Face Recognition: Different Encoding Methods on Newborn Infant Research

Many studies have investigated cognitive and social competences of newborn infants as face recognition, ability to imitate facial gestures and communicative skills. The preferential looking paradigm showed that newborns prefer to look at a human face-like stimulus compared to any other non human face-like stimulus. Furthermore, it is highlighted a newborn preference to look at a new face compared to a face previously seen from the habituation procedure (novelty effect); to explain this effect, it is assumed that the newborns are able to build a perceptive representation of the face that they looked at, like an expression of a motivation for novelty. This effect disappeared if the known face is the mother’s face which is looked at more when it is compared to a new face (familiarity effect); it is assumed that the newborns are able to build a dynamic “social” representation of the face through few communicative interactions with their mother in the first hours of life. The studies about the novelty effect and the familiarity effect are based on a preference-task, where a new face is compared with a known face. These studies show a large heterogeneity of the encoding methods related to the gazing behaviour of the newborns. The most widely used methods in literature are summarized in three big classes: live encoding, frame by frame video encoding, real time video encoding. This chapter will focus on the differences among encoding methods of the gazing behaviour of the newborns. Moreover the findings of an empirical study on the comparisons among different encoding methods will be reported.