Facial displays are an important form of social communication in nonhuman primates. Clues to the information conveyed by faces are the temporal and spatial characteristics of ocular viewing patterns to facial images. The present study compares viewing patterns of four rhesus monkeys ( Macaca mulatta ) to a set of 1- and 3-sec video segments of conspecific facial displays, which included open-mouth threat, lip-smack, yawn, fear-grimace, and neutral profile. Both static and dynamic video images were used. Static human faces displaying open-mouth threat, smile, and neutral gestures were also presented. Eye position was recorded with a surgically implanted eye-coil. The relative perceptual salience of the eyes, the midface, and the mouth across different expressive gestures was determined by analyzing the number of eye movements associated with each feature during static and dynamic presentations. The results indicate that motion does not significantly affect the viewing patterns to expressive facial displays, and when given a choice, monkeys spend a relatively large amount of time inspecting the face, especially the eyes, as opposed to areas surrounding the face. The expressive nature of the facial display also affected viewing patterns in that threatening and fear-related displays evoked a pattern of viewing that differed from that recorded during the presentation of submissive-related facial displays. From these results we conclude that (1) the most important determinant of the visual inspection patterns of faces is the constellation of physiognomic features and their configuration, but not facial motion, (2) the eyes are generally the most salient facial feature, and (3) the agonistic or affiliative dimension of an expressive facial display can be delineated on the basis of viewing patterns.

The problem of processing visual motion is underconstrained—many possible real world motions are compatible with any given dynamic retinal image. Recent psychophysical and neurophysiological experiments have shown that the primate visual system's normally veridical interpretation of moving patterns is attained through utilization of image segmentation cues unrelated to motion per se. These findings challenge notions of modularity in which it is assumed that the processing of specific scene properties, such as motion, can be studied in isolation from other visual processes. We discuss the implications of these findings with regard to both experimental and computational approaches to the study of visual motion.