Journal of the Royal Society of Medicine Volume 83 May 1990
325
On insight and psychosis: discussion paper
Anthony S David MRCPsych MPhil
Institute of Psychiatry, Denmark Hill, London SE5 8AF
Keywords: insight; psychosis; delusions; neuropsychology
Summary The concept of insight into psychosis has received scant attention in the psychiatric literature. Various types of insight are described after drawing on such sources as phenomenology, clinical observation and experimental psychology. It is proposed that insight is far from an all or none phenomenon but comprises three overlapping dimensions, namely, the recognition that one has a mental illness, compliance with treatment and the ability to re-label unusual mental events (delusions and hallucinations) as pathological.
Introduction A middle aged single woman is referred to the outpatient clinic because of back pain. No other problems are mentioned. She is a teacher with no previous medical or psychiatric history. On examination she has a large fungating carcinoma of the breast, which she later admits, had been present for over a year. Investigation reveals metastatic deposits in the spine and brain. Surely the patient knew what was wrong - had some insight - but was motivated to deny that particularly unpleasant piece of reality. At the same time, was coming to the clinic with the likelihood of being 'discovered', a way of having that denial broken? And what role did the cerebral deposits play in this behaviour?
Unlike the case history above, the psychiatrist has to consider mental rather than physical symptoms, and in the case of psychosis to be discussed here, the possibility of motivated denial is seldom so clear-cut. Nevertheless, for the psychiatrist, the assessment of insight is a constant preoccupation. More than 50 years ago, leading psychiatrist Aubrey Lewis remarked that little had been written about the concept of insight. He provided a temporary definition of the term: 'a correct attitude to morbid change in oneself',' but warned that the words 'correct', 'attitude', 'morbid' and 'change' each called for discussion. Zilboorg2 stated that 'amongst the unclarities which are of utmost clinical importance and which cause utmost confusion is the term insight'. Despite or perhaps because of these assertions, contemporary authors have tended to ignore the subject. Post3 discards it as a concept with 'limited value', and few would agree with the narrow use of insight as sole criterion upon which to distinguish neurosis from psychosis4. Yet leading textbooks from both the United Kingdom5 and the United States6 continue to recommend the assessment of insight as an informative aspect of a patient's mental state. Whether or not all psychiatrists acknowledge it, some assessment of insight, explicit or implicit, is crucial to the process of diagnosis, especially where psychosis is concerned. Before continuing we need to agree both what insight is and is not. The sudden appreciation of how parts relate to an organized whole with the
accompanying 'a-ha' experience is sometimes termed insight by Gestalt psychologists. This will not be considered further. Nor will the term be used in the popular, Oxford English Dictionary sense of, discernment, understanding and wisdom. The area encompassed by the term in psychoanalytic circles usually subdivided and prefaced by 'emotional' or 'intellectual'- is so vast and treacherous that it will be avoided where possible. Freud7, while not employing the term specifically, realized that what present day analysts would call insight was not merely rational self-evaluation, otherwise simply reading psychoanalytic texts would cure neurosis. Rather, it requires an appreciation of hidden truths which when uncovered lose their power to cause neurotic conflict. It is clear therefore that the conventional psychiatric definition differs from the Freudian notion. This article will survey briefly insight as applied to psychosis (that is severe mental disturbance characterized by delusions and hallucinations). This can be separated into two main areas: (i) the patient's recognition that he or she is suffering from an illness and the 'realization that the illness is mental" and (ii) the ability to re-label the experience of certain mental events as pathological. This would include realizing that 'hearing voices' when no-one is there, is in fact an auditory hallucination.
Based on Mental Health Foundation Prize
Essay, 1989 Section of Psychiatry
Recognition of illness Most normal people would concede that knowing oneself fully is an unattainable ideal. Realizing that one is ill (having insight according to our definition) might therefore be held up as a considerable achievement. Paradoxically, it requires that a person views his own subjective experience, objectively8. Any awareness of this kind demands higher than normal standards of self-knowledge given the unavoidable reliance on the apparatus of mind to carry out this task, it being by definition faulty. Such a formulation must imply a 'modularity of mind'9 whereby one faculty or module, in this case an observing agency, can remain functioning while another faculty is malfanctioning. In practice, many patients are aware of their illness despite severe psychiatric disturbance. For example in a classic study of melancholia'0, 18 out of 61 patients conceded illness of some kind while 14 (= 25%) considered that they had a mental disorder. But are the schizophrenic psychoses different? Again Jaspers9 stated emphatically that 'In psychosis there is no lasting or complete insight'. The World Health Organization's International 0141-0768/90/ 050325-05/$02.00/0 Pilot Study of Schizophrenia in different cultures" © 1990 confirmed Jaspers' view. Insight, said to be present The Royal if there was some awareness of 'emotional illness' and Society of absent if the patient vigorously denied that he Medicine
326
Journal of the Royal Society of Medicine Volume 83 May 1990
disturbed, was lacking in 97% of the sample. The study dealt with recent onset cases, while another study of 68 chronic schizophrenic patients found that only one third denied any mental illness'2. Thirty spoke of having 'an insanity' or 'nervous breakdown' and 16 clearly identified their illness as schizophrenia some even naming the subtype. The stage of illness at which insight is assessed is clearly important'3. The widely held view that insight during an acute psychosis is a contradiction in terms, renders the assessment of insight in such circumstances virtually impossible. Retrospective insight Francis Willis (1823)14 asserted: 'no man ... can be considered sane, until he freely and voluntarily confess his delusions'. A century later, Bleuler'5 wrote: 'Even the seemingly corrected delusions in schizophrenia should rather be considered as forgotten or pushed aside'. Jaspers was of a similar opinion believing that though patients may claim to be convinced of the unreality of their past experiences, when probed, many still harbour serious doubts8. If one cannot talk of full insight in this situation then perhaps one can talk of partial insight. More recently, Wing et al.'6 asked a large group of schizophrenic patients in London just prior to discharge whether they would classify themselves as having been mentally ill. One fifth said yes, with half conceding to 'nerves' or 'strain'. Twenty three per cent said that their own delusions and hallucinations indicated that they were or had been ill. An interesting finding emerged which reveals another facet of insight namely that 46% of the sample would regard a person who reported hearing voices as mentally ill and as many as 60% thought an average visitor would say the same. It appears then that insight into another's illness may be preserved despite the loss of personal insight'7. Cutting'8 asked 20 remitted schizophrenics whether they thought they had had a breakdown (14 said yes) or had been ill (17 said yes) and concluded that a surprising proportion of patients do possess insight, contrary to the expectations of many psychiatrists. Modern authors readily accept the notion that there are degrees of insight5 of which the retrospective kind is as valid as any other and that its development is an integral part of the recovery process.
Relationship of insight to compliance It would appear logical to assume that insight predicts treatment compliance. This can become circular when requests for treatment and compliance are used as evidence for the presence of insight. McEvoy et al.'9 asked 45 chronic schizophrenic patients whether they felt they were ill and required treatment. Only 13% agreed they were ill with 27% accepting a need for medication. A systematic study of this was conducted20 where insight was measured by asking 100 chronic schizophrenic patients three questions: Do you think you, (a) had to be in hospital? (b) had to see a psychiatrist? (c) had to see a doctor? Only 31 answered yes to one of the questions and of these, 14 adhered to their medication. Ofthe remaining 69, 12 took their medication. It should be emphasised that over half of the insightful patients did not take their medication and 17% ofthe insightless ones did. The same issue was examined in a clinical trial which compared 29 habitual drug-refusers with 30 drug-
compliers, all were schizophrenic21. Insight was determined using the WHO definition and it was found that seven drug-refusers had insight compared to 18 of the drug-compliers (P< 0.01). Again, it is clear that insight though related to compliance is a rather poor predictor of it. A similar design was employed22 in a study of 58 schizophrenic patients, 32 of whom were non-compliant. These patients were generally more disturbed and tended to lack insight that is, deny their illness. Looking at treatment in general13, 38 schizophrenic outpatients were followed to see whether they showed 'early insight' defined as 'a patient's ability during the early phase of decompensation, to recognize that he or she is beginning to suffer a relapse'. Twenty-four (63%) demonstrated insight and of these 22 were restabilized successfully as outpatients. In contrast, seven of the 14 subjects without early insight, who as a group were no more symptomatic, required hospitalization. Insight clearly aids compliance but what is peculiar about these results is that patients can have no insight into illness yet still accept and derive benefit from medication. It is therefore recommended that drug compliance and awareness of illness be regarded as separate though overlapping constructs which contribute to insight. Is insight a good thing? Insofar as insight tends to make treatment compliance more likely it must be judged desirable. However, the value of insight may be measured in other ways. Outcome for example may be influenced by the
patient's experience of illness, although Wing16 found no such effect. Two studies2324 investigated the views of patients who had recovered from an acute schizophrenic episode. Both found that a negative attitude or one which resulted in the illness being ignored or denied, led to poor psychosocial adjustment. Furthermore it emerged that an excessively insightful, overly positive attitude bordering on 'romantic idealization' also correlated with poor outcome. It was concluded that the absence of a negative attitude (rather than having positive one) is the critical factor4. Defining insight more broadly to include 'psychological mindedness', showed that insightful patients were better adjusted behaviourally during their hospital stay, yet they were more psychologically distressed25. This concurs with the frequent observation that insight may involve a tragic and painful struggle against the ravages of psychotic disturbance. That lack of insight goes along with elevated and even elated mood is supported by other workers'3 21'22. Is it possible to have too much insight, to be subjected to continuous torturing selfexamination as to whether or not one is sane? Both too much and too little could be construed as forms of abnormal illness behaviour26. A compromise might be sufficient insight to accept treatment but not so much that it encourages brooding on the 'reality' of how severely ill one is. This formulation has parallels in the more favourable prognosis achieved by cancer patients who adopt a 'fighting
spirit'27. The assessment of insight in a patient whose symptom is the belief that he or she is 'going mad' is genuinely problematic though easily ridiculed (see Szasz28). It does however have some basis in the real world, usually in the setting of depression, acute
Journal of the Royal Society of Medicine Volume 83 May 1990
anxiety or obsessional neurosis. The patient is distressed by the belief or realization that their 'mind is going' and this dominates the presentation. The clinician will avoid words like 'mad' and will concentrate on the form rather than content of the preoccupation and may eventually arrive at a psychiatric diagnosis. The patient has some insight in recognizing the presence of mental disorder but is in disagreement with the psychiatrist as to the correct application of terms (such as 'mad'). Insight is therefore partial at best.
Pseudo-insight Jaspers warned that listening to the patient's utterances out of context can lead to mistaken judgements about the presence of insight. Someone well-versed in psychological theory and jargon, who understands and uses technical terms appropriately, does not automatically possess insight. They may acknowledge 'morbid change' but this is not sufficient to be considered insightful. With 'pseudo-insight' the patient merely regurgitates overheard explanations. Factual accuracy is irrelevant, otherwise the discovery of the pathogenic role of a new neurotransmitter would negate the insight of someone whose understanding of mental disorder was built upon an old one. Insight need not imply knowledge of causality either, a view at odds with psychoanalytic formulations9. It simply requires the acceptance of personal illness affecting the mental apparatus (the ability to think, perceive, act, remember, etc) whose aetiology may be and often is, unknown. Nevertheless, pseudo-insight, like the retrospective variety, may be of value to the patient in establishing order in their lives and may initiate a process leading to what could be called true insight. The neuropsychology of insight There are many parallels between insight in the psychiatric context and the disturbances in selfawareness which occur following known brain lesions. A person's recognition that they have a mental illness is surely a specific kind of self-awareness or self-concem. The latter faculty is characteristically lost after frontal lobe damage30. Likewise 'excessive selfconcern' was once the prime indication for frontal leucotomy3l. However it is the syndrome of anosognosia, a term coined by Babinski in 1914 meaning lack of awareness of disease, which is of particular interest in this context. This syndrome is usually confined to lesions of the right hemisphere, usually the parietal lobe32. The accompanying affect especially where the lack of awareness amounts to denial, is often one of euphoria. Again, it appears that elevated mood and lack of insight go together. Denial of hemiplegia is often dramatic and bizarre, sometimes quasi-psychotic. The patient may be capable of seeing and moving his limbs yet flatly denies that they belong to him. Alternatively, someone with a flaccid and dense left hemiplegia may claim to be perfectly able. Thus, the patient lying paralysed in his hospital bed, does not see himself as ill. Similarly, the chronic schizophrenic accepts his admission to the psychiatric ward yet denies that he has a mental illness. In neither case do the denials appear to be contrived, 'psychological' defences. In the former the underlying lesion is selfevident, in the latter, there is no obvious cerebral pathology yet it is conceivable though highly speculative that similar brain mechanisms operate.
Re-labelling mental events as pathological 'A schizophrenic inmate of an asylum enters a room in a country inn . . . and can only be removed by force for he expects the Queen of Holland, who wishes to marry him to arrive at any moment. He is a little ungainly creature . . . without a single advantage ... It is impossible that the Queen ofHolland should know anything of this poor patient in Switzerland, and if she did, he would be the last man she would wish to marry .. . He imagines something absolutely impossible and . . believes it to be reality.' (Bleuler, 191333)
This section deals with insight into psychotic phenomena, especially delusions, defined as a subject's ability to re-label correctly the experience of these unusual mental events as pathological. Such a consideration runs contrary to most early and some modern psychopathologists' definitions of delusions, namely that they are false judgements held with 'extraordinary conviction. . . an incomparable subjective certainty; there is an imperviousness to other experiences and to compelling counter argument; the content is impossible'8. This definition is elaborated by Mullen3 and concurs with the American Psychiatric Association's criteria enshrined in the Diagnostic and Statistical Manual (DSM-III).35 According to its operational definition, delusions are: 'sustained in spite of what almost everyone else believes and in spite of what constitutes incontrovertible and obvious proof or evidence to the contrary' (p 356). This position has been challenged recently by authors who do not accept that delusions (and hallucinations) are unitary concepts. In addition and of most relevance here, they contest the notion of absolute conviction3638. It is assumed that as conviction diminishes so insight increases. (So-called pseudo-hallucinations will not be discussed. -These, according to some authorities, include by definition false perceptions plus awareness of their unreality, or in other words, the presence of insight39). The Present State Examination (PSE)40 a semistructured psychiatric interview, requires a present/absent rating on psychotic symptoms though leaves room for those which are 'partial' or 'questionable'. Reviewing the ratings on 11941 patients it was found that questionable delusions and hallucinations, ie those with 'intermediate levels of disbelief , accounted for over half of the sample's abnormal experiences. The author proposed that psychotic phenomena must be viewed as points on a continuum. Several researchers36 go further and demonstrate that delusions have several independent dimensions which correlate poorly with each other. Specifically, 'conviction' which most subjects scored highly, did not correlate with any other single item42. It can therefore be concluded that, like the allied concept of insight, delusions are most usefully regarded as multidimensional. The degree of conviction in the deluded, as well as being unrelated to other dimensions of belief, may itself vary considerably. This paradoxical situation has been called a 'double-awareness phase' in the recovery from delusions43 although similar intermediate states occur during their onset". These states may arise from rapid oscillations between belief and disbelief or because an individual becomes amenable to testing their still firmly held beliefs against reality. This aspect was studied in depth37 and it was found that some patients are influenced by evidence which disconfirms their delusions.
327
328
Journal of the Royal Society of Medicine Volume 83 May 1990
Unfortunately, the process can work in reverse, so that chance external events are construed as confirming a (false) belief. The author argues that both normals and psychotics are most susceptible to confirmatory evidence for current beliefs and that change occurs first through the acceptance of an alternative view and then the confirmation of it by experience and reasoning. Accepting the veracity of contradictory evidence for the initial belief is somewhat rarer. Nevertheless, despite the Jasperian assumption of resistance to compelling counterargument, some workers have found that direct confrontation ofdelusional beliefs with selected patients can be rewarding therapeutically45.
Mechanisms of delusion formation Many authors have proposed that delusions are understandable interpretations of abnormal perceptual experiences44-". However, this does not explain the primary delusions of Jaspers nor the maintenance of abnormal beliefs in the face of contradictory evidence, implying some aetiological role for disordered thinking47 or real world knowledge (common sense)48. There is a strong case for dissecting the logical processes leading up to the formation of beliefs and by comparing normal and psychotic people49. An effort has been made to do this experimentally5O which showed that deluded subjects predicted future events more readily and on the basis of inferior standards of proof in contrast to normal and psychiatric controls. It is unlikely that all delusional experiences have a single underlying mechanism but perhaps the combination of an abnormal perception which is particularly command-ing, and a liability to form inferences in an idio-syncratic way, is the most parsimonious explanation for delusion formation. This is an area which deserves more research. Knowing and not knowing Despite insight into psychotic phenomena being at least plausible, given the mind's modularity, it remains mysterious how contradictory beliefs can be maintained simultaneously33. Sigmund Freud51 claimed: 'The strange behaviour of patients, in being able to combine a conscious knowing with not knowing, remains inexplicable by what is called normal psychology.' He got round this by introducing a new psychology or metapsychology which could accommodate unconscious knowing. As stated in the introduction, this is awkward territory, for the present purposes, best avoided. Unfortunately, the vivid illustrations ofthis very paradox described by Bleuler as autistic thinking (see quotation), or 'double awareness'43, plus everyday clinical practice force us to attend to this problem. It is one which makes psychosis, especially schizophrenia seem so baffling. Since Freud, there have been theoretical advances in neuropsychology which might one day explain dual awareness such as the split-brain52 where the cognitive contents of one hemisphere are unavailable to the other. Also recognition without awareness can be demonstrated in some cases of agnosia. For example patients with prosopagnosia (inability to recognize faces usually due to bilateral parieto-occipital damage) may react electrophysiologically with a galvanic skin response to a familiar face yet maintain their lack of recognition despite repeated questioning53. It must be conceded that these are
rather curious instances of knowledge without insight from which it is difficult to generalize. An example of different levels of awareness with which we are all familiar occurs in relation to memory, when we recognize a name, person or place which we have been unable to recall. The knowing and not knowing conundrum may rest on a false dichotomy. Cognitive-behaviourism which distinguishes instead between knowing and doing, provides yet another framework for understanding aspects of human conduct which would otherwise remain as strange and inexplicable as they did to Freud. It is well recognized that therapies for phobias and obsessions which rest entirely on altering behaviour, eschewing insight, are highly effective. Modification of the underlying irrational fears and compulsions may then follow suit. Thus there can be a disjunction between thought and action54. In addition, exposure treatment for dysmorphophobia of delusional intensity, has been found to reduce delusional conviction as well as avoidance behaviour despite the erroneous belief being unalterable by rational argument55. It is a prerequisite of the behavioural approach that patients act against their beliefs regardless of -how tenaciously they are held. Therefore, it is less surprising to find that not all deluded patients automatically act on their delusions. Indeed, a leading psychopathologist stated that deluded people are less likely to act on their beliefs than those with overvalued ideas whose conviction is less firm and whose insight is correspondingly greater. In summary, despite Freud's pessimism, there are now explanatory models in experimental psychology which can account for knowing and not knowing. If developed, these models would assist greatly in the complex task of the assessment of insight. Occasionally patients do act on their delusions sometimes with catastrophic consequences bringing them to the attention of forensic psychiatrists. Although little is known about which factors determine this, some workers posit that certain kinds of delusions are particularly liable to be acted upon but demographic variables and other elements of the mental state exert a substantial influence too56. It is not yet known in what way strength of conviction is related to delusionally motivated behaviour though research into this difficult area is in progress (Wessely, personal communication).
Conclusion The concept of insight goes to the heart of our thinking about psychosis and has important implications for management From the preceding discussion it emerges that insight has at least three dimensions: (1) awareness of illness, (2) the capacity to re-label psychotic experiences as abnormal, and (3) treatment compliance. More research is needed in order to validate this scheme and to determine the true place of this neglected area in the understanding of mental disorders. Current models in neuropsychology may be usefully applied to uncover underlying mechanisms. Finally it is hoped that as clinicians rediscover the concept of insight they will feel more inclined to encourage patients to rediscover it too, allowing them to play a more active role in the recovery process.
Acknowledgments: This article is based on a more lengthy essay which was awarded the Royal Society of Medicine's
Journal of the Royal Society of Medicine Volume 83 May 1990 Mental Health Foundation prize for 1989. The author wishes to thank the many colleagues who discussed and helped shape the ideas presented and the Medical Research Council for their support.
References 1 Lewis A. The psychopathology of insight. Br J Med Psychol 1934;14:332-48 2 Zilboorg G. The emotional problem and the therapeutic role of insight. Psychoanal Q 1952;21:1--24 3 Post F. Clinical assessment of mental disorders. In: Shepherd M, Zangwill OL, eds. Handbook ofpsychiatry, vol 1. Cambridge: Cambridge University Press, 1983:219 4 Hamilton M. Fish's clinical psychopathology. Bristol: John Wright, 1974 5 Gelder M, Gath D, Mayou R. Oxford textbook of psychiatry. Oxford: Oxford University Press, 1983;29-30,49 6 Freedman AM, Kaplan HI, Sadock BJ. Comprehensive textbook of psychiatry, vol. 1, 2nd edn. Baltimore: Williams & Wilkins 1975:736 7 Freud S. On beginning the treatment (firther recommendations on the technique of psychoanalysis I), standard edition, vol. 12. London: Hogarth Press, 1958:123-44 8 Japsers K. General psychopathology. (trans. Hoenig J, Hamilton MW). Manchester: Manchester University Press, 1959 9 Fodor JA. The modularity of mind. Cambridge, Massachusetts: MIT, 1983 10 Lewis A. Melancholia: a clinical survey of depressive states. J Med Sci 1934;80:277-378 11 World Health Organization. Report ofthe International Pilot Study of Schiophrenia, vol. 1. Geneva: WHO, 1973 12 Brooks GW, Deane WN, Hugel RW. Some aspects ofthe subjective experience of schizophrenia. Dis Nervous System 1968;29 (suppl 5):78-82 13 Heinrichs DW, Cohen BP, Carpenter WT. Early insight and the management of schizophrenic decompensation. J Nerv Ment Dis 1985;173:133-8 14 Hunter R, MacAlpine I. 300 years of Psychiatry 1535-1860. Oxford: Oxford University Press 1963 15 Bleuler E. Textbook ofpsychiatry (trans A. A. Brill, 1951) Dover Publications 16 Wing JK, Monck E, Brown GW, Carstairs GM. Morbidity in the community of schizophrenic patients discharged from London mental hospitals in 1959. Br J Psych 1964;110:10-21 17 Brown R. Schizophrenia, language and reality. Am Psychol 1973;28:395-403 18 Cutting J. The psychology of schizophrenia. London: Churchill Livingstone, 1985 19 McEvoy JP, Aland J, Wilson WH, Guy W, Hawkins L. Measuring chronic schizophrenic patients' attitudes toward their illness and treatment. Hosp Community Psychiatry 1981;32:856-8 20 Lin IF, Spiga R, Fortsch W. Insight and adherence to medication in chronic schizophrenics. J Clin Psychiatry 1979;40:430-2 21 Van Putten T, Crumpton E, Yale C. Drug refusal in schizophrenia and the wish to be crazy. Arch Gen Psychiatry 1976;33:1443-6 22 Bartk6 G, Herczeg I & Zfdor G. Clinical ymptomatology and drug compliance in schizophrenic patients. Acta Psychiatr Scand 1988;77:74-6 23 Soskis DA, Bowers MB. The schizophrenic experience: a follow-up study of attitude and posthospital adjustment. J Nerv Ment Dis 1969;149:443-9 24 McGlashan TH, Carpenter WT. Does attitude toward psychosis relate to outcome? Am JPsychiatry 1981;138: 797-801 25 Roback HB, Abramowitz SI. Insight and hospital adjustment. Can J Psychiatry 1979;24:233-6 26 Pilowsky I. A general classification of abnormal illness behaviour. Br J Med Psychol 1978;51:131-7 27 Greer S. Cancer and the mind. Br J Psychiatry 1983;143:535-543 28 Szasz TS. The age of madness. London: Routledge & Kegan Paul, 1974:166-7
29 Reid JR, Finesinger JE. The role of insight in psychotherapy. Am J Psychiatry 1952;108:726-34 30 Lishman WA. Organic psychiatry. Oxford: Blackwell Scientific, 1987 31 Robinson MF, Freeman W. Psychosurgery and the self. New York: Grune & Stratton, 1954 32 McGlynn SM, Schacter DL. Unawareness of deficits in neuropsychological syndromes. J Clin Exp Neuropsychol 1989;11:143-205 33 Bleuler E. Autistic thinking. Am J Insanity 1913;69: 873-86 34 Mullen P. Phenomenology of disordered mental function. In: Essentials of postgraduate psychiatry (ed P. Hill, R. Murray & G. Thorley). London: Academic Press, 1979 35 American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 3rd edn. Washington, DC: American Psychiatric Association, 1980 36 Kendler KS, Glazer WM, Morgenstern H. Dimensions of delusional experience. Am J Psychiatry 1983;140: 466-9 37 Brett-Jones JR. Recovery from delusions: a methodological study. M.Phil thesis. London: Institute of Psychiatry, 1984 38 Garety PA. Delusions: problems of definition and measurement. Br J Med Psychol 1985;58:25-34 39 Kraupl Taylor F. On pseudo-hallucnations. Psychol Med
1981;11:265-71 40 Wing JK, Cooper JE, Sartorius N. Measurement and Classification of Psychiatric Symptoms. Cambridge: Cambridge University Press, 1974 41 Strauss JS. Hallucinations and delusions as points on continua function. Arch Gen Psychiatry 1969;21:581-6 42 Garety PA, Hemaley DR. Characteristics of delusional experience. Eur Arch Psychiatry Neurol Sci 1987;26: 294-8 43 Sacks MH, Carpenter WT, Strauss JS. Recovery from delusions. Arch Gen Psychiatry 1974;30:-117-120 44 Maher B, Ross JS. Delusions. In: Comprehensive handbook of psychopathology Adams HE, Sutker PB, eds. New York: Plenum Press, 1984:383409 45 Milton F, Patwa VK, Hafner RJ. Confrontation vs belief modification in persistently deluded patients. Br JMed Psychol 1978;51:127-30 46 Winters KC, Neale JM. Delusions and delusional thinking in psychotics: a review of the literature. Clin Psychol Rev 1983;3:227-53 47 Arieti S. The interpretation of schizophrenia, 2nd edn. London: Crosby Lockwood Staples, 1974 48 Cutting J, Murphy D. Schizophrenic thought disorder: a psychological and organic interpretation. Br J Psychiatry 1988;152:310-319 49 Hemsley DR, Garety PA. The formation of delusions: a Bayesian analysis. Br J Psychiatry 1986;149:51-6 50 Huq SF, Garety PA, Hemsley DR. Probabilistic judgements in deluded and non-deluded subjects. Q J Exp Psychol 1988;40A:801-2 51 Freud S. The dissection of the psychical personality. In The New Introductory Lectures on Psychoanalysis. Standard Edition (1964) 22 57-80. London: Hogarth Press, 1933 52 Sperry RW. Hemisphere disownnection and the unity of conscious awareness. Am Psychol 1968;23:723-33 53 Bauer RM. Autonomic recognition of names and faces in prosopagnosia: a neuropsychological application of the guilty knowledge test. Neuropsychologia 1984;22: 457-69 54 Rachman S. Irrational thinking, with special reference to cognitive therapy. Advances in Behavioural Therapy 1983;5:63-88 55 Marks IM, Mishan J. Dysmorphophobic avoidance from disturbed bodily perception: a pilot study of exposure therapy. Br J Psychiatry 1988;152:674-8 56 Taylor PJ. Motives for offending among violent and psychotic men. Br J Psychiatry 1985;147:491-8 (Accepted 6 September 1989)
329
325
On insight and psychosis: discussion paper
Anthony S David MRCPsych MPhil
Institute of Psychiatry, Denmark Hill, London SE5 8AF
Keywords: insight; psychosis; delusions; neuropsychology
Summary The concept of insight into psychosis has received scant attention in the psychiatric literature. Various types of insight are described after drawing on such sources as phenomenology, clinical observation and experimental psychology. It is proposed that insight is far from an all or none phenomenon but comprises three overlapping dimensions, namely, the recognition that one has a mental illness, compliance with treatment and the ability to re-label unusual mental events (delusions and hallucinations) as pathological.
Introduction A middle aged single woman is referred to the outpatient clinic because of back pain. No other problems are mentioned. She is a teacher with no previous medical or psychiatric history. On examination she has a large fungating carcinoma of the breast, which she later admits, had been present for over a year. Investigation reveals metastatic deposits in the spine and brain. Surely the patient knew what was wrong - had some insight - but was motivated to deny that particularly unpleasant piece of reality. At the same time, was coming to the clinic with the likelihood of being 'discovered', a way of having that denial broken? And what role did the cerebral deposits play in this behaviour?
Unlike the case history above, the psychiatrist has to consider mental rather than physical symptoms, and in the case of psychosis to be discussed here, the possibility of motivated denial is seldom so clear-cut. Nevertheless, for the psychiatrist, the assessment of insight is a constant preoccupation. More than 50 years ago, leading psychiatrist Aubrey Lewis remarked that little had been written about the concept of insight. He provided a temporary definition of the term: 'a correct attitude to morbid change in oneself',' but warned that the words 'correct', 'attitude', 'morbid' and 'change' each called for discussion. Zilboorg2 stated that 'amongst the unclarities which are of utmost clinical importance and which cause utmost confusion is the term insight'. Despite or perhaps because of these assertions, contemporary authors have tended to ignore the subject. Post3 discards it as a concept with 'limited value', and few would agree with the narrow use of insight as sole criterion upon which to distinguish neurosis from psychosis4. Yet leading textbooks from both the United Kingdom5 and the United States6 continue to recommend the assessment of insight as an informative aspect of a patient's mental state. Whether or not all psychiatrists acknowledge it, some assessment of insight, explicit or implicit, is crucial to the process of diagnosis, especially where psychosis is concerned. Before continuing we need to agree both what insight is and is not. The sudden appreciation of how parts relate to an organized whole with the
accompanying 'a-ha' experience is sometimes termed insight by Gestalt psychologists. This will not be considered further. Nor will the term be used in the popular, Oxford English Dictionary sense of, discernment, understanding and wisdom. The area encompassed by the term in psychoanalytic circles usually subdivided and prefaced by 'emotional' or 'intellectual'- is so vast and treacherous that it will be avoided where possible. Freud7, while not employing the term specifically, realized that what present day analysts would call insight was not merely rational self-evaluation, otherwise simply reading psychoanalytic texts would cure neurosis. Rather, it requires an appreciation of hidden truths which when uncovered lose their power to cause neurotic conflict. It is clear therefore that the conventional psychiatric definition differs from the Freudian notion. This article will survey briefly insight as applied to psychosis (that is severe mental disturbance characterized by delusions and hallucinations). This can be separated into two main areas: (i) the patient's recognition that he or she is suffering from an illness and the 'realization that the illness is mental" and (ii) the ability to re-label the experience of certain mental events as pathological. This would include realizing that 'hearing voices' when no-one is there, is in fact an auditory hallucination.
Based on Mental Health Foundation Prize
Essay, 1989 Section of Psychiatry
Recognition of illness Most normal people would concede that knowing oneself fully is an unattainable ideal. Realizing that one is ill (having insight according to our definition) might therefore be held up as a considerable achievement. Paradoxically, it requires that a person views his own subjective experience, objectively8. Any awareness of this kind demands higher than normal standards of self-knowledge given the unavoidable reliance on the apparatus of mind to carry out this task, it being by definition faulty. Such a formulation must imply a 'modularity of mind'9 whereby one faculty or module, in this case an observing agency, can remain functioning while another faculty is malfanctioning. In practice, many patients are aware of their illness despite severe psychiatric disturbance. For example in a classic study of melancholia'0, 18 out of 61 patients conceded illness of some kind while 14 (= 25%) considered that they had a mental disorder. But are the schizophrenic psychoses different? Again Jaspers9 stated emphatically that 'In psychosis there is no lasting or complete insight'. The World Health Organization's International 0141-0768/90/ 050325-05/$02.00/0 Pilot Study of Schizophrenia in different cultures" © 1990 confirmed Jaspers' view. Insight, said to be present The Royal if there was some awareness of 'emotional illness' and Society of absent if the patient vigorously denied that he Medicine
326
Journal of the Royal Society of Medicine Volume 83 May 1990
disturbed, was lacking in 97% of the sample. The study dealt with recent onset cases, while another study of 68 chronic schizophrenic patients found that only one third denied any mental illness'2. Thirty spoke of having 'an insanity' or 'nervous breakdown' and 16 clearly identified their illness as schizophrenia some even naming the subtype. The stage of illness at which insight is assessed is clearly important'3. The widely held view that insight during an acute psychosis is a contradiction in terms, renders the assessment of insight in such circumstances virtually impossible. Retrospective insight Francis Willis (1823)14 asserted: 'no man ... can be considered sane, until he freely and voluntarily confess his delusions'. A century later, Bleuler'5 wrote: 'Even the seemingly corrected delusions in schizophrenia should rather be considered as forgotten or pushed aside'. Jaspers was of a similar opinion believing that though patients may claim to be convinced of the unreality of their past experiences, when probed, many still harbour serious doubts8. If one cannot talk of full insight in this situation then perhaps one can talk of partial insight. More recently, Wing et al.'6 asked a large group of schizophrenic patients in London just prior to discharge whether they would classify themselves as having been mentally ill. One fifth said yes, with half conceding to 'nerves' or 'strain'. Twenty three per cent said that their own delusions and hallucinations indicated that they were or had been ill. An interesting finding emerged which reveals another facet of insight namely that 46% of the sample would regard a person who reported hearing voices as mentally ill and as many as 60% thought an average visitor would say the same. It appears then that insight into another's illness may be preserved despite the loss of personal insight'7. Cutting'8 asked 20 remitted schizophrenics whether they thought they had had a breakdown (14 said yes) or had been ill (17 said yes) and concluded that a surprising proportion of patients do possess insight, contrary to the expectations of many psychiatrists. Modern authors readily accept the notion that there are degrees of insight5 of which the retrospective kind is as valid as any other and that its development is an integral part of the recovery process.
Relationship of insight to compliance It would appear logical to assume that insight predicts treatment compliance. This can become circular when requests for treatment and compliance are used as evidence for the presence of insight. McEvoy et al.'9 asked 45 chronic schizophrenic patients whether they felt they were ill and required treatment. Only 13% agreed they were ill with 27% accepting a need for medication. A systematic study of this was conducted20 where insight was measured by asking 100 chronic schizophrenic patients three questions: Do you think you, (a) had to be in hospital? (b) had to see a psychiatrist? (c) had to see a doctor? Only 31 answered yes to one of the questions and of these, 14 adhered to their medication. Ofthe remaining 69, 12 took their medication. It should be emphasised that over half of the insightful patients did not take their medication and 17% ofthe insightless ones did. The same issue was examined in a clinical trial which compared 29 habitual drug-refusers with 30 drug-
compliers, all were schizophrenic21. Insight was determined using the WHO definition and it was found that seven drug-refusers had insight compared to 18 of the drug-compliers (P< 0.01). Again, it is clear that insight though related to compliance is a rather poor predictor of it. A similar design was employed22 in a study of 58 schizophrenic patients, 32 of whom were non-compliant. These patients were generally more disturbed and tended to lack insight that is, deny their illness. Looking at treatment in general13, 38 schizophrenic outpatients were followed to see whether they showed 'early insight' defined as 'a patient's ability during the early phase of decompensation, to recognize that he or she is beginning to suffer a relapse'. Twenty-four (63%) demonstrated insight and of these 22 were restabilized successfully as outpatients. In contrast, seven of the 14 subjects without early insight, who as a group were no more symptomatic, required hospitalization. Insight clearly aids compliance but what is peculiar about these results is that patients can have no insight into illness yet still accept and derive benefit from medication. It is therefore recommended that drug compliance and awareness of illness be regarded as separate though overlapping constructs which contribute to insight. Is insight a good thing? Insofar as insight tends to make treatment compliance more likely it must be judged desirable. However, the value of insight may be measured in other ways. Outcome for example may be influenced by the
patient's experience of illness, although Wing16 found no such effect. Two studies2324 investigated the views of patients who had recovered from an acute schizophrenic episode. Both found that a negative attitude or one which resulted in the illness being ignored or denied, led to poor psychosocial adjustment. Furthermore it emerged that an excessively insightful, overly positive attitude bordering on 'romantic idealization' also correlated with poor outcome. It was concluded that the absence of a negative attitude (rather than having positive one) is the critical factor4. Defining insight more broadly to include 'psychological mindedness', showed that insightful patients were better adjusted behaviourally during their hospital stay, yet they were more psychologically distressed25. This concurs with the frequent observation that insight may involve a tragic and painful struggle against the ravages of psychotic disturbance. That lack of insight goes along with elevated and even elated mood is supported by other workers'3 21'22. Is it possible to have too much insight, to be subjected to continuous torturing selfexamination as to whether or not one is sane? Both too much and too little could be construed as forms of abnormal illness behaviour26. A compromise might be sufficient insight to accept treatment but not so much that it encourages brooding on the 'reality' of how severely ill one is. This formulation has parallels in the more favourable prognosis achieved by cancer patients who adopt a 'fighting
spirit'27. The assessment of insight in a patient whose symptom is the belief that he or she is 'going mad' is genuinely problematic though easily ridiculed (see Szasz28). It does however have some basis in the real world, usually in the setting of depression, acute
Journal of the Royal Society of Medicine Volume 83 May 1990
anxiety or obsessional neurosis. The patient is distressed by the belief or realization that their 'mind is going' and this dominates the presentation. The clinician will avoid words like 'mad' and will concentrate on the form rather than content of the preoccupation and may eventually arrive at a psychiatric diagnosis. The patient has some insight in recognizing the presence of mental disorder but is in disagreement with the psychiatrist as to the correct application of terms (such as 'mad'). Insight is therefore partial at best.
Pseudo-insight Jaspers warned that listening to the patient's utterances out of context can lead to mistaken judgements about the presence of insight. Someone well-versed in psychological theory and jargon, who understands and uses technical terms appropriately, does not automatically possess insight. They may acknowledge 'morbid change' but this is not sufficient to be considered insightful. With 'pseudo-insight' the patient merely regurgitates overheard explanations. Factual accuracy is irrelevant, otherwise the discovery of the pathogenic role of a new neurotransmitter would negate the insight of someone whose understanding of mental disorder was built upon an old one. Insight need not imply knowledge of causality either, a view at odds with psychoanalytic formulations9. It simply requires the acceptance of personal illness affecting the mental apparatus (the ability to think, perceive, act, remember, etc) whose aetiology may be and often is, unknown. Nevertheless, pseudo-insight, like the retrospective variety, may be of value to the patient in establishing order in their lives and may initiate a process leading to what could be called true insight. The neuropsychology of insight There are many parallels between insight in the psychiatric context and the disturbances in selfawareness which occur following known brain lesions. A person's recognition that they have a mental illness is surely a specific kind of self-awareness or self-concem. The latter faculty is characteristically lost after frontal lobe damage30. Likewise 'excessive selfconcern' was once the prime indication for frontal leucotomy3l. However it is the syndrome of anosognosia, a term coined by Babinski in 1914 meaning lack of awareness of disease, which is of particular interest in this context. This syndrome is usually confined to lesions of the right hemisphere, usually the parietal lobe32. The accompanying affect especially where the lack of awareness amounts to denial, is often one of euphoria. Again, it appears that elevated mood and lack of insight go together. Denial of hemiplegia is often dramatic and bizarre, sometimes quasi-psychotic. The patient may be capable of seeing and moving his limbs yet flatly denies that they belong to him. Alternatively, someone with a flaccid and dense left hemiplegia may claim to be perfectly able. Thus, the patient lying paralysed in his hospital bed, does not see himself as ill. Similarly, the chronic schizophrenic accepts his admission to the psychiatric ward yet denies that he has a mental illness. In neither case do the denials appear to be contrived, 'psychological' defences. In the former the underlying lesion is selfevident, in the latter, there is no obvious cerebral pathology yet it is conceivable though highly speculative that similar brain mechanisms operate.
Re-labelling mental events as pathological 'A schizophrenic inmate of an asylum enters a room in a country inn . . . and can only be removed by force for he expects the Queen of Holland, who wishes to marry him to arrive at any moment. He is a little ungainly creature . . . without a single advantage ... It is impossible that the Queen ofHolland should know anything of this poor patient in Switzerland, and if she did, he would be the last man she would wish to marry .. . He imagines something absolutely impossible and . . believes it to be reality.' (Bleuler, 191333)
This section deals with insight into psychotic phenomena, especially delusions, defined as a subject's ability to re-label correctly the experience of these unusual mental events as pathological. Such a consideration runs contrary to most early and some modern psychopathologists' definitions of delusions, namely that they are false judgements held with 'extraordinary conviction. . . an incomparable subjective certainty; there is an imperviousness to other experiences and to compelling counter argument; the content is impossible'8. This definition is elaborated by Mullen3 and concurs with the American Psychiatric Association's criteria enshrined in the Diagnostic and Statistical Manual (DSM-III).35 According to its operational definition, delusions are: 'sustained in spite of what almost everyone else believes and in spite of what constitutes incontrovertible and obvious proof or evidence to the contrary' (p 356). This position has been challenged recently by authors who do not accept that delusions (and hallucinations) are unitary concepts. In addition and of most relevance here, they contest the notion of absolute conviction3638. It is assumed that as conviction diminishes so insight increases. (So-called pseudo-hallucinations will not be discussed. -These, according to some authorities, include by definition false perceptions plus awareness of their unreality, or in other words, the presence of insight39). The Present State Examination (PSE)40 a semistructured psychiatric interview, requires a present/absent rating on psychotic symptoms though leaves room for those which are 'partial' or 'questionable'. Reviewing the ratings on 11941 patients it was found that questionable delusions and hallucinations, ie those with 'intermediate levels of disbelief , accounted for over half of the sample's abnormal experiences. The author proposed that psychotic phenomena must be viewed as points on a continuum. Several researchers36 go further and demonstrate that delusions have several independent dimensions which correlate poorly with each other. Specifically, 'conviction' which most subjects scored highly, did not correlate with any other single item42. It can therefore be concluded that, like the allied concept of insight, delusions are most usefully regarded as multidimensional. The degree of conviction in the deluded, as well as being unrelated to other dimensions of belief, may itself vary considerably. This paradoxical situation has been called a 'double-awareness phase' in the recovery from delusions43 although similar intermediate states occur during their onset". These states may arise from rapid oscillations between belief and disbelief or because an individual becomes amenable to testing their still firmly held beliefs against reality. This aspect was studied in depth37 and it was found that some patients are influenced by evidence which disconfirms their delusions.
327
328
Journal of the Royal Society of Medicine Volume 83 May 1990
Unfortunately, the process can work in reverse, so that chance external events are construed as confirming a (false) belief. The author argues that both normals and psychotics are most susceptible to confirmatory evidence for current beliefs and that change occurs first through the acceptance of an alternative view and then the confirmation of it by experience and reasoning. Accepting the veracity of contradictory evidence for the initial belief is somewhat rarer. Nevertheless, despite the Jasperian assumption of resistance to compelling counterargument, some workers have found that direct confrontation ofdelusional beliefs with selected patients can be rewarding therapeutically45.
Mechanisms of delusion formation Many authors have proposed that delusions are understandable interpretations of abnormal perceptual experiences44-". However, this does not explain the primary delusions of Jaspers nor the maintenance of abnormal beliefs in the face of contradictory evidence, implying some aetiological role for disordered thinking47 or real world knowledge (common sense)48. There is a strong case for dissecting the logical processes leading up to the formation of beliefs and by comparing normal and psychotic people49. An effort has been made to do this experimentally5O which showed that deluded subjects predicted future events more readily and on the basis of inferior standards of proof in contrast to normal and psychiatric controls. It is unlikely that all delusional experiences have a single underlying mechanism but perhaps the combination of an abnormal perception which is particularly command-ing, and a liability to form inferences in an idio-syncratic way, is the most parsimonious explanation for delusion formation. This is an area which deserves more research. Knowing and not knowing Despite insight into psychotic phenomena being at least plausible, given the mind's modularity, it remains mysterious how contradictory beliefs can be maintained simultaneously33. Sigmund Freud51 claimed: 'The strange behaviour of patients, in being able to combine a conscious knowing with not knowing, remains inexplicable by what is called normal psychology.' He got round this by introducing a new psychology or metapsychology which could accommodate unconscious knowing. As stated in the introduction, this is awkward territory, for the present purposes, best avoided. Unfortunately, the vivid illustrations ofthis very paradox described by Bleuler as autistic thinking (see quotation), or 'double awareness'43, plus everyday clinical practice force us to attend to this problem. It is one which makes psychosis, especially schizophrenia seem so baffling. Since Freud, there have been theoretical advances in neuropsychology which might one day explain dual awareness such as the split-brain52 where the cognitive contents of one hemisphere are unavailable to the other. Also recognition without awareness can be demonstrated in some cases of agnosia. For example patients with prosopagnosia (inability to recognize faces usually due to bilateral parieto-occipital damage) may react electrophysiologically with a galvanic skin response to a familiar face yet maintain their lack of recognition despite repeated questioning53. It must be conceded that these are
rather curious instances of knowledge without insight from which it is difficult to generalize. An example of different levels of awareness with which we are all familiar occurs in relation to memory, when we recognize a name, person or place which we have been unable to recall. The knowing and not knowing conundrum may rest on a false dichotomy. Cognitive-behaviourism which distinguishes instead between knowing and doing, provides yet another framework for understanding aspects of human conduct which would otherwise remain as strange and inexplicable as they did to Freud. It is well recognized that therapies for phobias and obsessions which rest entirely on altering behaviour, eschewing insight, are highly effective. Modification of the underlying irrational fears and compulsions may then follow suit. Thus there can be a disjunction between thought and action54. In addition, exposure treatment for dysmorphophobia of delusional intensity, has been found to reduce delusional conviction as well as avoidance behaviour despite the erroneous belief being unalterable by rational argument55. It is a prerequisite of the behavioural approach that patients act against their beliefs regardless of -how tenaciously they are held. Therefore, it is less surprising to find that not all deluded patients automatically act on their delusions. Indeed, a leading psychopathologist stated that deluded people are less likely to act on their beliefs than those with overvalued ideas whose conviction is less firm and whose insight is correspondingly greater. In summary, despite Freud's pessimism, there are now explanatory models in experimental psychology which can account for knowing and not knowing. If developed, these models would assist greatly in the complex task of the assessment of insight. Occasionally patients do act on their delusions sometimes with catastrophic consequences bringing them to the attention of forensic psychiatrists. Although little is known about which factors determine this, some workers posit that certain kinds of delusions are particularly liable to be acted upon but demographic variables and other elements of the mental state exert a substantial influence too56. It is not yet known in what way strength of conviction is related to delusionally motivated behaviour though research into this difficult area is in progress (Wessely, personal communication).
Conclusion The concept of insight goes to the heart of our thinking about psychosis and has important implications for management From the preceding discussion it emerges that insight has at least three dimensions: (1) awareness of illness, (2) the capacity to re-label psychotic experiences as abnormal, and (3) treatment compliance. More research is needed in order to validate this scheme and to determine the true place of this neglected area in the understanding of mental disorders. Current models in neuropsychology may be usefully applied to uncover underlying mechanisms. Finally it is hoped that as clinicians rediscover the concept of insight they will feel more inclined to encourage patients to rediscover it too, allowing them to play a more active role in the recovery process.
Acknowledgments: This article is based on a more lengthy essay which was awarded the Royal Society of Medicine's
Journal of the Royal Society of Medicine Volume 83 May 1990 Mental Health Foundation prize for 1989. The author wishes to thank the many colleagues who discussed and helped shape the ideas presented and the Medical Research Council for their support.
References 1 Lewis A. The psychopathology of insight. Br J Med Psychol 1934;14:332-48 2 Zilboorg G. The emotional problem and the therapeutic role of insight. Psychoanal Q 1952;21:1--24 3 Post F. Clinical assessment of mental disorders. In: Shepherd M, Zangwill OL, eds. Handbook ofpsychiatry, vol 1. Cambridge: Cambridge University Press, 1983:219 4 Hamilton M. Fish's clinical psychopathology. Bristol: John Wright, 1974 5 Gelder M, Gath D, Mayou R. Oxford textbook of psychiatry. Oxford: Oxford University Press, 1983;29-30,49 6 Freedman AM, Kaplan HI, Sadock BJ. Comprehensive textbook of psychiatry, vol. 1, 2nd edn. Baltimore: Williams & Wilkins 1975:736 7 Freud S. On beginning the treatment (firther recommendations on the technique of psychoanalysis I), standard edition, vol. 12. London: Hogarth Press, 1958:123-44 8 Japsers K. General psychopathology. (trans. Hoenig J, Hamilton MW). Manchester: Manchester University Press, 1959 9 Fodor JA. The modularity of mind. Cambridge, Massachusetts: MIT, 1983 10 Lewis A. Melancholia: a clinical survey of depressive states. J Med Sci 1934;80:277-378 11 World Health Organization. Report ofthe International Pilot Study of Schiophrenia, vol. 1. Geneva: WHO, 1973 12 Brooks GW, Deane WN, Hugel RW. Some aspects ofthe subjective experience of schizophrenia. Dis Nervous System 1968;29 (suppl 5):78-82 13 Heinrichs DW, Cohen BP, Carpenter WT. Early insight and the management of schizophrenic decompensation. J Nerv Ment Dis 1985;173:133-8 14 Hunter R, MacAlpine I. 300 years of Psychiatry 1535-1860. Oxford: Oxford University Press 1963 15 Bleuler E. Textbook ofpsychiatry (trans A. A. Brill, 1951) Dover Publications 16 Wing JK, Monck E, Brown GW, Carstairs GM. Morbidity in the community of schizophrenic patients discharged from London mental hospitals in 1959. Br J Psych 1964;110:10-21 17 Brown R. Schizophrenia, language and reality. Am Psychol 1973;28:395-403 18 Cutting J. The psychology of schizophrenia. London: Churchill Livingstone, 1985 19 McEvoy JP, Aland J, Wilson WH, Guy W, Hawkins L. Measuring chronic schizophrenic patients' attitudes toward their illness and treatment. Hosp Community Psychiatry 1981;32:856-8 20 Lin IF, Spiga R, Fortsch W. Insight and adherence to medication in chronic schizophrenics. J Clin Psychiatry 1979;40:430-2 21 Van Putten T, Crumpton E, Yale C. Drug refusal in schizophrenia and the wish to be crazy. Arch Gen Psychiatry 1976;33:1443-6 22 Bartk6 G, Herczeg I & Zfdor G. Clinical ymptomatology and drug compliance in schizophrenic patients. Acta Psychiatr Scand 1988;77:74-6 23 Soskis DA, Bowers MB. The schizophrenic experience: a follow-up study of attitude and posthospital adjustment. J Nerv Ment Dis 1969;149:443-9 24 McGlashan TH, Carpenter WT. Does attitude toward psychosis relate to outcome? Am JPsychiatry 1981;138: 797-801 25 Roback HB, Abramowitz SI. Insight and hospital adjustment. Can J Psychiatry 1979;24:233-6 26 Pilowsky I. A general classification of abnormal illness behaviour. Br J Med Psychol 1978;51:131-7 27 Greer S. Cancer and the mind. Br J Psychiatry 1983;143:535-543 28 Szasz TS. The age of madness. London: Routledge & Kegan Paul, 1974:166-7
29 Reid JR, Finesinger JE. The role of insight in psychotherapy. Am J Psychiatry 1952;108:726-34 30 Lishman WA. Organic psychiatry. Oxford: Blackwell Scientific, 1987 31 Robinson MF, Freeman W. Psychosurgery and the self. New York: Grune & Stratton, 1954 32 McGlynn SM, Schacter DL. Unawareness of deficits in neuropsychological syndromes. J Clin Exp Neuropsychol 1989;11:143-205 33 Bleuler E. Autistic thinking. Am J Insanity 1913;69: 873-86 34 Mullen P. Phenomenology of disordered mental function. In: Essentials of postgraduate psychiatry (ed P. Hill, R. Murray & G. Thorley). London: Academic Press, 1979 35 American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 3rd edn. Washington, DC: American Psychiatric Association, 1980 36 Kendler KS, Glazer WM, Morgenstern H. Dimensions of delusional experience. Am J Psychiatry 1983;140: 466-9 37 Brett-Jones JR. Recovery from delusions: a methodological study. M.Phil thesis. London: Institute of Psychiatry, 1984 38 Garety PA. Delusions: problems of definition and measurement. Br J Med Psychol 1985;58:25-34 39 Kraupl Taylor F. On pseudo-hallucnations. Psychol Med
1981;11:265-71 40 Wing JK, Cooper JE, Sartorius N. Measurement and Classification of Psychiatric Symptoms. Cambridge: Cambridge University Press, 1974 41 Strauss JS. Hallucinations and delusions as points on continua function. Arch Gen Psychiatry 1969;21:581-6 42 Garety PA, Hemaley DR. Characteristics of delusional experience. Eur Arch Psychiatry Neurol Sci 1987;26: 294-8 43 Sacks MH, Carpenter WT, Strauss JS. Recovery from delusions. Arch Gen Psychiatry 1974;30:-117-120 44 Maher B, Ross JS. Delusions. In: Comprehensive handbook of psychopathology Adams HE, Sutker PB, eds. New York: Plenum Press, 1984:383409 45 Milton F, Patwa VK, Hafner RJ. Confrontation vs belief modification in persistently deluded patients. Br JMed Psychol 1978;51:127-30 46 Winters KC, Neale JM. Delusions and delusional thinking in psychotics: a review of the literature. Clin Psychol Rev 1983;3:227-53 47 Arieti S. The interpretation of schizophrenia, 2nd edn. London: Crosby Lockwood Staples, 1974 48 Cutting J, Murphy D. Schizophrenic thought disorder: a psychological and organic interpretation. Br J Psychiatry 1988;152:310-319 49 Hemsley DR, Garety PA. The formation of delusions: a Bayesian analysis. Br J Psychiatry 1986;149:51-6 50 Huq SF, Garety PA, Hemsley DR. Probabilistic judgements in deluded and non-deluded subjects. Q J Exp Psychol 1988;40A:801-2 51 Freud S. The dissection of the psychical personality. In The New Introductory Lectures on Psychoanalysis. Standard Edition (1964) 22 57-80. London: Hogarth Press, 1933 52 Sperry RW. Hemisphere disownnection and the unity of conscious awareness. Am Psychol 1968;23:723-33 53 Bauer RM. Autonomic recognition of names and faces in prosopagnosia: a neuropsychological application of the guilty knowledge test. Neuropsychologia 1984;22: 457-69 54 Rachman S. Irrational thinking, with special reference to cognitive therapy. Advances in Behavioural Therapy 1983;5:63-88 55 Marks IM, Mishan J. Dysmorphophobic avoidance from disturbed bodily perception: a pilot study of exposure therapy. Br J Psychiatry 1988;152:674-8 56 Taylor PJ. Motives for offending among violent and psychotic men. Br J Psychiatry 1985;147:491-8 (Accepted 6 September 1989)
329
