In spite of the astonishing progress we have made in medicine over the past hundred and thirty or so years, there are still those unfortunate times when someone’s limb is so damaged by disease or accident that it must be amputated. The vast majority of people who undergo amputation have a clear sensation that the limb is still there—a phantom limb. So vivid is this impression that many patients, upon regaining consciousness after the operation, do not realize that the limb is gone until they lift the covers and see. Their somatosensory experience of the limb seems to be the same as before. They experience it as being in a certain position, as being hot or cold, as being in pain, and sometimes even as moving in response to their intentions (Ramachandran and Hirstein, 1998).
The arms of phantom limb patients feel the same from the inside, as it were, but obviously do not look the same on the outside. There is another disorder, called asomatognosia, in which the patient’s limb looks the same externally but lacks certain internal sensations. Strangely enough, these patients often claim that the limb is not their own, and that it belongs to someone else. This internal/external dichotomy also applies to the perception of entire human bodies. Some neurological patients can no longer visually recognize people they know, but they continue to attribute the same personalities and other mental traits to them—they cannot recognize them externally, but they recognize them internally, as it were. Just as in the case of limbs, there is a sort of opposite here. Patients with Cotard’s and Capgras’ syndromes recognize familiar people externally, but attribute either no mind at all to them, and hence speak of them as “dead” or “robots” (Cotard’s), or they perceive them to have strange, unknown minds, unrecognizable insides, and hence speak of them as “impostors” (Capgras’).
Things are not the same on the two sides of this internal/external dichotomy, however. While failure in the external domain does not affect the rationality of the patients, failure in the internal domain does, causing them to make strange claims about the identities of their limbs and loved ones. In this paper, I will try to glean insights about how we identify and emotionally respond to human bodies and their limbs, by examining the contrasts between these two classes of disorders. I will formulate some hypotheses about why failures of the internal representation systems affect the patients’ rationality, while failures of the external systems do not.
Certain types of paralysis caused by damage to the parietal cortex of the brain’s right hemisphere are in one respect the opposite of phantom limbs: The sight of the arm is the same, but aspects of the internal, somatosensory experience are missing. Sometimes these patients will actually claim that the paralyzed limbs are not their own, a type of what is classically known as asomatognosia, which means lack of knowledge about the body. Asomatognosia is often seen in conjunction with a more common syndrome, known as anosognosia, which means lack of knowledge, or unawareness of, illness (Babinski, 1914). This unawareness seems to play a role in causing a phenomenon known as denial in some patients. They will calmly but firmly deny that they are paralyzed, or affected in any way by their brain damage. There is evidence that some anosognosic patients have some types of somatosensation (see Davies et al., 2005), but one suspects that some higher-level inner representation of the arm is damaged, specifically, a representation of the arm as something that one can accomplish tasks with. We accomplish tasks with tools, but we move them from the outside, whereas we move our arms and legs from the inside. Our limbs respond not to our actions, as tools do, but to our wills. In the asomatognosic patient, the amputation, performed by nature itself, is a sort of inner amputation. A portion of the will’s domain has been cut off.
The psychologist V. S. Ramachandran performed an experiment in which the subject puts his hands into an opening in the bottom of a large wooden box, and looks down into the box from above. But instead of seeing his own hands, he is actually seeing the hands of another person, reflected via two mirrors to have the position and orientation of his hands. The subject is asked to tap his hands rhythmically, and as long as the other person is able to duplicate the rhythm of the tapping, the subject will believe the hands he sees are his hands. But as soon as the rhythm is broken, the subject experiences an intense startle, and realizes that the hands he is seeing are not his. Apparently some of the brain’s body-representing areas employ the following rule: Something is a part of you if and only if it moves in response to your will. When the asomatognosic patient’s arm does not respond to his will, he concludes that it is not his arm. Thus, what humans consider to be themselves and not themselves depends on how they represent themselves. Another rule, devised by Dennett (1991), to capture these phenomena is, “I am what I control and care for.” The patient can no longer control the limb, and cares nothing for it. This, along with other factors, causes him to believe that it is no longer part of him.
Neurologist Todd Feinberg’s fascinating book, Altered Egos (2001), contains an extensive discussion of asomatognosia and what it might tell us about the human sense of self. In one study he reports, he and his colleagues tested patients with right hemisphere stroke for asomatognosia (Feinberg et al., 1990). Their test simply consisted of holding up the patient’s left hand and asking, “What is this?”
Within the asomatognosia group, the most common misidentification was calling the limb “your hand” or “your arm.” Several of these patients also referred to the limb as “the doctor’s” hand. One patient referred to the limb as “a breast” and a “deodorant.” One patient called it “my mother-in-law’s” hand (1990, 1391).
Asomatognosics may claim that the arm “belongs to a fellow patient previously transported by ambulance, or that it had been forgotten in bed by a previous patient” (Bisiach and Geminiani, 1991). One of Feinberg’s other patients called his left arm “a useless piece of machinery” (2001). Another patient called his arm a piece of “dead wood” (Weinstein, 1991); yet another called his “a piece of dead meat” (Critchley, 1974). As the tone of these responses implies, patients may display hatred of the paralyzed limb, a condition known as misoplegia. The arm involved is almost always the left arm, and asomatognosia also occurs at a high frequency among patients without specific right hemisphere damage whose right hemisphere is temporarily put to sleep. Meador et al. (2000) found that 88% of 62 epileptic patients whose right hemisphere was temporarily deactivated by an amobarbital injection could not recognize their own left hands.
Just as we represent body parts two ways, internally and externally, we also have two ways of representing people, I suggest. The analogue of phantom limb applied to the person might be prosopagnosia. Prosopagnosia is the inability to recognize peoples’ faces. Prosopagnosics typically must wait until familiar people speak before they can recognize them. Or, they may depend on certain clues, such as the type of glasses or tie someone typically wears. For the prosopagnosic, the familiar face is not present, but the person is. There is a significant difference between the sort of representation failure that occurs with phantom limbs and with prosopagnosia, of course. In the prosopagnosic, face representations are damaged or inaccessible, whereas in the phantom limb patients, representations of the missing limb are presumably intact, but not satisfied. Perhaps then, an even closer analogue of phantom limb applied to the person is a case in which the interior, the mind, of a person is somehow there, without a body. Claims about ghosts, imaginary friends, and some conceptions of gods and spirits may be cases of this.
On the other side, the analogue of asomatognosia applied to the person is Cotard’s syndrome. The patient with this syndrome says that he and/or others are “dead,” or “empty.” There is a body present, but no person, rather like a dense neurological version of what psychiatrists call depersonalization. The patient of Wright et al. (1993) reported, “feeling nothing inside.” Another patient “felt that her brain was dead” (Young et al., 1994). In 1788, Charles Bonnet described a patient who, after what appears to have been a stroke, “demanded that the women should dress her in a shroud and place her in a coffin since she was in fact already dead” (Förstl and Beats, 1992).
Just as some patients with asomatognosia will claim that the arm they see is someone else’s arm, some patients will claim that people have actually been replaced by someone else, who looks nearly identical to the original person, a condition known as Capgras’ syndrome. Patients with Capgras’ syndrome claim that people close to them, usually a parent, spouse, or child, have been replaced by impostors. The patient’s attitude toward the alleged impostor can vary. The majority of Capgras’ patients are suspicious of the “impostor” at the very least. Many are paranoid about the “impostor,” and attribute intent to harm to him, but there are also cases where the patient’s attitude is positive. The negative reactions toward previously loved and familiar people are on this view the analogue of the asomatognosic’s misoplegia for her newly unfamiliar and unloved arm.
I have hypothesized that both Cotard’s and Capgras’ syndromes are due to failures of the human mind-reading system (Hirstein, 2005), a set of brain process we use to understand and predict the behavior of others. This failure can be understood in this context as a failure to construct an internal representation of a person. Once we get to know a person well, we develop an individualized internal representation of her mind, on this approach. We also possess generic internal representation that can be applied to people we do not yet know well. Capgras’ syndrome occurs when individual internal representations are damaged or inaccessible and are replaced by generic internal representations. This would help explain why one Capgras’ patient we saw believed his father was an impostor, and referred to him generically as, “that nice Jewish gentleman.” He also tended to categorize other people by religion (Hirstein and Ramachandran, 1997). Apparently, his generic internal representations were organized according to religion.
When both individual and generic internal representations are unusable, Cotard’s syndrome results. The patient can only represent people externally, and this causes him to see them as empty and without minds. This can also explain why some patients, such as that of Wright et al. (1993) alternate between Cotard’s and Capgras’ syndromes: The patient’s specific internal representations are inaccessible, while his generic internal representations are intermittently on and off-line, perhaps due to irregular blood flow.
Let me be clearer about what I mean by speaking of internal and external representations of people and their parts. One can make an initial division, at a higher level, between two types of representations the brain contains, analog, and conceptual. Mental images are a good example of analog representations. Analog representations possess the same significant properties that are possessed by what they represent. Pictures, for instance, mirror the basic shapes, dimensions, and other geometric spatial relations, of that which they depict. Conceptual representations are more abstract. They contain information that does not possess the relevant properties of the represented object. Our concepts themselves often bundle together analog and conceptual representations. My concept of the Eiffel Tower, for instance, contains an analog representation, an image of it, as well as conceptual representations of its being in Paris, and made of steel.
Both internal and external representations are types of analog representations. An internal representation of an agent is a (spatial) analog representation from the unique point of view of that agent, from inside that agent’s head, of processes and events within that agent’s body. An external representation of an agent is a spatial analog representation of the external features of that agent. (The phrase, “internal representation of x” is a bit deceptive, since we tend to think of x externally when we hear the phrase, at least at first. The phrase “X, represented internally,” captures the thought processes involved more accurately.) What I will call our hybrid representation system contains an internal representation system nested within an external representation system. The skin seems to mark the boundary between what the internal and external systems represent. We experience the entities represented by the internal representation system as being inside the skin, whereas the external representation system represents the skin from the outside perspective.
When we observe people, we represent them from our point of view, and from their points of view. When I look at someone, my internal representation of myself is active in my mind, as well as a representation of her internal representation of herself. My hybrid representation contains an internal representation of me, and an external representation of her. But (perhaps using the same processes for simulation purposes) my brain also constructs a representation of her hybrid representation system. This contains an internal representation of her, and an external representation of me. Given that these two hybrid representations may share some resources, it may be that they cannot both be completely active at the same time. Hence the saying, When you speak to someone you hold her prisoner: We cannot fully listen and think our own thoughts simultaneously.
Internal representations primarily involve somatosensory representations, but also involve our awareness of our conscious thoughts and emotions. External representations are primarily visual, but also involve auditory representations, as well as representations deriving from other sense modalities. Many of the existing dichotomies are not adequate to capture what we need here. One such dichotomy is the distinction between egocentric and allocentric representations. There is a sense in which all analog representations are egocentric (and hence both internal and external representations are egocentric), since they always represent their object from a particular point of view (with the possible exception of holograms). While it is true that I can recognize people and objects from many different perspectives, nevertheless, the particular orientation they stand in to me is vital. For instance, whether someone is facing me or turned away is vital in interactions.
We may be seeing both internal and external representation systems at work in the linguistic phenomenon philosophers refer to as opacity. Lois Lane is not aware that the normal man she knows as “Clark Kent” is also the man she knows as “Superman,” who possesses special powers, including the ability to fly:
Lois Lane believes Superman can fly.
Superman is Clark Kent.
Lois Lane believes Clark Kent can fly.
Is the conclusion correct? If we represent Lois Lane and Superman/Clark Kent externally, it is, because Lois Lane’s beliefs are about that man, called by either name. She believes that Clark Kent can fly, whether she knows it or not, we might say. But when we understand Lois internally, it is clear there is a problem. She would never think the thought that would lead her to utter that conclusion. Virtually all uses of the intentional verbs, such as “see,” “love,” “know,” “desire,” etc., involve this sort of internal/external double entendre.
The right hemisphere of the brain seems to contain areas, such as the inferior parietal cortex, that serve the dual function of representing ourselves on certain occasions, and other people on other occasions. One such area is the right supramarginal gyrus, lying at the junction of the parietal and temporal lobes. I suggest that this area, working in concert with interconnected areas in the temporal lobes and the front of the brain, is responsible for the sense of oneself as a being with an inner mental life that produces actions directed out into the world. It represents a mind, with intentionality directed at objects nearby—a mind, enclosed in a body, which is in turn situated in a spatial milieu. Alternatively, when we understand the intentional actions of others, this area serves as an internal representation of them. All of the interconnected areas together function as a hybrid representation system.
Nielsen (1938) was the first to locate the crucial site of damage leading to asomatognosia in a part of the parietal cortex called the right supramarginal gyrus or in its connections to subcortical areas. The study mentioned earlier by Feinberg et al. (1990) involving twelve asomatognosic patients supported this. The Cotard’s patient of Young et al. (1994) had a “temporoparietal contusion” along with bilateral frontal damage. The damage in Capgras’ may be anterior to this area, in portions of the temporal lobes and in the orbitofrontal cortex (Signer, 1994; Mattioli et al., 1999). The idea that the site of damage in Capgras’ might be just anterior to that in Cotard’s also nicely fits a truism of the cognitive neuropsychology of perception, that perception is more generic at the back of the brain, and becomes more specific as processing progresses forward toward the poles of the temporal lobes and the prefrontal cortices.
Damage to these areas area should produce problems in the patient’s hybrid representation of himself, as well as in his hybrid representation of others. Ramachandran and Rogers-Ramachandran (1996) found that patients with anosognosia also claimed that other patients had moved their arms in response to a command, when they had not. Adolphs et al. (1996) found activity in the right parietal cortex during a mind-reading task. This caused them to suggest in a subsequent work that, “in addition to retrieval of visual and somatic records, the difficulty of the particular task we used may make it necessary for subjects to construct a somatosensory representation by internal simulation. Such an internally constructed somatic activity pattern—which could be either covert or overt—would simulate components of the emotional state depicted by the target face. In essence, the brain would reconstruct a state akin to that of the person shown in the stimulus” (Adolphs et al., 2001). Representations of emotional states would be part of the internal representation system; they are a way of representing the mind of the person of interest.
Adolphs et al. (1996) also found that damage to the right inferior parietal cortex can produce a deficit in recognizing facial expressions. Starkstein et al. (1992) found in their large study of anosognosic patients that they had deficits in recognizing facial emotions as well as emotional tonality in voices. They argue that this finding is consistent with the idea that damage to the same areas is associated with anosognosia and flattening of affect. Anosognosic patients also seem to show a failure in what we might call emotion sharing. We are set up to register emotions similar to those we are with, perhaps to synchronize our emotional responses. What the relatives of these patients find so alarming when they visit them in the hospital is the way that they fail to share their concern about their brain damage. Normally this sort of concern is quite infectious.
One phenomenon that is capable of yielding more clues about how to unite our understandings of asomatognosia and the misidentification syndromes such as Capgras’ and Cotard’s is called mirror agnosia. Ramachandran and his colleagues discovered that certain anosognosic patients have lost their ability to interpret mirror images. Their test is simple: A mirror is held in front of the patient. As the patient watches, an assistant holds a pen just above the patient’s left shoulder, so that the pen is visible only in the mirror. The patient is then instructed to grab the pen with his right hand. Amazingly, some patients will repeatedly bump their hands into the mirror, or try to reach around behind it, rather than simply reaching back toward their left to grasp the pen (Ramachandran et al., 1997).
Some Capgras’ patients experience the impostor delusion when they look at themselves in a mirror. The patient of Silva et al. (1995) “developed the delusion that his facial reflection in mirrors and other glass surfaces was no longer his own face but the work of an autonomous evil being. The image of his face was identical to his own face except that the image appeared to be very sinister and aggressive.” Breen et al. (2000) describe a patient with “marked right hemisphere dysfunction” who came to believe that his mirror image was actually another person. He did not believe that the person in the mirror wanted to hurt him, and consistent with the hypothesis described earlier, his internal representation of the man seemed to be generic. When asked what the man in the mirror thought, the patient answered that he hadn’t been able to get him to talk. The patient was not shocked by these events, and impassively noted that the man shaved when he shaved, and dressed as he dressed. When Breen and her colleagues tested the patient for mirror agnosia, he repeatedly bumped his hand into the mirror. This approach described here is capable of explaining the co-presence of misidentifications and mirror agnosia. Understanding how to interpret mirrors requires knowledge of how to move from the default origin of the hybrid representation system, our own minds, out into the world, reflected at the appropriate angle off the mirror. We also need to shift the origin of the hybrid representation system when we observe other people, from our own minds to that of the persons of interest. Damage to the internal representation system compromises both of these abilities.
In the cases we have reviewed, damage to internal representations affects the patients’ emotional responses to their body parts and loved ones, while damage to external representations does not. Prosopagnosia involves damage to external visual representations of people, but it does not affect emotional responses. Its opposites, Cotard’s and Capgras’ syndromes involve changes in emotional reactions.
Internal representations seem to carry more weight in determining identity than external representations. If internal representation fails, even if external representations are intact and satisfied, the patient is likely to misidentify. As one asomatognosic patient said when it was pointed out that the arm she was disowning was connected to her own shoulder, “But my eyes and my feelings don’t agree, and I must believe my feelings” (Nielsen, 1938). Failure of internal representations can cause the patient to doubt whether what he sees is even real. As Cotard himself put it, “Nothing exists any longer, not even themselves” (in Förstl and Beats, 1992). The ability to be rational depends heavily on the ability to identify. A person contradicts herself when she claims that the same thing both has and lacks the same property at the same time. Contradiction is the ultimate enemy of rationality, but contradiction only exists when the same thing is referred to or thought of in both instances. The irrationality of the patients seems traceable entirely to their problem identifying. Since they cannot change their identifications, they alter other parts of their beliefs systems to accommodate this.
Perhaps because of the way it affects our judgments of identity, damage to the internal representation systems can make the person irrational in selected domains. I am of course relying on an intuitive conception of rationality here: It is irrational to claim that a limb that is clearly emerging from one’s shoulder or hip is not one’s own. It is similarly irrational to claim that a person in your company, who looks exactly like your husband, is in fact an impostor of your husband. Asomatognosia, Capgras’, and Cotard’s patients have at least islands of irrationality. Prosopagnosics and phantom limb patients are not irrational in any domain.
Damage to the internal representation systems alters, sometimes quite subtly, the way that the person perceives her body, or other people. But instead of saying that her limbs doesn’t feel the same, or that something is different about her father, which would be the rational thing to do, she claims that the limb is not her limb, and that her father is not her father. When the internal representation system is intact, the patient can use reason to overrule mere appearances: Even though it feels as if my arm is still there, it isn’t (phantom limb), even though that person looks like a stranger, he isn’t (prosopagnosia). But when the internal representation system is damaged, what merely seems to be the case has the power to become reality. That arm does not feel like my arm, therefore it isn’t (asomatognosia). That man does not seem to be my father, therefore he isn’t (Capgras’, Cotard’s). Given its use in simulating the perspectives of others, perhaps the explanation for this is that the hybrid representation system allows us to envisage alternative realities, alternative perspectives, without believing that they represent reality. In the case of language, as in the Lois Lane case above, such a system (or perhaps its left-hemisphere analogue) gives us the power to imagine the world from Lois’s point of view, using the names and concepts she would use.
Perhaps another factor in producing the irrationality is due to the way that the internal representation system has the power to overrule any external representations system, as it does in the Lois Lane case. We might initially consent to the conclusion, but with a few seconds of thought, we realize Lois would not have that belief. This positions the internal representation system as a second line of defense against candidate beliefs that are false because of malfunctions in the external representation systems. But it is the second and final line of defense, so that when it fails, irrationality and its errors can manifest. One well-documented manifestation of these is confabulation, which I have argued is a tendency to confidently make ill-grounded claims that one should normally know are ill-grounded. The claims by anosognosics, and particularly asomatognosics about their limbs, as well as the claims by misidentification patients about people, are classically considered to be confabulations.
Even though the person with a phantom limb has lost something of great value, the limb itself, she retains something of even greater value, the internal representation of the limb. This allows her to maintain a rational attitude toward her predicament. More practically, the intactness of the internal representation system is an important factor in the success of prosthetic devices. The patient comes to accept the device as part of herself because the internal representation of her former arm successfully inhabits it. The prosopagnosic has lost the ability to identify people by their faces, but he has not lost the people themselves. He retains his representations of their minds, and these minds can re-inhabit the visually unfamiliar people around him. But the patients suffering from Capgras’ and Cotard’s syndromes have in a very real sense lost their loved ones, the ones they need most at this difficult time. They are surrounded instead by strange or empty people, some of whom mimic or menace them from behind mirrors. And the asomatognosic has an alien, unwanted appendage, someone else’s property, dangling from her own shoulder. I have argued here that these strange symptoms are not the result of deep and mysterious psychiatric conflicts, as had been thought in the past, but rather of the way our systems of representation are organized, and the way they respond to different types of damage.
1. Adolphs, R., Damasio, H., Tranel, D., and Damasio, A. R. 1996. Cortical systems for the recognition of emotion in facial expression. Journal of Neuroscience 16: 7678-7687.
2. Adolphs, R., Tranel, D., and Damasio, A. R. 2001. Neural systems subserving emotion: Lesion studies of the amygdala, somatosensory cortices, and ventromedial prefrontal cortices. In Handbook of Neurophysiology, ed. G. Gainotti. New York: Elsevier.
3. Babinski, J. 1914. Contribution a l’etude des troubles mentaux dans l’hemiplegie organique cerebrale (anosognosie). Revue Neurologique (Paris) 27: 845-848.
4. Bisiach, E., and Geminiani, G. 1991. Anosognosia related to hemiplegia and hemianopia. In Awareness of Deficit After Brain Injury: Clinical and Theoretical Issues, eds. G. P. Prigatano and D. L. Schacter. Oxford: Oxford University Press.
5. Breen, N., Caine, D., Coltheart, M., Hendy, J., and Roberts, C. 2000. Towards an understanding of delusions of misidentification: Four case studies. In Pathologies of Belief, eds. M. Coltheart and M. Davies. Oxford: Blackwell.
6. Critchley, M. 1974. Misoplegia or hatred of hemiplegia. Mt. Sinai Journal of Medicine 41: 82-87.
7. Davies, M., Davies, A A., and Coltheart, M. 2005. Anosognosia and the two-factor theory of delusions. Mind & Language 20: 209-236.
8. Dennett, D. 1991. Consciousness Explained. Boston: Little, Brown and Company.
9. Feinberg, T. E., Haber, L. D., and Leeds, N. E. 1990. Verbal asomatognosia. Neurology 40: 1391-1394.
10. Feinberg, T. E., Roane, D. M., Kwan, P. C., Schindler, R. J., and Haber, L. D. 1994. Anosognosia and visuoverbal confabulation. Archives of Neurology 31: 468-473.
11. Feinberg, T. E., and Roane, D. M. 1997. Anosognosia, completion and confabulation: The neutral-personal dichotomy. Neurocase 3: 73-85.
12. Feinberg, T. E. 2001. Altered Egos: How the Brain Creates the Self. Oxford: Oxford University Press.
13. Förstl, H., and Beats, B. 1992. Charles Bonnet’s description of Cotard’s delusion and reduplicative paramnesia in an elderly patient (1788). British Journal of Psychiatry 160: 416-418.
14. Hirstein, W., and Ramachandran, V. S. 1997. Capgras syndrome: A novel probe for understanding the neural representation of the identity and familiarity of persons. Proceedings of the Royal Society of London, Series B 264: 437-444.
15. Hirstein, W. 2005. Brain Fiction: Self-Deception and the Riddle of Confabulation. Cambridge, Mass.: The M. I. T. Press.
16. Mattioli, F., Miozzo, A., and Vignolo, L.A. 1999. Confabulation and delusional misidentification: A four year follow-up study. Cortex 35: 413-422.
17. Meador, K. J., Loring, D. W., Feinberg, T. E., Lee, G. P., and Nichols, M. E. 2000. Anosognosia and asomatognosia during intracarotid amobarbital inactivation. Neurology 55: 816-820.
18. Nielsen, J. M. 1938. Gerstmann syndrome; finger agnosia, agraphia, confusion of right and left, acalculia; comparison of this syndrome with disturbances of body scheme resulting from lesions of right side of brain. Archives of Neurological Psychiatry 39: 536-560.
19. Ramachandran, V. S., and Rogers-Ramachandran, D. 1996. Denial of disabilities in anosognosia. Nature 382: 501.
20. Ramachandran, V. S., Altschuler, E. L., and Hillyer, S. 1997. Mirror agnosia. Proceedings of the Royal Society of London, Series B 264: 645-647.
21. Ramachandran, V. S., and Hirstein, W. 1998. The perception of phantom limbs: The D. O. Hebb lecture. Brain 121: 1603-1630.
22. Signer, S. F. 1994. Localization and lateralization in the delusion of substitution. Psychopathology 27: 168-176.
23. Silva, J. A., Leong, G. B., Wine, D. B., and Saab, S. 1992. Evolving misidentification syndrome and facial recognition deficits. Canadian Journal of Psychiatry 37: 239-241.
24. Starkstein, S. E., Federoff, J. P., Price, T. R., Leiguarda, R., and Robinson, R. G. 1992. Anosognosia in patients with cerebrovascular lesions. Stroke 23: 1446-1453.
25. Weinstein, E. A. 1991. Anosognosia and denial of illness. In Awareness of Deficit After Brain Injury: Clinical and Theoretical Issues, eds. G. P. Prigatano and D. L. Schacter. Oxford: Oxford University Press.
26. Wright, S., Young, A. W., Hellawell, D. J. 1993. Sequential Cotard and Capgras delusions. British Journal of Clinical Psychology 32: 345-349.