Prosopagnosia Research Paper

Face recognition impairments are socially disabling, and most patients develop strategies for tackling the problem. They may note carefully what clothes someone is wearing, ask relatives always to wear a particular distinctive item, or become adept at initiating or maintaining a conversation whilst they work out who they are talking to. These strategies are not completely effective. One prosopagnosic patient lost a legal action when he discussed his case with his opponent’s lawyer—he thought it was his own advocate because he was wearing similar court clothes.

Prosopagnosia should not be confused with other types of impairment that can compromise face recognition (Young 1992). In prosopagnosia, recognition from other cues than the face is usually possible. In contrast, cases of inability to recognize people from face, voice, and name have been described—these seem to reflect loss of semantic information about the identities of individuals. There are also reports of problems of name retrieval (anomia), in which familiar people are successfully recognized and appropriate semantic information is accessed, but their names cannot be recalled. Face recognition impairments therefore reflect damage to an underlying system which can break down in different ways.

Although reports of problems affecting face recognition after brain injury can be traced back to the nineteenth century, prosopagnosia was identified as a distinct neuropsychological problem by Bodamer in 1947 (Ellis and Florence 1990). Whilst it is generally considered a rare deficit, there are now several hundred case descriptions in the literature. These include cases where brain injury early or late in life has led to inability to recognize previously familiar faces, and developmental cases where the disorder is present from birth. People with prosopagnosia know when they are looking at a face, and can describe its features, but the loss of any sense of overt recognition is often complete, with no feeling of familiarity. In contrast, other aspects of face processing, such as the ability to interpret facial expressions, or to match views of unfamiliar faces, can remain remarkably intact in some (though not all) cases. The ability to recognize everyday objects other than faces may also remain good, and many prosopagnosic patients are able to read without difficulty. In one of the cases of early acquired deficit, a child who had remained unable to recognize any faces since suffering meningitis and subsequent complications in infancy was nonetheless able to learn to read (Young and Ellis 1989); this underlines the implication that reading and face recognition depend on different types of visual analysis.

Deficits commonly associated with prosopagnosia include a visual field defect in the left upper quadrant, achromatopsia (loss of color vision), and topographical disorders (inability to find one’s way about). These are useful clinical pointers, but they are thought to be due to anatomical proximity of otherwise unrelated processes rather than having any direct functional significance; cases of prosopagnosia without each of these associated deficits have been described, and they have also each been reported in the absence of prosopagnosia.

Studies of prosopagnosia initially concentrated on the location of the lesions responsible, and to what extent the problem is specific to face recognition. More recent work has asked whether there exist different underlying forms of deficit, and has explored the implications of findings of covert recognition.

1. Lesions Causing Prosopagnosia

The underlying pathology involves lesions affecting ventromedial regions of occipitotemporal cortex; these include the lingual, fusiform, and parahippocampal gyri, and more anterior parts of the temporal lobes. Functional imaging studies of normal observers have confirmed the importance of these regions to face perception (Kanwisher et al. 1997). Bilateral lesions are usually present in the relatively small number of cases that have come to autopsy (Damasio et al. 1982), but cases involving unilateral lesions of the right cerebral hemisphere have been reported in several computerized tomography (CT) and magnetic resonance imaging (MRI) studies, and the functional imaging findings with normal people also emphasize the importance of the right hemisphere.

2. A Face-Specific Deficit?

Prosopagnosia involves a severe impairment of face recognition, yet a number of prosopagnosic patients can read and can recognize many everyday objects, so the deficit does not compromise all aspects of visual recognition. But is it only face recognition that is impaired?

In most cases, the answer is clearly ‘no.’ So, for example, recognition of familiar buildings, flowers, types of car, or animal species may all pose difficulties. Like faces, these are all visual categories with a high degree of interitem similarity.

This has led to the hypothesis that what is lost in prosopagnosia is the ability to recognize the individual members of categories that contain several items of similar appearance. Ability to identify the general category to which the items belong (face, building, flower, etc.) is preserved, but within-category recognition is defective. Although attractive, this idea has been undermined by (rare) reports of highly circumscribed impairments. One of De Renzi’s (1986) cases could find his own belongings when they were mixed in with similar objects, identify his own handwriting, pick out a Siamese cat among photographs of other cats, recognize his own car without reading the number plate, and sort domestic coins from foreign coins. In all of these tasks he was therefore able to identify individual members of visual categories with high interitem similarity, yet he could not recognize the faces of relatives and close friends. Just as strikingly, the deficit can affect mainly human faces: when McNeil and Warrington’s (1993) prosopagnosic patient took up farming, he was able to learn to recognize his sheep, and correctly identified several of them from photographs of their faces!

Although they are exceptionally rare, such cases imply that the possibility of face-specific deficits must be taken seriously. However, it may be important to reconsider the underlying issue. The presumption behind the interest in face-specific deficits has been that these might indicate whether the brain has an evolved neural substrate for the important social task of face recognition. But this rationale would only be valid if the organization of the evolved substrate was so rigid that it could not become involved in other important visual discriminations. Studies of the properties of visual cells in the primate brain (Logothetis and Pauls 1995) and acquired expertise in humans (Diamond and Carey 1986) suggest, in contrast, a more plastic organization. It is possible that much of our visual expertise involves extending the sphere of involvement of cells that would otherwise become dedicated to face recognition.

3. Different Forms Of Prosopagnosia

There has been much discussion as to whether prosopagnosia can have more than one different underlying cause. Often, a distinction is made between cases that seem to reflect defective face perception and those that are more like a loss of face memories. When face perception is defective, patients may be poor at telling male from female faces, or matching photographs of unfamiliar faces, whereas patients with a face memory problem may remain able to perform such tasks fairly normally.

For many years, this debate proved inconclusive because it was conceivable that the so-called perceptual cases simply reflected a more severe form of the same impairment. However, studies of ability to form mental images of premorbidly familiar faces have shown that someone suffering from the perceptual form of prosopagnosia can perform better than someone with a loss of face memories (Young et al. 1994). A plausible interpretation is that this happens because we need the same representation to conjure up a mental image of someone’s face that we need to recognize that face. In perceptually based cases of prosopagnosia, these stored representations of familiar faces could remain intact, but a seen face could not access them because it is not perceived sufficiently accurately; imaging faces which cannot be recognized would then remain possible. In memory-based cases, stored memories cannot be used for imagery or for recognition—either because they are themselves damaged or because access to the store is defective.

4. Covert Recognition

The interpretation of some cases of prosopagnosia as a domain-specific impairment of face memory is strengthened by findings of covert recognition, which show clear parallels to priming effects found more generally in amnesia. Prosopagnosic patients usually fail all tests of overt recognition of familiar faces. They cannot name the face, give the person’s occupation or other biographical details, or even state whether a face is that of a familiar person (all faces seem unfamiliar). Surprisingly, however, there is substantial evidence of covert recognition from physiological and behavioral measures (Bruyer 1991).

Bauer (1984) measured skin conductance whilst a prosopagnosic patient, LF, viewed a familiar face and listened to a list of five names. When the name belonged to the face LF was looking at, there was a greater skin conductance change than when someone else’s name was read out. Yet if LF was asked to choose which name in the list was correct for the face, his performance was at chance level. Hence, there was a marked discrepancy between LF’s inability to identify the face overtly and the relatively good recognition shown using the indirect skin conductance measure.

A number of other indices of nonconscious, covert recognition in prosopagnosia have also been developed. Matching of familiar faces is better than for unfamiliar faces, priming has been found from familiar faces onto the recognition of name targets, and learning of correct face-name pairings is better than learning of incorrect pairings (Bruyer 1991). All of these effects can be demonstrated when overt recognition of the faces is at chance level.

Not all prosopagnosic patients show covert recognition, however. As might be expected, when there is evidence of substantial impairment of face perception, covert as well as overt performance can fall to chance level.

There is a parallel between preserved covert recognition abilities and those aspects of recognition that operate automatically for normal people. We cannot look at a familiar face and decide not to recognize it—the mechanisms responsible for visual recognition are not open to conscious control in this way. It seems that some of these automatic aspects of recognition continue to function in some cases of prosopagnosia. A simulation of this pattern can be made by halving some of the connection strengths in a computer model of the architecture of the recognition system (Young and Burton 1999). The network is then no longer able to classify face inputs as familiar, yet it continues to display priming effects. This helps us to understand how covert responses can be preserved when there is no overt discrimination, but it does not solve the more difficult issue of what is involved in being aware of recognizing a face.

Despite the extensive range of covert effects demonstrated in the laboratory, prosopagnosic patients do not act as if they recognize faces in everyday life. Instead, most are acutely conscious of their problems in face recognition. However, some patients lack insight into their disability and show unawareness of impairment (anosognosia). Case SP was very poor at recognizing familiar faces, yet she maintained that she recognized faces ‘as well as before,’ even when directly confronted with her failure to recognize photographs of familiar faces (Young et al. 1990). In contrast, SP showed adequate insight into other physical and cognitive problems produced by her illness—her lack of insight into her face recognition impairment involved a deficit-specific anosognosia. Such deficit-specific anosognosias may reflect impairment to mechanisms used for monitoring performance in everyday life; we can all make occasional errors of recognition, so we constantly monitor our own performance to correct these errors quickly and avoid embarrassment whenever possible.

Remarkably, the deficits of overt recognition noted in prosopagnosia need not be absolute. Sergent and Poncet (1990) found that their patient, PV, could achieve overt recognition of some faces if several members of the same semantic category were presented together. This only happened when PV could determine the category herself; otherwise she continued to fail to recognize the faces overtly even when the occupational category was pointed out to her. Sergent and Poncet (1990) suggested that the simultaneous presentation of several members of the same category may have temporarily raised their activation above the threshold at which recognition could be achieved.

Such findings in prosopagnosia show that the boundary between awareness and lack of awareness is not as completely impassable as it seems to the patients’ everyday experience. However, the circumstances under which this has been found to happen are at present very limited, and it has not yet proved possible to find a way to provide significant remedial assistance in real life.

Bibliography:

1. Bauer R M 1984 Autonomic recognition of names and faces in prosopagnosia: A neuropsychological application of the guilty knowledge test. Neuropsychologia 22: 457–69
2. Bruyer R 1991 Covert face recognition in prosopagnosia: A review. Brain and Cognition 15: 223–35
3. Damasio A R, Damasio H, Van Hoesen G W 1982 Prosopagnosia: Anatomic basis and behavioral mechanisms. Neurology 32: 331–41
4. De Renzi E 1986 Current issues in prosopagnosia. In: Ellis H D, Jeeves M A, Newcombe F, Young A (eds.) Aspects of Face Processing. Martinus Nijhoff, Dordrecht, The Netherlands, pp. 243–52
5. Diamond R, Carey S 1986 Why faces are and are not special: An effect of expertise. Journal of Experimental Psychology: General 115: 107–17
6. Ellis H D, Florence M 1990 Bodamer’s (1947) paper on prosopagnosia. Cognitive Neuropsychology 7: 81–105
7. Kanwisher N, McDermott J, Chun M M 1997 The fusiform face area: A module in human extrastriate cortex specialized for face perception. Journal of Neuroscience 17: 4302–11
8. Logothetis N K, Pauls J 1995 Psychophysical and physiological evidence for viewer-centred object representations in the primate. Cerebral Cortex 5: 270–88
9. McNeil J E, Warrington E K 1993 Prosopagnosia: A face specific disorder. Quarterly Journal of Experimental Psychology, Series A 46: 1–10
10. Sergent J, Poncet M 1990 From covert to overt recognition of faces in a prosopagnosic patient. Brain 113: 989–1004
11. Young A W 1992 Face recognition impairments. Philosophical Transactions of the Royal Society, London, Series B 335: 47–54
12. Young A W, Burton A M 1999 Simulating face recognition: Implications for modelling cognition. Cognitive Neuropsychology 16: 1–48
13. Young A W, De Haan E H F, Newcombe F 1990 Unawareness of impaired face recognition. Brain and Cognition 14: 1–18
14. Young A W, Ellis H D 1989 Childhood prosopagnosia. Brain and Cognition 9: 16–47
15. Young A W, Humphreys G W, Riddoch M J, Hellawell D J, De Haan E H F 1994 Recognition impairments and face imagery. Neuropsychologia 32: 693–702