The self in neuroscience and psychiatry
http://www.100md.com
《神经病学神经外科学杂志》
Edited by Tilo Kircher, Anthony David. Cambridge: Published by Cambridge University Press, 2003, pp 466. ￡30.00 (paperback), ISBN 0-521-53350-3
This is a thought provoking book. Tilo Kircher and Anthony David have persuaded 22 authors to write on aspects of the self. The title might be "The self and schizophrenia" for this is the central focus. In his chapter Parnas makes this explicit:
"The most fundamental level of selfhood that appears to be affected in early schizophrenia is the automatic pre-reflective articulation of the first person perspective" (page 217)
and quotes Bleuler:
"Ganz intakt ist dennoch das Ich nirgends"
"The I is, however, never completely intact" to this effect. The proposal is that by attacking the self as a concept and dealing with its scientific and philosophical implications one can arrive at a more fundamental understanding of the nature of schizophrenia than Bleuler’s concept that it is a disorder of association. However, the implications are wider than this.
To be sure there are some disagreements and differences in emphasis between the contributors. Thus Berrios and Marková in a characteristically thorough historical survey conclude that the self is "fundamentally a western construct" to be explained by means other than neurobiology. Bentall writes:
"To many hard-nosed biological scientists, the self might seem an almost metaphysical concept" (page 301)
and quotes Baumeister:
"Providing a satisfactory definition of the self has proven fiendishly difficult. It is what you mean when you say "I". Most people use "I" and "self" many times each day, and so most people have a secure understanding of what the self is – but articulating that understanding is not easy".
Phillips writes:
"Sartre and Heidegger come to mind as philosophers who reject the notion that there is anything like a human nature or natural self-development ..." (page 320)
These quotations provide a background of cautious scepticism but there is much in this volume to convince the reader that addressing the problem of the self and its relationship to the phenomena of psychosis takes one into the core problem of the neural basis of psychotic symptoms and indeed the nature of human consciousness.
For example, there is the question of the relationship between the self and the central executive—a key concept in "cognitive psychology". Blakemore and Frith (chapter 19) outline the theory that auditory hallucinations and delusions of control arise because of a failure in the mechanism by which the predicted consequences of self-produced actions are derived from an internal "forward" model that predicts and cancels the sensory consequences of self-produced actions. This widely discussed model has a nice wiring diagram (figure 19.1) according to which a patient lacks awareness of initiating an action or thought or of its predicted consequences. However, as Zahavi argues on pp 67–68 there is a problem. The account is based on the assumption that the "mineness" of experience comes about as the result of high order representation, and that such meta-representational accounts of conscious experience are flawed in that they depend on an infinite regress—that is, they require the invocation of an "homunculus" whose function in turn needs to be accounted for by a further homunculus at a higher level. Similar criticisms apply to theory of mind formulations discussed by Kai Vogeley in chapter 17. A more parsimonious account of the differences between cortical function in other primates, mammals, and humans is required.
Where then is the neural basis of the self? According to Markowitsch (chapter 9) it is located in the right pre-frontal and anterior temporal cortex on the basis of evidence relating to the hemispheric encoding and retrieval asymmetry (HERA) hypothesis of episodic memory, according to which left pre-frontal cortex is engaged in recording episodic memories while right pre-frontal cortex is relevant to retrieval. Keenan et al (chapter 8) agree on the basis of studies that implicate right hemisphere/pre-frontal cortex in self-recognition and self-face recognition as well as episodic memory. These chapters therefore introduce the topic of cerebral asymmetry; but do not tell us why it is relevant.
I missed reference to what I have called the Broca-Annett axiom—the concept that cerebral asymmetry is the characteristic that defines the human brain.1,2 This is the notion that at some point in hominid evolution the brain lateralised and that this innovation was critical to the evolution of language. On page 198 of this volume Panksepp refers to the problems of animal models of schizophrenia:
"Not only is it risky to infer mental processes simply from the study of behavioural responses, but some of the key symptoms of schizophrenia (eg thought disorder) may be impossible to model in creatures of modest cortical endowment".
If, as Panksepp also suspects may be the case, the disease is specific to the human capacity for language, the problem is deeper than this. Miskolczy3 considered that schizophrenia was a disease of Homo sapiens-specific cerebral cortex; the Broca-Annett axiom is a lead to the nature of its anatomical basis. Gallup et al (page 147) place the problem of self-awareness in the context of primate evolution by considering the point in the primate lineage at which evidence of self-recognition emerges in mirror experiments. They equate this point with the development of the capacity for a "theory of mind and appreciation of the existence of others" and relate this capacity to "enlargement of the frontal lobes and vulnerability to the phenomena of psychosis". However, theory of mind interpretations of mirror recognition have been contested.4 Susceptibility to schizophrenia may be more precisely related to the capacity for language and the origin of the species5, and cerebral correlates of this transition may be more recent and more precisely defined. Tomasello6 has suggested that the correlate in human ontogeny of the capacity for language is the development of the ability to share with another individual a focus of attention in the outside world—he asks "why don’t chimpanzees point?" This is a plausible correlate of cerebral lateralisation and yields a radical alternative to frontal lobe explanations of the origin of psychotic symptoms.
One can ask further questions about how the phenomena of psychosis considered as deviations in the integrity of the self match up with what we know about the epidemiology and biology of schizophrenia. From the evidence of the WHO and other studies7 schizophrenic disorders have an approximately uniform incidence in all populations. Onsets occur, with a characteristic sex difference, in the reproductive phase of life and are associated with structural changes in the brain as reflected in a degree of ventricular enlargement.8 Thus we are dealing with variation, associated with variations in brain structure, that is maintained across the species in the face of biological disadvantage. On page 236 Parnas writes that:
"...disruption in the ontogenesis of corticocortical connectivity may represent a crucial aetiological element in the origins of schizophrenia ... Development of the elemental self-object relation is ... most likely heavily dependent on a progressive sophistication of intra- and intermodal sensory and sensorimotor integrations. An intact and very precise intracortical and corticocortical connectivity is a necessary condition for such developments".
On page 375 Vogeley writes of "converging evidence for schizophrenia as a disconnectivity syndrome". To be sure much evidence is consistent with schizophrenia as a misconnectivity syndrome and this implies a uniformity of pathogenesis incompatible with a polygenic theory. The key question is where are the misconnections? If schizophrenia is a Homo sapiens-specific condition associated with disorders of self-concept that emerged with the evolution of language, the anatomical basis of these developments is the central issue. There appears to be no alternative to the Broca-Annett axiom that the brain lateralised in course of hominid evolution, and that the final step was the advent of the capacity for language. The critical connections are those between the hemispheres and this must surely be where the misconnectivity occurs.9 It is of particular interest that development of the corpus callosum continues into the third and fourth decades of life10 and is associated with a sex difference (continuing later in females) that is a possible correlate of the sex difference in age of onset of psychosis.
It is a remarkable but commonplace fact that psychotic symptoms, including those relating to the core of the self, sometimes respond to neuroleptic medication. In their contribution to "The Self" Walter and Spitzer consider the implications of this observation. They also point out the awkwardness of having two apparently unrelated explanations of ichst?rungen—"disorders of the I"—as internal monitoring deficits in the case of passivity phenomena on the one hand and, in terms of the theory of mind, for experiences of reference, on the other. This is unparsimonious Walter and Spitzer point out. Elsewhere they11 have elegantly demonstrated that hypofrontality in schizophrenia is not a general deficit—there is a failure of lateralisation, verbal activation to the left, and spatial activation to the right. What is required is some integration of dopaminergic and laterality hypotheses. Here the observations of Hsiao et al12 are relevant. These authors have shown that dopamine uptake (or possibly turnover) in the striatum is not grossly changed in individuals with schizophrenic disorders, but whereas in controls there is marked lateralisation, in patients this is lost. If the torque is the defining characteristic of the human cerebral cortex, corticostriatal influences could be central to an understanding of the nature of the disturbance in psychosis and its response to medication.
Perhaps the most distinctive contribution to this volume is by Maxim Stamenov, who places the concept of the self within the framework of linguistic theory. He gives an account of the symbolic representation in language of the first person personal pronoun—"this I is the most unusual symbol humans have at their disposal". He explains that there are two aspects of the function of the I—it refers in one direction to the implicit central executive, while it gets its meaning from another world, the external world of current explicit cognitive representation. These requirements are incommensurable—the first negating the possibility of having representational content; the second requiring it, and thus the linguistic I acquires the characteristic of "nonspecularity", becoming part of the explicit linguistic content without being capable of being visualised in the way that one can visualise objects in the external world.
Stamenov has a section entitled "Every sentence must have a self". The philosopher WV Quine asked the question, "Why does every sentence have a tense?" These propositions seem to be related and both perhaps are consistent with the performative hypothesis (associated with JR Ross and R Lakoff) that every sentence has an implicit superordinant clause in the first person and the present tense. This in turn relates to the concept of Buehler13 that the structure of language has, at its core, a deictic origin of the I, here, and now. Somehow language is built around this origin and, if it is lost, Buehler suggests language begins to fall apart. This is perhaps what we see happening in schizophrenic thought disorder.14 Sass remarks on page 254 that auditory hallucinations lack deixis—they are disembodied. Stamenov’s chapter is the most cogent exposition of the methodology of modern linguistics and its application to the complex philosophical and scientific problems raised by the phenomena of psychosis that I know.
Stamenov and Kircher and David’s concluding chapter both draw attention to the unsolved problem of the nature of consciousness. This surely is closely related to the problem of the evolution of language. I have argued that the Broca-Annett axiom is crucial and that the cerebral torque is all that we have available as an anatomical basis of the capacity for language. De Saussure made a key contribution in separating the phonological engram from the other components (concepts and meanings) of the sign and this separation must relate to the differential functions of the hemispheres. The anatomical disposition of the torque along the anteroposterior axis allows the motor and sensory engrams in Broca’s and Wernicke’s area respectively to interact with differing polarities with the corresponding areas of heteromodal association cortex in the non-dominant hemisphere. Thus language is conceived as a bihemispheric phenomenon with a deictic focus in Broca’s area and its relationship on the one hand to the internal word of thoughts in the right dorsolateral prefrontal cortex and on the other hand to the external world of perceived speech in Wernicke’s area. In this way can be conceptualised the critical role of the self in the origin of language and the phenomena of psychosis.
Kircher and David are to be congratulated on bringing together a set of seminal contributions on which further conceptual and scientific advances in understanding the phenomena of psychosis and their relationship to the human capacity of language can be based.
References
Crow TJ.Directional asymmetry is the key to the origin of Modern Homo Sapiens (the Broca-Annett axiom). Response to Lesley Rogers. Laterality 2004;9:233–242.
Crow TJ.What Marian Annett can teach Noam Chomsky and could have taught Stephen Jay Gould if he’d had time to listen. Cortex 2004;40:120–134.
Miskolczy D .Uber das anatomische korrelat der schizophrenie. Zeitschr Neurol 1933;147:509–544.
Povinelli DJ.Folk Physics For Apes. The Chimpanzees’ Theory of How the World Works. New York: Oxford University Press, 2000.
Crow TJ.Schizophrenia as the price that Homo sapiens pays for language: a resolution of the central paradox in the origin of the species. Brain Res Reviews 2000;31:118–129.
Tomasello M .On the different origins of symbols and grammar. In: Christiansen MH, Kirby S, eds. Language evolution Oxford: Oxford University Press, 2003;:94–110.
Crow TJ.A continuum of psychosis, one human gene and not much else - the case for homogeneity. Schiz Res 1995;17:135–145.
Vita A, Dieci M. Silenzi C, et al. Cerebral ventricular enlargement as a generalized feature of schizophrenia: a distribution analysis on 502 subjects. Schiz Res 2000;44:25–34.
Crow TJ.Schizophrenia as a transcallosal misconnection syndrome. Schiz Res 1998;30:111–114.
Pujol J, Vendrell P, Junque C, et al. When does human brain development end? Evidence of corpus callosum growth up to adulthood. Ann Neurol 1993;34:71–75.
Walter H, Wunderlich AP, Blankenhorn M. et al No hypofrontality, but absence of prefrontal lateralization comparing verbal and spatial working memory in schizophrenia. Schiz Res 2003;61:175–184.
Hsiao M-C, Lin K-J. Liu C-Y. transporter change in drug-na?ve schizophrenia: an imaging study with 99mTc-TRODAT-1. Schiz Res 2003;65:39–46.
Buehler K .Sprachtheorie. Translated (1990) by DW Goodwin as Theory of Language: The Representational Function of Language. Amsterdam: J Benjamins, 1934.
Crow TJ.Auditory hallucinations as primary disorders of syntax: An evolutionary theory of the origins of language. Cog Neuropsychiatry 2004; 9:125–145.(T J Crow)
This is a thought provoking book. Tilo Kircher and Anthony David have persuaded 22 authors to write on aspects of the self. The title might be "The self and schizophrenia" for this is the central focus. In his chapter Parnas makes this explicit:
"The most fundamental level of selfhood that appears to be affected in early schizophrenia is the automatic pre-reflective articulation of the first person perspective" (page 217)
and quotes Bleuler:
"Ganz intakt ist dennoch das Ich nirgends"
"The I is, however, never completely intact" to this effect. The proposal is that by attacking the self as a concept and dealing with its scientific and philosophical implications one can arrive at a more fundamental understanding of the nature of schizophrenia than Bleuler’s concept that it is a disorder of association. However, the implications are wider than this.
To be sure there are some disagreements and differences in emphasis between the contributors. Thus Berrios and Marková in a characteristically thorough historical survey conclude that the self is "fundamentally a western construct" to be explained by means other than neurobiology. Bentall writes:
"To many hard-nosed biological scientists, the self might seem an almost metaphysical concept" (page 301)
and quotes Baumeister:
"Providing a satisfactory definition of the self has proven fiendishly difficult. It is what you mean when you say "I". Most people use "I" and "self" many times each day, and so most people have a secure understanding of what the self is – but articulating that understanding is not easy".
Phillips writes:
"Sartre and Heidegger come to mind as philosophers who reject the notion that there is anything like a human nature or natural self-development ..." (page 320)
These quotations provide a background of cautious scepticism but there is much in this volume to convince the reader that addressing the problem of the self and its relationship to the phenomena of psychosis takes one into the core problem of the neural basis of psychotic symptoms and indeed the nature of human consciousness.
For example, there is the question of the relationship between the self and the central executive—a key concept in "cognitive psychology". Blakemore and Frith (chapter 19) outline the theory that auditory hallucinations and delusions of control arise because of a failure in the mechanism by which the predicted consequences of self-produced actions are derived from an internal "forward" model that predicts and cancels the sensory consequences of self-produced actions. This widely discussed model has a nice wiring diagram (figure 19.1) according to which a patient lacks awareness of initiating an action or thought or of its predicted consequences. However, as Zahavi argues on pp 67–68 there is a problem. The account is based on the assumption that the "mineness" of experience comes about as the result of high order representation, and that such meta-representational accounts of conscious experience are flawed in that they depend on an infinite regress—that is, they require the invocation of an "homunculus" whose function in turn needs to be accounted for by a further homunculus at a higher level. Similar criticisms apply to theory of mind formulations discussed by Kai Vogeley in chapter 17. A more parsimonious account of the differences between cortical function in other primates, mammals, and humans is required.
Where then is the neural basis of the self? According to Markowitsch (chapter 9) it is located in the right pre-frontal and anterior temporal cortex on the basis of evidence relating to the hemispheric encoding and retrieval asymmetry (HERA) hypothesis of episodic memory, according to which left pre-frontal cortex is engaged in recording episodic memories while right pre-frontal cortex is relevant to retrieval. Keenan et al (chapter 8) agree on the basis of studies that implicate right hemisphere/pre-frontal cortex in self-recognition and self-face recognition as well as episodic memory. These chapters therefore introduce the topic of cerebral asymmetry; but do not tell us why it is relevant.
I missed reference to what I have called the Broca-Annett axiom—the concept that cerebral asymmetry is the characteristic that defines the human brain.1,2 This is the notion that at some point in hominid evolution the brain lateralised and that this innovation was critical to the evolution of language. On page 198 of this volume Panksepp refers to the problems of animal models of schizophrenia:
"Not only is it risky to infer mental processes simply from the study of behavioural responses, but some of the key symptoms of schizophrenia (eg thought disorder) may be impossible to model in creatures of modest cortical endowment".
If, as Panksepp also suspects may be the case, the disease is specific to the human capacity for language, the problem is deeper than this. Miskolczy3 considered that schizophrenia was a disease of Homo sapiens-specific cerebral cortex; the Broca-Annett axiom is a lead to the nature of its anatomical basis. Gallup et al (page 147) place the problem of self-awareness in the context of primate evolution by considering the point in the primate lineage at which evidence of self-recognition emerges in mirror experiments. They equate this point with the development of the capacity for a "theory of mind and appreciation of the existence of others" and relate this capacity to "enlargement of the frontal lobes and vulnerability to the phenomena of psychosis". However, theory of mind interpretations of mirror recognition have been contested.4 Susceptibility to schizophrenia may be more precisely related to the capacity for language and the origin of the species5, and cerebral correlates of this transition may be more recent and more precisely defined. Tomasello6 has suggested that the correlate in human ontogeny of the capacity for language is the development of the ability to share with another individual a focus of attention in the outside world—he asks "why don’t chimpanzees point?" This is a plausible correlate of cerebral lateralisation and yields a radical alternative to frontal lobe explanations of the origin of psychotic symptoms.
One can ask further questions about how the phenomena of psychosis considered as deviations in the integrity of the self match up with what we know about the epidemiology and biology of schizophrenia. From the evidence of the WHO and other studies7 schizophrenic disorders have an approximately uniform incidence in all populations. Onsets occur, with a characteristic sex difference, in the reproductive phase of life and are associated with structural changes in the brain as reflected in a degree of ventricular enlargement.8 Thus we are dealing with variation, associated with variations in brain structure, that is maintained across the species in the face of biological disadvantage. On page 236 Parnas writes that:
"...disruption in the ontogenesis of corticocortical connectivity may represent a crucial aetiological element in the origins of schizophrenia ... Development of the elemental self-object relation is ... most likely heavily dependent on a progressive sophistication of intra- and intermodal sensory and sensorimotor integrations. An intact and very precise intracortical and corticocortical connectivity is a necessary condition for such developments".
On page 375 Vogeley writes of "converging evidence for schizophrenia as a disconnectivity syndrome". To be sure much evidence is consistent with schizophrenia as a misconnectivity syndrome and this implies a uniformity of pathogenesis incompatible with a polygenic theory. The key question is where are the misconnections? If schizophrenia is a Homo sapiens-specific condition associated with disorders of self-concept that emerged with the evolution of language, the anatomical basis of these developments is the central issue. There appears to be no alternative to the Broca-Annett axiom that the brain lateralised in course of hominid evolution, and that the final step was the advent of the capacity for language. The critical connections are those between the hemispheres and this must surely be where the misconnectivity occurs.9 It is of particular interest that development of the corpus callosum continues into the third and fourth decades of life10 and is associated with a sex difference (continuing later in females) that is a possible correlate of the sex difference in age of onset of psychosis.
It is a remarkable but commonplace fact that psychotic symptoms, including those relating to the core of the self, sometimes respond to neuroleptic medication. In their contribution to "The Self" Walter and Spitzer consider the implications of this observation. They also point out the awkwardness of having two apparently unrelated explanations of ichst?rungen—"disorders of the I"—as internal monitoring deficits in the case of passivity phenomena on the one hand and, in terms of the theory of mind, for experiences of reference, on the other. This is unparsimonious Walter and Spitzer point out. Elsewhere they11 have elegantly demonstrated that hypofrontality in schizophrenia is not a general deficit—there is a failure of lateralisation, verbal activation to the left, and spatial activation to the right. What is required is some integration of dopaminergic and laterality hypotheses. Here the observations of Hsiao et al12 are relevant. These authors have shown that dopamine uptake (or possibly turnover) in the striatum is not grossly changed in individuals with schizophrenic disorders, but whereas in controls there is marked lateralisation, in patients this is lost. If the torque is the defining characteristic of the human cerebral cortex, corticostriatal influences could be central to an understanding of the nature of the disturbance in psychosis and its response to medication.
Perhaps the most distinctive contribution to this volume is by Maxim Stamenov, who places the concept of the self within the framework of linguistic theory. He gives an account of the symbolic representation in language of the first person personal pronoun—"this I is the most unusual symbol humans have at their disposal". He explains that there are two aspects of the function of the I—it refers in one direction to the implicit central executive, while it gets its meaning from another world, the external world of current explicit cognitive representation. These requirements are incommensurable—the first negating the possibility of having representational content; the second requiring it, and thus the linguistic I acquires the characteristic of "nonspecularity", becoming part of the explicit linguistic content without being capable of being visualised in the way that one can visualise objects in the external world.
Stamenov has a section entitled "Every sentence must have a self". The philosopher WV Quine asked the question, "Why does every sentence have a tense?" These propositions seem to be related and both perhaps are consistent with the performative hypothesis (associated with JR Ross and R Lakoff) that every sentence has an implicit superordinant clause in the first person and the present tense. This in turn relates to the concept of Buehler13 that the structure of language has, at its core, a deictic origin of the I, here, and now. Somehow language is built around this origin and, if it is lost, Buehler suggests language begins to fall apart. This is perhaps what we see happening in schizophrenic thought disorder.14 Sass remarks on page 254 that auditory hallucinations lack deixis—they are disembodied. Stamenov’s chapter is the most cogent exposition of the methodology of modern linguistics and its application to the complex philosophical and scientific problems raised by the phenomena of psychosis that I know.
Stamenov and Kircher and David’s concluding chapter both draw attention to the unsolved problem of the nature of consciousness. This surely is closely related to the problem of the evolution of language. I have argued that the Broca-Annett axiom is crucial and that the cerebral torque is all that we have available as an anatomical basis of the capacity for language. De Saussure made a key contribution in separating the phonological engram from the other components (concepts and meanings) of the sign and this separation must relate to the differential functions of the hemispheres. The anatomical disposition of the torque along the anteroposterior axis allows the motor and sensory engrams in Broca’s and Wernicke’s area respectively to interact with differing polarities with the corresponding areas of heteromodal association cortex in the non-dominant hemisphere. Thus language is conceived as a bihemispheric phenomenon with a deictic focus in Broca’s area and its relationship on the one hand to the internal word of thoughts in the right dorsolateral prefrontal cortex and on the other hand to the external world of perceived speech in Wernicke’s area. In this way can be conceptualised the critical role of the self in the origin of language and the phenomena of psychosis.
Kircher and David are to be congratulated on bringing together a set of seminal contributions on which further conceptual and scientific advances in understanding the phenomena of psychosis and their relationship to the human capacity of language can be based.
References
Crow TJ.Directional asymmetry is the key to the origin of Modern Homo Sapiens (the Broca-Annett axiom). Response to Lesley Rogers. Laterality 2004;9:233–242.
Crow TJ.What Marian Annett can teach Noam Chomsky and could have taught Stephen Jay Gould if he’d had time to listen. Cortex 2004;40:120–134.
Miskolczy D .Uber das anatomische korrelat der schizophrenie. Zeitschr Neurol 1933;147:509–544.
Povinelli DJ.Folk Physics For Apes. The Chimpanzees’ Theory of How the World Works. New York: Oxford University Press, 2000.
Crow TJ.Schizophrenia as the price that Homo sapiens pays for language: a resolution of the central paradox in the origin of the species. Brain Res Reviews 2000;31:118–129.
Tomasello M .On the different origins of symbols and grammar. In: Christiansen MH, Kirby S, eds. Language evolution Oxford: Oxford University Press, 2003;:94–110.
Crow TJ.A continuum of psychosis, one human gene and not much else - the case for homogeneity. Schiz Res 1995;17:135–145.
Vita A, Dieci M. Silenzi C, et al. Cerebral ventricular enlargement as a generalized feature of schizophrenia: a distribution analysis on 502 subjects. Schiz Res 2000;44:25–34.
Crow TJ.Schizophrenia as a transcallosal misconnection syndrome. Schiz Res 1998;30:111–114.
Pujol J, Vendrell P, Junque C, et al. When does human brain development end? Evidence of corpus callosum growth up to adulthood. Ann Neurol 1993;34:71–75.
Walter H, Wunderlich AP, Blankenhorn M. et al No hypofrontality, but absence of prefrontal lateralization comparing verbal and spatial working memory in schizophrenia. Schiz Res 2003;61:175–184.
Hsiao M-C, Lin K-J. Liu C-Y. transporter change in drug-na?ve schizophrenia: an imaging study with 99mTc-TRODAT-1. Schiz Res 2003;65:39–46.
Buehler K .Sprachtheorie. Translated (1990) by DW Goodwin as Theory of Language: The Representational Function of Language. Amsterdam: J Benjamins, 1934.
Crow TJ.Auditory hallucinations as primary disorders of syntax: An evolutionary theory of the origins of language. Cog Neuropsychiatry 2004; 9:125–145.(T J Crow)