Face inversion disproportionately disrupts detection of vertical versus horizontal, not long-range versus short-range, spatial relations

Abstract

Presenting a face inverted disrupts sensitivity to the spacing between the features. Recent evidence suggests that inversion disproportionately affects sensitivity to vertical over horizontal changes in eye position. One explanation is that inversion disrupts the processing of long-range (e.g., eye to mouth) more than short-range (e.g., interocular) spatial relations. Here we investigated whether the size of the shift or the direction (i.e., eyes up versus eyes down; eyes in versus eyes out) affects the pattern of the inversion effect. Our results replicated the finding of poor detection of vertical changes in inverted faces (even when these changes were very large). In the vertical condition the inversion effects were just as large for the “eyes down” (short-range, as the eye-mouth distance was small) as the “eyes up” (long-range), while horizontally “eyes in” (short-range, as the eyes were very close) and “eyes out” (long-range) both produced small inversion effects. These results argue against a long-range versus short-range spatial relations explanation of the horizontal versus vertical difference in the inversion effect. Rather, it appears the axis of change rather than the spatial distances between features is behind the difference in the size of the inversion effect.