Neural systems for recognition of familiar faces
Introduction
The recognition of familiar individuals is critical for appropriate social interactions. The capacity to readily access information about a person we encounter determines how we should interact with that particular individual. It is a common experience that our behavior changes rapidly based on whether we are interacting with a friend, a son or daughter, or the boss, and that this “changing of gears” is totally automatic.
In the present manuscript, we propose a model for the distributed neural systems that participate in the recognition of familiar faces, highlighting that this spatially distributed process involves not only visual areas but also areas that primarily have cognitive and social functions other than visual perception. We hypothesize that visual familiarity is only a partial aspect of how we recognize familiar individuals and that person knowledge and emotional responses are also essential requirements for the successful identification of someone we know.
We will focus our attention on different aspects of person knowledge that are activated during face recognition. “Person knowledge” refers to a broad class of information about an individual that encompasses subjective characteristics, such as personal traits, intentions, attitudes, and transient mental states, and objective information, such as biographical facts and episodic memories. In particular, we will highlight the role of a set of areas that have been associated with the performance of tasks involving “theory of mind”. Theory of mind refers to the capacity to represent the mental state of others and to interpret and predict someone else's behavior based on that representation (Gallagher & Frith, 2003; Leslie, 1994). The anterior paracingulate cortex, the posterior superior temporal sulcus (pSTS)/temporoparietal junction (TPJ), and the precuneus have been associated with “theory of mind” tasks (Frith & Frith, 1999). We are proposing a new function of this set of areas: the spontaneous retrieval of some aspects of person knowledge in the act of face recognition.
Finally, we propose that the emotional response to a familiar individual plays a key role in person recognition. The modulation of activity in the amygdala and in the insula based on familiarity supports the hypothesis that these structures play a fundamental role during social interactions and familiarity recognition.
Numerous neuroimaging experiments on familiar face recognition exist in the literature, but the findings in the face-responsive regions of the ventral extrastriate cortex have been inconsistent (Dubois et al., 1999; Gobbini, Leibenluft, Santiago, & Haxby, 2004; Gorno-Tempini et al., 1998; Henson, Shallice, & Dolan, 2000; Leibenluft, Gobbini, Harrison, & Haxby, 2004; Leveroni et al., 2000; Nakamura et al., 2000; Rossion, Schiltz, Robaye, Pirenne, & Crommelinck, 2001; Rotshtein, Henson, Treves, Driver, & Dolan, 2005). In some cases, familiar faces evoked a stronger response (Henson et al., 2000, Leveroni et al., 2000, Rotshtein et al., 2005), in other cases a weaker response (Dubois et al., 1999, Rossion et al., 2001), and in others no modulation at all (Gorno-Tempini et al., 1998, Shah et al., 2001). The inconsistency of these findings could be due to the different types of familiar faces that were used in these experiments. In some experiments, the effect of simple, experimentally learned visual familiarity has been investigated (Dubois et al., 1999, Leveroni et al., 2000, Rossion et al., 2001), in others the familiarity associated with individuals known through the media (Gorno-Tempini et al., 1998; Henson et al., 2000; Leveroni et al., 2000; Sergent, Ohta, & MacDonald, 1992) and in other cases the familiarity associated with personal acquaintances (Bartels & Zeki, 2000; Gobbini et al., 2004, Leibenluft et al., 2004, Nakamura et al., 2000).
In contrast to the inconsistent effect of familiarity on responses in the extrastriate visual areas, anterior temporal regions consistently show stronger responses to a variety of familiar stimuli, including faces (Gorno-Tempini et al., 1998, Leveroni et al., 2000, Nakamura et al., 2000, Rotshtein et al., 2005, Sergent et al., 1992), names (Gorno-Tempini et al., 1998; Grabowski et al., 2001) and landscapes (Nakamura et al., 2000), suggesting a possible role for these anatomical structures in the retrieval of biographical or autobiographical information. Medial temporal regions also respond more strongly to familiar, as compared to unfamiliar, faces and names, probably reflecting the general role of these regions in long-term memory retrieval (Douville et al., 2005; Haist, Bowden Gore, & Mao, 2001; Leveroni et al., 2000).
We argue that visual familiarity and biographical information are only two aspects of the recognition of familiar individuals. We propose that the emotional response that we experience when we meet someone we know and the spontaneous retrieval of information about that person's personality, outlook, and intentions are also integral components of the representation of a familiar individual that play key roles in recognition.
Our hypotheses are based largely on the results of two fMRI experiments that we have conducted to investigate the effect of the type of familiarity that accrues naturally with years of social interactions and long exposure. In the present manuscript, we briefly describe the findings of these two fMRI experiments and the relevant related literature; then we describe the areas we identified and amplify our hypothesis about their role in the representation of familiar individuals. Finally, we propose a modified and updated version of our model for the human face perception neural system (Haxby, Hoffman, & Gobbini, 2000) with a specific focus on the components that play a role in familiar face recognition.
Section snippets
fMRI studies of recognition of personally familiar and famous faces
To investigate the roles of social attachment and emotion in the representation of familiar individuals, we designed two fMRI experiments employing different types of familiar faces. In one fMRI experiment we compared the response to personally familiar faces (faces of friends and family), to famous familiar faces, and to faces of strangers (see Gobbini et al., 2004 for details on the experimental design). In the second fMRI experiment, we studied mothers looking at pictures of their own child,
Retrieval of person knowledge
A key component of the neural representation of a familiar individual concerns information about that person, such as personal traits, intentions, attitudes, transient mental states, biographical information and episodic memory. We use the term “person knowledge” to refer to this broad class of information.
Our hypothesis is that, the “theory of mind” areas encode aspects related to personal traits, intentions and transient mental states (Allison, Puce, & McCarthy, 2000; Mitchell, Heatherton, &
Emotional response
Another key component of the neural representation of familiar individuals that we are emphasizing in our model concerns the emotional response that we experience when we meet someone we know. Simple visual familiarity with faces that are learned in an experimental setting is sufficient to induce a weaker response in the amygdala as compared to faces that are completely new (Dubois et al., 1999, Schwartz et al., 2003). In our fMRI experiments, we detected a weaker response in the amygdala for
Visual familiarity
As mentioned above, the effect of familiarity on the neural response to faces in the ventral occipital temporal cortex has not been consistent across different experiments reported in the literature.
We found that activation in face-responsive regions of the extrastriate cortex did not show a simple monotonic modulation of the response based on the type of familiarity. Comparing personally familiar faces to famous familiar faces showed a stronger response to the more familiar faces (the ones of
New model for face recognition
Recognizing a familiar individual entails recognition of his or her visual appearance, the spontaneous activation of person knowledge and an appropriate emotional response. These components are all integral to successful recognition of a familiar individual. Neuropsychological studies demonstrate that these different components are dissociable and that the failure to access one of these components can lead to impaired recognition.
Bruce and Young's cognitive model on face perception and
Conclusions
Perception of a familiar individual modulates neural activity in a distributed network of areas that extends beyond the visual extrastriate cortex. Spontaneous retrieval of semantic information and information about the personality, attitude and outlook of a familiar individual is tightly linked to the recognition of the visual appearance of his or her face. The appropriate emotional response to a familiar face is also an essential component of successful face recognition. The immediate
References (97)
- et al.
Social perception from visual cues: Role of the STS region
Trends in Cognitive Science
(2000) The primate amygdala and the neurobiology of social behavior: Implications for understanding social anxiety
Biological Psychiatry
(2002)- et al.
The neural correlates of maternal and romantic love
Neuroimage
(2004) Autonomic recognition of names and faces in prosopagnosia: A neuropsychological application of the guilty knowledge test
Neuropsychologia
(1984)- et al.
Response and habituation of the human amygdala during visual processing of facial expression
Neuron
(1996) - et al.
A temporoparietal and prefrontal network for retrieving the spatial context of lifelike events
Neuroimage
(2001) - et al.
Reading the mind from eye gaze
Neuropsychologia
(2002) - et al.
Movement and mind: A functional imaging study of perception and interpretation of complex intentional movement patterns
Neuroimage
(2000) - et al.
Volitional control of autonomic arousal: A functional magnetic resonance study
Neuroimage
(2002) - et al.
Neural systems behind word and concept retrieval
Cognition
(2004)