Developmental prosopagnosia: a window to content-specific face processing
Introduction
In a landmark paper, Bodamer [1] coined the term ‘prosopagnosia’ to describe the selective degradation of face perception and face memory in three individuals suffering from brain damage. Since the work of Bodamer, a small yet steady trickle of reported cases of acquired prosopagnosia has sustained the notion that human recognition of faces is carried out by content-specific (also called domain-specific) mechanisms that are specialized for face processing. Support for this idea also comes from an extraordinary condition that displays this is the opposite pattern from prosopagnosia, that is, object agnosia with normal face recognition [2, 3]. The existence of face-specific processing mechanisms is consistent both with the presence of cells selectively tuned to faces in the temporal lobes of monkeys [4••] and humans [5] and with findings that specific regions in the human temporal lobes become active when presented with facial images [6].
Prosopagnosia can be a devastating loss to patients, because face recognition is an integral part of our social lives. Its onset is usually dramatic, because the afflicted person becomes painfully aware that something so effortlessly achieved in the past is no longer possible.
Although there are a number of documented cases of individuals who acquire prosopagnosia during their lifetime, it appears that the number who have developmental prosopagnosia (DP), that is, individuals who never acquired the ability to recognize faces, might be much larger. Developmental prosopagnosics (DPs) have not suffered any obvious brain damage, yet they have deficits in face recognition that can be as severe and as selective as those seen in acquired prosopagnosics. Until recently, it appeared to be a very rare condition, but a number of research groups have tested DPs in recent years [7, 8, 9•, 10], and our website has been contacted by more than 500 self-diagnosed DPs (http://www.faceblind.org).
For example, JK, a woman in her early 30s who has recently completed her Ph.D., discusses a recent episode. “This week I went to the wrong baby at my son's daycare and only realized that he was not my son when the entire daycare staff looked at me in horrified disbelief”. Social embarrassment and consequent social isolation are common. “A few years ago I introduced myself to a woman at a wedding only to find that I knew her quite well... she worked in the lab down the hall and we frequently had lunch and chatted together in the lounge. She, incidentally, has never spoken to me again”. Interestingly, JK's sister and father also have face recognition deficits.
Despite these dramatic difficulties associated with developmental prosopagnosia, the existence of and the nature of their impairment is often not evident to those with DP. Having no obvious standards of comparison, those who become aware of their deficits do so gradually in adolescence and adulthood. In addition, many cope as best as they can by relying on hairstyles, clothing, context and voices.
In this review, we only discuss individuals who have no apparent brain damage. We do not discuss individuals who suffered brain damage as children, and who are sometimes categorized as developmental prosopagnosics [11, 12]. We expect that this group of subjects will be an important source of information about face recognition and its development, but they are likely to have a different profile to those prosopagnosics without brain damage.
Because a number of research groups are now working with DPs, there has been a substantial increase in the number of papers on DP. They have addressed the cognitive, neural and developmental basis of the condition. Here, we discuss these results and their contribution to our understanding of DP and face processing more generally.
Section snippets
Developmental prosopagnosia and inferences to functional organization
During the past several years, DP has been investigated at a number of different levels, but the most fruitful work has addressed the cognitive basis of the disorder. A correct cognitive description of DP is undoubtedly a crucial step in our understanding of the condition, but we believe it will also contribute importantly to theories of normal face processing. Although some have argued that developmental disorders are unlikely to shed light on normal functional organization [13], we believe
Demonstrating that prosopagnosia and agnosia are separate conditions
Prosopagnosia is characterized by impaired face recognition, but it is often accompanied by object agnosia [9•, 19]. Prosopagnosia and agnosia could stem from deficits to the same mechanisms, and if this were the case, there would be no theoretical reason to differentiate between prosopagnosia and agnosia. Gauthier, Behrmann, and Tarr [20] were the most recent to take this position. They pointed out that nearly all past reports of dissociations between impaired face and normal object
Impairments in developmental prosopagnosia: face-specific or general purpose?
The previous results demonstrate that deficits to an assumed unitary visual recognition system cannot account for developmental prosopagnosia. More recent work has focused on the mechanisms that failed to develop in DP. In particular, these studies examined whether faces are handled by face-specific mechanisms or mechanisms used with a wider range of object classes, such as mechanisms for processing the precise spatial configuration of the features of an object or processing of object classes
Investigating the architecture of face processing through developmental prosopagnosia
Research into the cognitive and neural basis of developmental prosopagnosia has benefited from the models of normal face processing developed during the past twenty years. In turn, we think developmental prosopagnosia provides an opportunity to corroborate and further refine these models, because DPs report considerable variability within face processing. The classic model of Bruce and Young [32] proposes that a number of independent modules are involved in face processing, with some operating
Neural basis of developmental prosopagnosia
The neural basis of developmental prosopagnosia has recently been investigated with both fMRI and MEG. It is now well established that neurotypical subjects have an area in fusiform gyrus that responds much more strongly to faces than it does to other objects. This area is often called the fusiform face area (FFA), and Avidan et al. [44•] and Hasson et al. [45] used fMRI to investigate whether DPs showed an atypical FFA. Surprisingly, they found no difference in the FFA between the DPs and the
Developmental course and the genetic basis of developmental prosopagnosia
Currently, little is known about the developmental course of DP. The few studies conducted in children with DP suggest that the deficit is apparent early in life [42]. Our laboratory has been contacted by many parents who are concerned that their young children might have prosopagnosia, and many adult DPs report that their difficulties with faces started very early on in childhood.
De Gelder and Stekelenburg [8] recently investigated whether DP might originate in defective subcortical
Conclusions
Recent work with DPs has made significant advances. We now have much stronger evidence than before that DP is a disorder that sometimes results from deficits with content-specific mechanisms specialized for processing faces. Qualitative differences in the cognitive and neural basis of DP in different DPs have shown that the condition has multiple types. However, work on DP has only begun to scratch the surface, and DP appears likely to provide insight into a number of key cognitive, neural,
References and recommended reading
Papers of particular interest, published within the annual period of review, have been highlighted as:
• of special interest
•• of outstanding interest
Acknowledgments
This work was supported by a grant to K Nakayama from the National Institutes of Health (R01 EY13602). N Kanwisher, P Glimcher, and L Garrido provided valuable suggestions.
References (52)
- et al.
Dissociations of face and object recognition in developmental prosopagnosia
J Cogn Neurosci
(2005) - et al.
Less impairment in face imagery than face perception in early prosopagnosia
Neuropsychologia
(2003) - et al.
Developmental disorders of vision
Neurol Clin
(2003) - et al.
Age-related cortical gray matter reductions in non-demented down's syndrome adults determined by MRI and voxel-based morphometry
Brain
(2004) - et al.
Theories of developmental dyslexia: insights from a multiple case study of dyslexic adults
Brain
(2003) - et al.
Can face recognition really be dissociated from object recognition?
J Cogn Neurosci
(1999) Looking at upside-down faces
J Exp Psychol
(1969)- et al.
FFA: a flexible fusiform area for subordinate-level visual processing automatized by expertise
Nat Neurosci
(2000) - Yovel G, Duchaine B: Specialized face perception mechanisms extract both part and spacing information from faces:...
- et al.
Training “greeble” experts: a framework for studying expert object recognition processes
Vision Res
(1998)