Abstract
Normal faces rapidly presented in the visual peripher are perceived as grotesque and distorted (Tangen, Murphy, Thompson, 2011). This phenomenon, "The flashed-face distortion effect" (FFD) is a powerful illusion that reveals important properties of how faces are coded in peripheral vision. Presently, we manipulated aspects of facial appearance and spatial/temporal properties of the flashed-face stimulus to determine what factors modulate the illusion's strength. In our first study, we examined how face orientation (upright vs. inverted), makeup, and image size each affected FFD strength. Face inversion was intended as a means of determining the illusion's face-specificity, while varying makeup and image size was intended to reveal how local contrast and spatial scale impacted the FFD. We recruited 28 participants to complete two experimental blocks: One in which makeup and orientation were varied across trials, and another in which makeup and image size (at a fixed eccentricity) were varied. Participants rated apparent distortion using a 1-7 scale. We found that neither the presence of makeup nor face orientation impacted the FFD, but image size did. Specifically, faces presented at an intermediate size (~4 degrees of visual angle) were rated as more distorted than smaller or larger faces (p=0.003). In our second study, we examined how perceived distortion was impacted by makeup and both the eccentricity and duration of the images comprising the FFD sequences. Briefly, we again found no impact of makeup on the FFD, but did find that both longer image durations (p< 0.001) and greater eccentricities (p< 0.001) led to significantly more distortion. These results suggest that the FFD is not a face-specific effect, but instead depends on low-level spatial and temporal factors. The FFD may thus reflect the dynamics of pooling operations in the visual periphery and should be observable with other object classes.
Meeting abstract presented at VSS 2017