Abstract
The eye-size illusion refers to the fact that when the size of a face’s frame is changed but the eye size is unchanged, observers judge the eye size to be different from that in the original face frame (Rakover, 2011; Xiao et al., 2014). Indirect evidence suggests holistic processing to be involved in engendering this illusion (Xiao et al., 2014). To test directly the role of holistic processing in the eye-size illusion, we used the composite face paradigm (Young et al., 1987), a gold standard test of holistic processing. Participants (n=30) judged which eyes were bigger in two conditions (Figure 1). In the original-composite face condition (Figure 1a & c), one face is the original face and another is the composite face (the top is the same as the original but the bottom is a different person’s). In the composite-composite condition (Figure 1b & d), two faces are identical composite faces. In each condition, the faces were either aligned (Figure 1a & b) or misaligned (Figure 1c & d). The face sizes differed from the original by 6%, 10%, or 14% with the eye size unchanged. Participants’ eye-size illusion was greater with a smaller face frame, replicating the robust eye-size illusion, F(2, 28) = 11.59, p < .001, η2 = .45. Their eye-size illusion was greater in the aligned than misaligned faces, suggesting the involvement of holistic processing, F(1, 29) = 8.62, p < .01, η2 = .23. Participants also showed a greater eye-size illusion in the original-composite condition than in the composite-composite condition, F(1, 29) = 18.88, p < .001, η2 = .39, further supporting the role of holistic processing. These results together suggest that holistic face processing plays an important in the perception of the eye size illusion.
Meeting abstract presented at VSS 2015