http://www.tandfonline.com/doi/abs/10.1080/17470218.2016.1195851#.V2guKLsrJhE
Face Learning: Successfully recognizing the face of a familiar person requires activation of a stable face representation. Such representations must be sensitive to structural and textural differences between different identities, but tolerant of transient within-person variability in appearance. 

The same person can appear visually different on different occasions, and this variability can sometimes exceed the differences between two people (Adini, Moses & Ullman, 1996). 

The ability to identify a familiar face is thus a remarkable challenge to the visual system, yet familiar observers are able to do so with ease and accuracy. By contrast, recognizing or even matching unfamiliar faces from new instances is surprisingly hard (Bruce et al., 1999; Clutterbuck & Johnston, 2002; Jenkins, White, van Montfort & Burton, 2011). 

Although this remarkable difference in processing familiar and unfamiliar faces has been shown in a number of studies (Bruce et al., 2001; Jenkins et al., 2011), we remain largely unclear about the processes involved in the transition between these two states, i.e., face learning. Specifically, the precise mechanisms of forming representations that allow identification of a person across different instances are largely unknown - Andrews et al (2016). 

To explore how stable face representations develop, in this study we employed incidental learning in the form of a face sorting task. In each trial, multiple images of two facial identities were sorted into two corresponding piles.

Figure 1: Examples of ambient face images from just two identities. 

Following the sort, participants showed evidence of having learnt the faces, performing more accurately on a matching task with seen than unseen identities. Furthermore, ventral temporal event-related potentials were more negative in the N250 time range for previously-seen than previously-unseen identities.

These effects appear to demonstrate some degree of abstraction, rather than simple picture learning, as the neurophysiological and behavioural effects were observed with novel images of the previously-seen identities.

The results provide evidence of the development of facial representations, allowing a window onto natural mechanisms of face learning.