Abstract
Face composite task has been one of the most prevalent research paradigms to demonstrate holistic processing of upright faces. Many researchers examined the exact mechanism underlying holistic processing, but holistic processing and its representation still remain elusive and controversial, some studies have suggested that holistic processing may not be evenly distributed in that processing the top-half of a face might induce stronger holistic processing than processing its bottom-half counterpart. However, the previous studies may have suffered from inadequate control of perceptual similarity between the top and bottom parts of face stimuli. In the present study, we manipulated independently perceptual similarity between the relevant (top) and irrelevant (bottom) parts of face stimuli, such that the relevant part of high versus low similarity were joined by the irrelevant part of high versus low similarity to form composite faces. Participants were asked to judge whether relevant parts of two composite faces were the same or different in a complete design. The results revealed that when the irrelevant parts were of low similarity, holistic processing as indexed by congruency effect was stronger than when irrelevant parts were highly similar. In contrast, whether or not the relevant parts were of high or low similarity, comparable levels of holistic processing were obtained. Taken together, these findings suggest that perceptual similarity, especially of the irrelevant parts, can modulate holistic processing of composite faces. Moreover, previous findings claiming asymmetry of holistic processing between top and bottom parts may be attributed to differences in perceptual similarity of the specific face stimuli that were used.
Meeting abstract presented at VSS 2016