Inverted faces benefit from whole-face processing

Abstract Upright faces are thought to engage holistic processing whereby local regions are integrated into a unified whole for the purposes of rapid, efficient analysis. In contrast, inverted faces are thought to recruit a slower, less-accurate serial analysis of local features. Aperture paradigms, whereby a target face is revealed by a dynamic viewing window that shifts over the stimulus image, offer a compelling test of this view. If upright faces are processed holistically, perceptual judgements ought to be substantially disrupted when stimuli are viewed through apertures. In contrast, aperture viewing should produce little or no decrement in perceptual decisions when judging inverted faces, as they are thought to be subjected to serial feature-based analysis. Here we present four experiments that elucidate the effects of aperture viewing on the perception of upright and inverted faces. In our first two experiments, we find evidence of disproportionate aperture effects for upright faces relative to inverted faces. However, these findings are qualified by the fact that observers found it harder to discriminate inverted faces presented in the ‘baseline’ whole-face condition. When observers’ ability to discriminate faces in the whole-face condition was matched for difficulty (Experiments 3 and 4), we show that upright and inverted faces produce very similar aperture effects. These findings indicate that both upright and inverted faces benefit from whole-face processing and accord with other lines of evidence that faces engage qualitatively similar types of processing in both orientations.