Facial Motion as a Cue to Identity
THE HUMAN FACE provides the most reliable and accessible means of person identification 1—more reliable than other sources such as voice or gait (see later this chapter). Each of us can recognize thousands of different faces, an amazing feat when we consider how similar human faces are. All human faces conform to the same basic configuration of eyes above nose above mouth, constrained by mechanical functions such as eating and breathing. The human face also provides the viewer with a wide range of other social signals. From a person's expression, we infer the emotional state of a person. We are able to categorize expressions into a number of distinct categories (see Ekman, 1982), and movements of the lips and tongue help distinguish different speech sounds (for example see, Campbell, 1986). Direction of head and eye gaze can signal the focus of another's attention (Kleinke, 1986; Langton, 2000), and regulate turn-taking during conversation.
Until recently, most investigations of face recognition have utilized static images of faces (for example, photographs). However, the majority of our experience with faces is not with static images but with moving faces. We encounter moving faces directly, by interacting with people on a daily basis, and indirectly via images shown on our television screens. When we yawn, when we laugh, when we talk, and when we smile, our face moves in a complex manner. Faces move in both rigid and nonrigid ways. During rigid motion, the face maintains its three-dimensional form while the whole head changes its relative position and/or orientation. In nonrigid motion, individual parts of the face move in relation to one other, for example, during the formation of expressions or articulation of speech. Typically, a complex combination of both rigid and nonrigid motion is required for everyday interaction, with head and face movements superimposed onto larger body movements (Munhall & Vatikiotis-Bateson, 1998). This chapter explores whether such patterns of facial motion are important in the recognition of identity. This research area has been largely neglected, as typically we are highly skilled at recognizing known people from the spatial information available in a static face image (for a review, see Bruce, 1994). However, it has become increasingly important to gain an understanding of how human observers process moving faces, from both an applied and theoretical perspective.
First, in terms of application, facial animation has become a developing computer technology, highly important in the games and entertainment industry (see Parke & Waters, 1996). Since the earliest cartoon films, film producers have attempted to make the faces of their characters move in realistic ways. However, the movements made by cartoon characters do not need to be fully realistic to be convincing and entertaining. But as film production increasingly involves the use of simulated humans, as in the 1997 film Titanic, so companies want their simulants to have animated human faces (a more