Abstract
It is a common conception among beginner face illustrators that drawing or copying a face that is vertically inverted (upside-down) will improve the accuracy of the drawing because there will be less interference from how we expect the face to look. This study looks into that conception; decades of research have shown that holistic processing of faces is disrupted when they are inverted. The disruption of holistic processing should thus lead to more accurate face drawings. We tested this theory by having participants draw upright and inverted parameterized face profiles and analyzing their accuracy. We constructed a parameterized profile face space by recording the location of 66 landmark points on a collection of 720 profile faces (see Davidenko, 2007). For this study, we generated 16 parameterized faces by sampling from the first 8 dimensions of this face space. In each trial, participants were shown a face on the left side of the screen and had 45 seconds to copy it on the right side of the screen, using a stylus on a Windows 8 Surface tablet. Each participant drew each of 16 faces both upright and inverted (the presentation order was randomized and counterbalanced across subjects). We carefully recorded the location of the 66 landmark points on each face drawing, allowing us to compute a distance metric between each drawing and its corresponding original face [see supplementary materials fig.1]. This distance metric served as a measure of accuracy, with higher distances corresponding to greater errors. Contrary to common belief, people's drawings were significantly more accurate for upright versus inverted faces (t(15) = 4.9; p=0.0002) [see supplementary materials fig.2]. Our results suggest that holistic processing may be vital to accurate face drawing.
Meeting abstract presented at VSS 2015