Abstract
It is widely believed that recognition of face identity is achieved via holistic processing of the representation integrated over parts of a face, whereas processing of facial expressions can be undertaken based on representation of separate parts. In two experiments, we examined how perceptual field may affect processing identity and expression of faces by using gaze-contingent control display in conjunction with an adaptive up-down procedure to adjust the size of perceptual field. In Experiment 1, we created a baseline condition to assess the size of perceptual field required for processing face identity. The results indicated that on average participants needed a perceptual field of about 2.54° of visual arc to process face identity. Moreover, although there was a tendency for participants to evenly distribute their eye movements among various parts of a face, the eyes appeared to play a slightly greater role than nose and mouth for identifying a face. The results of Experiment 2, on the other hand, indicate that (a) in judging whether a face was happy, participants would disproportionally rely upon the mouth region with a minimal size of perceptual field (0.63°), and (b) in judging whether a face was sad, both the eye regions and the mouth were critical to arrive at that judgment, and it required a substantially larger perceptual field (2.26°). These findings were further corroborated from results of temporal analyses of eye movements, showing the dynamic changes in processing different regions of a face for identifying specific facial expression. Taken together, these findings not only provide further evidence to lend support to the notion that processing identity and expression of faces may require different styles of processing (holistic vs. analytic) but also point out in details how judgment of various facial expressions may require different sizes of perceptual field.
Meeting abstract presented at VSS 2018