F aces give us clues to the identity, race, sex, attractiveness, mood, and inten-tions of others and are therefore probably the most biologically and socially signicant visual stimuli in our human environment. It is no real surprise then that so many of us are fascinated with faces, seeing them in clouds and on Mars! The signicance of faces in our daily lives has led researchers to argue that faces are likely to be processed “automatically” (Öhman, 2002; Öhman & Mineka, 2001). Automatic processes have four main characteristics:
They happen very quickly (e.g., Batty & Taylor, 2003; Öhman, 1997), • although exactly how fast they need to be to be classied as automatic is far from clear (see Compton, 2003, for further discussion). An automatic process should happen, at least in part, unconsciously (e.g., • Bargh, 1997; Öhman, 2002; Robinson, 1998). Automatic processing is mandatory or obligatory (e.g., Wojciulik, • Kanwisher, & Driver, 1998). Face processing that is equivalent under all conditions would provide evidence for mandatory processing in the strict sense. However, a process that always reliably occurs to some extent, even if it is “turned down,” may still be considered mandatory. Processing that is capacity free, requiring minimal attentional resources • (e.g., Schneider & Chein, 2003; Vuilleumier, Armony, Driver, & Dolan,
2001) is considered automatic. If performing another task concurrently does not alter performance on a face-processing task, then that aspect of face processing would be considered capacity free. The nature of the dual task is important, however, because a competing task involving faces may be detrimental to performance, whereas a task involving a different stimulus type may not.