Understanding the recognition of facial identity and facial expression

The dominant view in current theories of face perception is that facial identity (recognizing who a person is) and facial expression (interpreting their moods and feelings) are processed by distinct parallel visual routes. Although there is considerable evidence to support the independent coding of identity and expression, it is not clear whether the idea of distinct parallel visual routes provides the best fit to the data.We conclude that there is clear evidence for some separation between the coding of facial identity and expression, the concept of independent visual pathways is not strongly supported. The data are consistent with other potential frameworks that deserve to be more fully explored.One alternative framework derives from image-based analysis of faces using principal component analysis (PCA). This shows that the independent perception of facial identity and facial expression can be modelled within a single representational framework in which some dimensions (principal components) code facial identity, some code facial expressions and others code both. PCA therefore indicates that the dissociation of identity and expression might be partial rather than absolute.We also focus on Haxby and colleagues' observations that facial expressions and other 'changeable' facial cues (such as lipspeech and gaze) are associated with the inferior occipital gyrus and superior temporal sulcus (STS), whereas 'invariant' facial cues (such as facial identity) are associated with the inferior occipital gyrus and lateral fusiform gyrus. This distinction begs more fundamental questions, such as why are facial characteristics divided in this manner and why is the STS more interested in facial expressions, lipspeech and gaze?One potential explanation lies in the fact that the STS is not only sensitive to changeable facial characteristics, but also to other perceptual dimensions that are inherently linked with them (such as their associated vocalizations and dynamic information). There is evidence that the STS might be involved in the integration of these different channels. Consequently, we propose that the prominent role of the STS in coding changeable facial characteristics might reflect an increased reliance on integrative mechanisms for interpreting changeable social signals.In summary, an approach to face perception that emphasizes the different physical properties and information-processing demands (such as reliance on integrative mechanisms) of different facial characteristics has considerable value. This differs from the classic approach, which has tended to emphasize distinctions based mainly on informational content (for example, identity versus expression).