Face Processing Systems: From Neurons to Real-World Social Perception

Abstract

Primate face processing depends on a distributed network of interlinked face-selective areas composed of face-selective neurons. In both humans and macaques, the network is divided into a ventral stream and a dorsal stream, and the functional similarities of the areas in humans and macaques indicate they are homologous. Neural correlates for face detection, holistic processing, face space, and other key properties of human face processing have been identified at the single neuron level, and studies providing causal evidence have firmly established that face-selective brain areas are central to face processing. These mechanisms give rise to our highly accurate familiar face recognition but also to our error prone performance with unfamiliar faces. This limitation of the face system has important implications for consequential situations such as eyewitness identification and policing.

Spatiotemporal dynamics of similarity-based neural representations of facial identity

“…The observed pattern provides evidence that spatiotemporal patterns of activity in at least some face-selective regions in human cortex encode qualitatively different information about face identity at different times over the first few 100 ms after stimulus onset, with a transition from a lower level representation to a higher level representation occurring around 200-300 ms. This pattern suggests that models of the neural basis of face recognition that assign a single function to each face-selective cortical region (14) are likely to be incomplete, because they do not account for the possibility that a given face-selective region may play different functional roles at different times. Given that the identity-based representation used in the current study represents any exemplar of the same identity identically, whereas the image-based representation does not, this late transition could reflect the temporal emergence of a neural representation with high tolerance to identity-preserving transformations (19).…”

“…For example, within a given region, different information relevant to discrimination may be represented at different times over the first few 100 ms after stimulus onset. However, current models of the neural basis of face recognition in humans do not typically allow for this possibility, because they usually assign a single functional role to each face-selective region (14).A few previous studies (15-18) have explored the temporal properties of the neural representation of individual face identities in humans. However, the measurements of representations in these studies were based on variations in the amplitude of neural activity at just one or two time points, typically sampling from different sensors for different time points (16-18) or on the temporal dynamics of signal from a small number of intracranial electrodes in fusiform gyrus (15).…”

“…For example, within a given region, different information relevant to discrimination may be represented at different times over the first few 100 ms after stimulus onset. However, current models of the neural basis of face recognition in humans do not typically allow for this possibility, because they usually assign a single functional role to each face-selective region (14).…”

New insights on real-world human face recognition

“…People can detect with high accuracy minute alterations to the facial configurations of famous (Ge, Luo, Nishimura, & Lee, 2003) or personally familiar faces (Brédart & Devue, 2006;Devue et al, 2007). But on the other hand, we can fail spectacularly to recognise faces of people we have just recently met, or even to match two views of the same unfamiliar individual Young & Burton, 2017; for reviews on the differences between familiar and unfamiliar face processing see Freiwald, Yovel, & Duchaine, 2016;Johnston & Edmonds, 2009;Natu & O'Toole, 2011).…”

A geometric morphometric evaluation of the Belanglo ‘Angel’ facial approximation

“…A facial approximation is undertaken to stimulate leads to identification from colleagues, friends and family members [6,7], but face recognition studies show that individuals who are unfamiliar with the person depicted utilise different neural mechanisms, attend to different aspects of the head and face [8], and essentially engage in face-matching [for reviews, see 9,10,11]. Face-matching is frequently confounded by variation in head pose even when the images are captured under the same photographic conditions, and the most salient feature is the shape and hue of terminal head hair [8,[12][13][14].…”