photographed at a series of angles. Eye measurements were distorted when individuals were turned 30 to 70° away from the camera. However, all antemortem photographs displayed individuals turned 30° or less from the camera. Wilcoxon signed rank tests revealed no significant differences between the hand-drawn approximations and the antemortem photograph measurements values for all feature relationships except for the eye width to iris diameter ratio, likely because the Taylor (2001) method does not reference the underlying bone when approximating the size and appearance of the eye.