Face perception is mediated by a distributed cortical network
Introduction
Face perception is a highly developed visual skill in primates. Converging evidence from neuropsychology, neurophysiology, and cognitive development indicates that face perception may be mediated by a specialized neural system in the brain [9]. In the past decade, with the advent of functional brain imaging techniques, it has been shown that face perception consistently activates a region in the lateral fusiform gyrus [17], [30], [33], [46]. It has been proposed that the ‘fusiform face area’ is a module specialized for face perception [30]. This region, however, also responds significantly to other categories of objects (e.g., houses, chairs, and tools), thus ruling out its status as a face “module” [4], [21], [27], [28]. Furthermore, it has been shown that neural activation within the fusiform face-selective region is modulated by expertise [12], [13], attention [37], visual imagery [24], [26], [36], and emotion [51]. These findings suggest that the response to faces in extrastriate cortex is not the result of a mere hierarchical, bottom-up, ‘feature’ analysis, but is modulated by top-down effects, likely originating in parietal and frontal regions.
The recognition of identity is based on invariant facial features, while changeable aspects of the face, such as speech-related movement and expression, contribute to social communication. When looking at faces, we automatically perceive the gender, expression, and mood. Processing the information gleaned from the faces of others therefore requires the integration of activation within a network of cortical regions. Numerous face perception studies have reported activation not only in the fusiform gyrus, but also in other regions in the visual cortex, limbic system, and prefrontal cortex [24], [25]. Haxby et al. have suggested that face perception is mediated by a distributed neural network in the human brain and have proposed a new model that includes a ‘core’ system and an ‘extended’ system [20]. The core system is comprised of the inferior occipital gyrus (IOG), fusiform gyrus (FG), and the superior temporal sulcus (STS). The ventral regions (IOG and FG) mediate the recognition of individuals. The more dorsal regions (STS) participate in the perception of social signals such as the direction of gaze, speech-related lip movements, and facial expressions [22], [41]. The extended system includes the amygdala and the insula, which mediate the perception of emotional facial expressions, in particular fear, anger, and disgust [2], [25], [34], [39]. Additionally, it has been shown that assessment of facial attractiveness evoked activation in prefrontal cortex [35] and the reward circuitry, in particular the amygdala and nucleus accumbens [1]. It is important to note, however, that not a single study has shown activation within all regions of both the core and the extended systems.
The aim of this study was to test whether activation in regions of the distributed neural system associated with face perception could be localized with a simple task, namely passive viewing of various face stimuli. Previous studies focused on activation within the fusiform gyrus [16], or compared activation evoked by specific face stimuli, such as familiar versus unfamiliar faces [44]. We adopted a more general approach, instructing subjects to view different face stimuli (line drawings of unfamiliar faces and photographs of unfamiliar, famous, and emotional faces) and analyzing all face-responsive regions in the brain. Our results indicate that viewing faces evokes activation in a distributed network that includes multiple, bilateral regions in the visual cortex, limbic system, and prefrontal cortex. Within this network, the response to famous and emotional faces is stronger than the response to unfamiliar faces.
Section snippets
Subjects
Thirteen healthy, drug-free volunteers (eight males, five females, mean age 26 ± 4 years) with normal vision participated in the study. All subjects gave written informed consent prior to the examination.
Stimuli and task
Subjects were presented with four different types of face stimuli: black and white line drawings of unfamiliar faces, and gray scale photographs of unfamiliar, famous, and emotional faces. Phase scrambled versions of these faces were used as visual baseline. The scrambled pictures were generated
Activation evoked by visual perception of faces
Visual perception of faces, as compared with scrambled faces, significantly activated the IOG, FG, STS, amygdala, hippocampus, IFG, and the OFC (Fig. 1). Within these face-responsive regions, bilateral activation was found in all subjects (see Table 1 for cluster size and brain atlas coordinates), however stronger responses were observed in the right hemisphere. In the IOG, FG, STS, and IFG this hemispheric asymmetry was manifested by larger clusters of activation (P < 0.0001) and higher t-values
Discussion
The neural response evoked by passive viewing of faces was investigated using fMRI. Activation was found in a network of face-responsive regions, including the IOG, FG, STS, amygdala, hippocampus, IFG, and OFC. Within these regions, all stimuli (line drawings of unfamiliar faces and photographs of unfamiliar, famous, and emotional faces) evoked significant activation, with stronger responses in the right hemisphere. Furthermore, the response to famous and emotional faces was stronger than the
Acknowledgements
We thank Victor Candia, Daniel Kiper, Kevan Martin, and Elena Yago for their comments on previous drafts, Philips Medical systems for their continuous support, and the ETH Zurich program of Strategic Excellence Projects (TH 7/02-02).
References (52)
- et al.
Response and habituation of the human amygdala during visual processing of facial expression
Neuron
(1996) - et al.
Association of the distinct visual representations of faces and names: a PET activation study
NeuroImage
(2001) - et al.
Effect of familiarity on the processing of human faces
NeuroImage
(1999) Is face recognition ‘special’? Evidence from Neuropsychology
Behav. Brain Res.
(1996)- et al.
Analysis of fMRI time-series revisited
NeuroImage
(1995) - et al.
The effect of face inversion on activity in human neural systems for face and object perception
Neuron
(1999) - et al.
The distributed human neural system for face perception
Trends Cogn. Sci.
(2000) - et al.
Face and voice expression identification in patients with emotional and behavioural changes following ventral frontal lobe damage
Neuropsychologia
(1996) - et al.
Visual imagery of famous faces: effects of memory and attention revealed by fMRI
NeuroImage
(2002) - et al.
Distributed Neural Systems for the Generation of Visual Images
Neuron
(2000)