Mental Disorders: Genetic Aspects

In International Encyclopedia of the Social Sciences, Macmillan 1968

    People who are related to one another by blood tend to resemble one another in, among other things, their mental make‑up and their liability to mental illness. Both genetic and environmental factors may play a part in this resemblance. The most widely accepted view of the nature of the interaction between heredity and environment has been called the diathesis‑stress theory (Rosenthal 1963). In its application to mental illness this view suggests that the susceptibility to mental illness, insofar as it is genetically based, varies along a con­tinuum ranging from high to low extremes, most people clustering about an average of moderate susceptibility. Environmental stresses, also, vary from severe to slight. Accordingly, when a mental breakdown occurs, a combination of both factors is involved; thus, we should expect a high rate of breakdown among normal individuals subjected to severe stress, and a high rate also among very susceptible people placed under even mild stress. Such a quantitative relationship has been shown to hold in fact, e.g., in the neurotic illnesses of combat troops (Symonds 1943). Whether psy­chotic illnesses follow the same general law is a matter more difficult to decide.

Personality deviations and neurotic illness

    Twin studies. An important part of the work in the field of personality deviations and neurotic illness has been studies of twins. One‑egg, or monozygotic (MZ), twins, whose entire genetic equipment is identical, are of the same sex and are very much alike in physical characteristics. Two­egg, or dizygotic (DZ), twins ordinarily resemble each other no more than any pair of brothers or sisters and are as likely to be of opposite sexes as of the same sex. As a rule only the same‑sexed DZ twin pairs are taken by investigators for com­parison with MZ pairs. Twin pairs are said to be concordant when it is found that the twin of a proband (index case) with a particular deviation also shows the same anomaly. Concordance rates are usually given as percentages. Genetic hypoth­eses lead one to suppose that concordance rates should be much higher in MZ than in DZ pairs and that variability within MZ pairs should be smaller than within DZ pairs. Table 1 lists the percentage of concordances found for a variety of conditions, in which the statistics are based on at least thirty pairs.

    Criminality and delinquency. Much effort has been put into the investigation of criminality and behavior disorder. The results of Rosanoff in ju­venile delinquency are noteworthy (Rosanoff et al. 1934; 1941). There are high rates of concordance both in MZ and in DZ twin pairs, and little dif­ference between them. This suggests that the simi­larity in the twins' behavior is due to common fac­tors in their environment. It is at least possible that each of the twins constitutes part of the stress fac­tor for his twin partner. The same suggestion arises from the observations of behavior disorder and neurotic traits in school children. Shields found that in these children the degree of neurotic re­action was more noticeably related to environmen­tal factors than genetic constitution; the hereditary factor showed itself in the type of reaction (1954).

    Adult neuroses. Concordance rates are lower in neurotic adults, both in the MZ and in the DZ pairs. The greatly increased variability within both kinds of pairs can be put down to the much wider range of experience, and wider variety of stresses, to which adults are subjected.

    Male homosexuality. At the opposite extreme are Kaflmann's findings of 100 per cent concordance in MZ pairs, as against 12 per cent concordance in DZ pairs, for male homosexuality. This would suggest that in these cases the genetic factors ac­count for the greater part of the variance. Some caution in interpretation is needed, however. The importance that the genetic contribution acquires here may be due to the fact that there was a pre­dominance of the more constitutional type of homo­sexual in the sample studied by Kailmann and his team.

    Studies of relatives. Attempts to estimate the importance of hereditary factors in causing neu­rotic illness have been made by investigating the frequency of such illnesses among the relatives of neurotic patients. Findings have varied greatly from observer to observer. One of the early work­ers, Brown (1942), started by investigating the first‑degree relatives of patients who had been di­agnosed as suffering from obsessional neurosis, anxiety neurosis, and hysteria, as well as those of a control group. Among the relatives of all groups he found individuals suffering from obsessional neurosis, anxiety neurosis, and hysteria, as well as from other personality deviations of a kind not easily named and classified. There were three sig­nificant findings: the relatives of the control group had much less psychiatric abnormality than the relatives of any of the other three groups; all three diagnoses were represented among the relatives of all three patient groups; among those relatives who were classifiable under the three named diagnoses, it was found that there was a tendency for them to be in the same diagnostic category as the related patient. For example, of the nine obsessional rela­tives discovered, seven were related to the obses­sional patients. This finding suggests a certain degree of specificity, which is best seen in the in­vestigations which have been made on the relatives of obsessional patients (Luxenburger 1930; Lewis 1936; Rüdin 1953). Among the 100 parents of 50 obsessional patients Lewis found that 37 showed pronounced obsessional traits in one form or an­other; 21 per cent of the 206 siblings also showed obsessional traits.

    At the opposite pole are the findings in the rela­tives of patients diagnosed as suffering from "hys­teria." The best family study was made by Ljung­berg (1957), who found that among the fathers, brothers, and sons of hysterics, 2 per cent, 3 per cent, and 5 per cent respectively were themselves hysterics; and among the mothers, sisters, and daughters, 7 per cent, 6 per cent, and 7 per cent respectively. His observations also suggested that hysterical symptoms were not necessarily associ­ated with hysterical personalities, and of the 363 hysterics whose personality structures were ana­lyzed, 55 per cent were found to be nondeviant.

    Similar conclusions were reached by the writer (Slater 1961) from a study of 24 pairs of twins, 12 MZ and 12 DZ, in which the proband had been diagnosed as suffering from hysteria. In none of these pairs was there a co‑twin who had ever been diagnosed as suffering from hysteria, though ab­normalities of personality and psychiatric illness were common. Among the relatives of these pairs the incidence of hysteria was even lower than in Ljungberg's material, and the anomalies found to be most noticeably in excess were manic‑depres­sive and endogenous affective psychoses.

    It seems probable that environmental factors are more important than genetic ones in determining whether or not a man breaks down with a neurotic illness. But it would seem that genetic factors in­fluence the predisposition to such breakdown, help to determine whether the personality is a stable one, and, in the event of breakdown, have a large effect on the type of symptoms which are likely to be shown.

Manic‑depressive illness

    The risk of affective psychoses in the first‑degree relatives of manic‑depressive patients is shown in Table 2. It will be seen that there is much variation in the data reported by different observers. Never­theless, these risk figures are all very much higher than would be expected of a sample taken from the general population, in which the incidence level is probably of the order of 0.4 per cent.


Approximately 15 per cent of the first‑degree relatives of manic‑depressives may themselves have affective disorders of the same generic group.

    Single‑gene explanation. One explanation sug­gests that there is a single dominant autosomal gene which predisposes a spontaneous variation in mood. Most people with such a tendency are likely to remain healthy throughout their lives, though their cyclothymic temperament may be clearly rec­ognizable both to themselves and to their families. If such cyclothymic individuals are subjected to stresses, the spontaneous swing of mood into de­pression or elation may go so far and last so long that medical treatment becomes necessary; and then the patient may be stigmatized as a "manic­depressive." The genetic hypothesis proposes, in fact, to account for only a part of the causation. Environmental factors and threshold effects must also be playing a part. The theory requires that 50 per cent of the first‑degree relatives of manic­depressives should be gene carriers; if only 15 per cent of the relatives show themselves as such, this can be put down to low penetrance of the gene.

    Polygenic explanation. A single‑gene hypothe­sis, however, is not the only possibility; polygenic inheritance is an alternative explanation. Edwards (1960; 1963) has drawn attention to the fact that, in the case of common conditions, it is not easy to distinguish between the expected consequences of a single gene with diminished penetrance and of multifactorial inheritance with a threshold ef­fect. Assuming that the predisposition to a condi­tion, such as schizophrenia or diabetes, is quanti­tatively graded, with a normal distribution, Edwards suggests that when p = the frequency of the dis­order, the incidence of the disorder in the first‑degree relatives of persons suffering from the disorder will be approximately V. If the frequency of manic‑depressive illness in the general popula­tion is approximately 0.004, then the frequency of manic‑depressive illness in the first‑degree rela­tives of manic‑depressives should be about 6 per cent to 7 per cent. The observations are about double that figure, but it is not possible to say that observation and expectation, on the multifac­tonal hypothesis, are irreconcilable.


    Extensive investigations of the hereditary factor in schizophrenia have been made by Kallmann and his associates in the New York State Psychiatric Institute and Hospital (Columbia University). A synopsis of the results obtained is given in Table 3. These risk figures should be compared with the estimated risk of schizophrenia for a member of the general population of 0.9 per cent. Environ­mental effects show up: association with a schizo­phrenic proband in the same home virtually doubles the risk of schizophrenia for step‑sibs and spouses; and there is a greater risk for the non~ separated MZ co‑twin of a schizophrenic than for an MZ co‑twin who has lived apart from the pro­band for five years or more. However, the table also shows the risk of schizophrenia running up steeply with increasingly close degrees of blood relation­ship. In view of these figures it is difficult and, the writer feels, unrealistic to dispute the conclusion that genetic factors play a significant role in the causation of schizophrenia.


    The risk figures published by Kallmann are some­what higher than those obtained by other workers but are in the main reconcilable with them. The writer, for instance (Slater 1953), found a risk of 76 per cent for the MZ co‑twin and 14 per cent for the DZ co‑twin, in a sample of 41 MZ and 115 DZ pairs. However, it is noteworthy that Essen‑Möller (1941) in Sweden found larger differences within MZ pairs than were found by other workers, and investigations in Norway and Finland show tend­encies in the same direction. Thus Kringlen (1966) collected 50 MZ and 94 same‑sexed pairs, with concordance rates of 38 per cent and 14 per cent; and Tienari (1963) found that none of the 16 male MZ pairs he studied were concordant. It is possible that what is diagnosed as schizophrenia in Scandi­navia is not quite the same as schizophrenia in Germany, Britain, the United States, and Japan. That there may be peculiar features about the gene distributions in northern lands is also suggested by the work of Böök (1953). In a remote part of Sweden north of the Arctic circle, in a population of farmers and lumbermen, he found a high inci­dence of schizophrenia, in a predominantly cata­tonic form; the incidence of schizophrenia in the relatives of probands, however, was also high, sug­gesting an intermediate gene with 20 per cent penetrance in the heterozygote. The possible preva­lence of different genetic predisposing factors for schizophrenia in different parts of the world is a possibility which deserves investigation.

    The findings of Tienari, by themselves, are anom­alous and should not be taken as throwing doubt on the results obtained by others; they should be considered in relation to those of other observers, summarized in Table 4. In this table the figures relating to schizophrenia are derived from work in Germany, the United States, Sweden, and England. To these should be added the results obtained by the Japanese workers Kurihara (1959) and Inouye (1961). Inouye found concordance for schizo­phrenia in 60 per cent of 55 MZ pairs and in 12 per cent (two pairs) of the DZ pairs. Kurihara also found that 29 of 45 MZ pairs were concordant for schizophrenic symptomatology, but none of the 9 DZ pairs were.

      The twin work on schizophrenia has been ana­lyzed and discussed critically by Rosenthal in a number of papers (1962a; 1962b; 1963). He con­cludes that concordance rates have been artificially inflated by the sampling methods employed. Cases have been largely taken from standing populations and include an unrepresentative proportion of se­vere cases; if genetic factors are connected with reduced chances of remission (which has yet to be shown), this sampling would obviously bias the results. Clearly, sampling from consecutive admis­sions, or better still from birth registers, would be an improvement. Rosenthal criticizes standards of diagnosis, both of zygosity and clinical classifi­cation, and considers that these diagnoses should be made independently of one another by different observers. Not all recorded work is equally open to such criticism; it is, for instance, a considerable safeguard to publish protocols in full, as the writer did, to make them available to rediagnosis by the reader.

    Etiological theories. Rosenthal classifies the etiological theories of schizophrenia into (1) mono­genic‑biochemical, (2) diathesis‑stress, and (3) life‑experience theories, his own views inclining to a theory of the second type. This is no place for the discussion of the difficult problems involved, but the writer inclines to a theory of the first type. It can be shown (Slater 1958) that a monogenic theory fits fairly well with the empirically obtained figures for the frequency of schizophrenia in the siblings of schizophrenics, in the children of one schizophrenic parent, and in the children of par­ents both of whom had schizophrenic illnesses. These data can be reconciled with a gene of inter­mediate type, manifesting itself in all homozygotes but in only 26 per cent of heterozygotes; all but 3 per cent of schizophrenics would be heterozygous for the gene. This hypothesis clearly involves a massive environmental contribution in the causa­tion of manifest illness and therefore differs from theories of type (2) only in supposing that the element of specificity in determining the type of psychosis is provided by the genetic constitution. One may expect that decisive support or refutation of type (1) theories will depend on biochemical investigations.

Presenile and senile dementias

    Inheritance in Huntington's chorea is dependent on an autosomal dominant gene. Its incidence in the sexes is about equal. The age of onset, according to Panse (1942), extends from early childhood to the late sixties, with a mean at age 36; Wendt and his colleagues estimate the mean age of onset at 44 (1960). These estimates mean that most of the children of gene carriers are born before the parent has developed the disease. Elimination of the patho­genic gene by processes of natural selection is, ac­cordingly, very slow. The disease is slowly progres­sive and fatal, with a mean duration of 13 years (Wendt et al. 1960).

    Unusual forms of presenile dementia are Pick's disease and Alzheimer's disease; both have a ge­netic basis. Compared with Huntington's chorea, they occur later in life, with a mean age of onset of 55, and with about seven years as the mean duration. Although the conditions are distinct path­ologically, they are often difficult to distinguish clinically. According to Sjögren (Sjögren et al. 1952), the genetic factor in Pick's disease is most probably a dominant major gene, its manifestation subject to modifying genes; in Alzheimer's disease Sjögren thinks multifactorial inheritance more probable.

    The problem of genetic determination in senile dementia is even more difficult. The most generally accepted opinion in the past has been that senile dementia is but one aspect of senescence and that specific genetic causation is improbable. However, this established viewpoint has been challenged by the work of Larsson, Sjögren, and Jacobson (1963). In a large study in Stockholm they found that se­nile dementia was not correlated with senescence. The relatives of patients suffering from senile de­mentia were not more than normally subject to other conditions, although their risk of senile de­mentia itself was increased; no instances of Pick's or Alzheimer's disease were found among them. There did not appear to be special factors for lon­gevity whose presence or absence was connected with senile dementia. No evidence could be found of environmental factors of a sociomedical kind playing a part in determining the onset of senile dementia. Furthermore, there was no secular change in the incidence of the disease. Particularly in favor of an explanation in terms of a single gene rather than multifactorial inheritance were the variations in geographical distribution and the absence of intermediate states between senile de­mentia and normal aging in the siblings and chil­dren of the probands.

    The morbidity risk for senile dementia was found to be greatly increased among the relatives of the patients. These researchers consider that the best working hypothesis for the explanation of their findings is that of an autosomal major dominant gene. This gene would be subject to dimin­ished penetrance, the manifestation rate increasing with age. Only a minority even of the gene carriers would ever develop senile dementia; and since the calculated gene frequency was 0.12, the great bulk of the population would be immune.



BANSE, J. 1929 Zum Problem der Erbprognosebestim­mung: Die Erkrankungsaussichten der Vettern und Basen von Manisch‑Depressiven. Zeitschrift für die gesamte Neurologie und Psychiatrie 119:576‑612.

BÖÖK, J. A. 1953 A Genetic and Neuropsychiatric Inves­tigation of a North‑Swedish Population: I. Psychoses. Acta genetica et statistica medica (Basel) 4:1‑100.

BORGSTROEM, C. A. 1939 Eine Serie von kriminellen Zwillingen. Archiv für Rassen‑ und Gesellschafts­biologie 33:334‑343.

BROWN, FELIX W. 1942 Heredity in the Psychoneuroses. Royal Society of Medicine, Proceedings 35:785‑790.

EDWARDS, J. H. 1960 The Simulation of Mendelism. Acta genetica et statistica medica (Basel) 10:63‑70.

EDWARDS, J. H. 1963 The Genetic Basis of Common Disease. American Journal of Medicine 34:627‑638.

Es SEN‑MÖLLER, ERIK 1941 Psychiatrische Untersuchungen an einer Serie von Zwillingen. Acta psychiatrica et neurologica scandinavica Supplement 23.

ESSEN‑MÖLLER, ERIK 1963 Twin Research and Psychiatry. Acta psychiatrica scandinavica 39, fasc. 1:65‑77.

FONSECA, ANTONIO F. DA 1959 Análise heredo‑clinica das perturbacöes afectivas: Estudo de 60 pares de gémeos a seus consanguineos. Universidad de Porto (Portugal): Faculdade de Medicina.

INOUYE, Eiji 1961 Similarity and Dissimilarity of Schizophrenia in Twins. Volume 1, pages 524‑530 in World Congress of Psychiatry, Third, Montreal, Pro­ceedings. Univ. of Toronto Press.

KAIJ, LENNART 1960 Alcoholism in Twins: Studies on the Etiology and Sequels of Abuse of Alcohol. Stock­holm: Almqvist & Wiksell.

KALLMANN, FRANZ J. 1950 The Genetics of Psychoses: An Analysis of 1,232 Twin Index Families. American Journal of Human Genetics 2:385‑390.

KALLMANN, FRANZ J. 1952 Comparative Twin Studies on the Genetic Aspects of Male Homosexuality. Jour­nal of Nervous and Mental Disease 115:283‑298.

KRANZ, HEINRICH 1936 Lebensschicksale krimineller Zwillinge. Berlin: Springer.

KRINGLEN, EINAR 1966 Schizophrenia in Twins: An Epidemiological‑Clinical Study. Psychiatry 29:172‑184.

KTJRIHARA, M. 1959 A Study of Schizophrenia by Twin Method. Psychiatria et neurologia japonica (Seishin shinkeigaku zasshi) 61:1721‑1741. ‑ Text in Japa­nese; title and summary in English.

LANGE, JOHANNES (1929) 1931 Crime as Destiny: A Study of Criminal Twins. London: Allen & Unwin. ‑ First published as Verbrechen als Schicksal.

LARSSON, TAGE; SJÖGREN, TORSTEN; and JACOBSON, GEORGE 1963 Senile Dementia: A Clinical Sociomedical and Genetic Study. Copenhagen: Munsgaard.

LEWIS, AUBREY 1936 Problems of Obsessional Illness. Royal Society of Medicine, Proceedings 29:325‑336.

LJUNGBERG, L. 1957 Hysteria: A Clinical, Prognostic and Genetic Study. Acta psychiatrica et neurologica scandinavica Supplement 112.

LUXENBURGER, H. 1930 Psychiatrisch‑neurologische Zwillingspathologie. Zeitschrift für die gesamte Neurologie und Psychiatrie 56:145‑180.

PANSE, FRIEDRICH 1942 Die Erbchorea: Eine klinisch‑genetische Studie. Leipzig: Thieme.

RÖLL, A.; and ENTRES, J. L. 1936 Zum Problem der Erbprognosebestimmung: Die Erkrankungsaussichten der Neffen und Nichten von Manisch‑Depressiven. Zeitschrift für die gesamte Neurologie und Psychiatrie 156:169‑202.

ROSANOFF, AARON J.; HANDY, L. M.; and PLESSET, I. R.1941 The Etiology of Child Behavior Difficulties, Juvenile Delinquency and Adult Criminality, With Special Reference to Their Occurrence in Twins. Sac­ramento: California State Printing Office.

ROSANOFF, AARON J.; HANDY, L. M.; and ROSANOFF, I. A. 1934 Criminality and Delinquency in Twins. Journal of Criminal Law and Criminology 24:923‑934.

ROSENTHAL, DAVID 1962a Familial Concordance by Sex With Respect to Schizophrenia. Psychological Bulletin 59:401‑421.

ROSENTHAL, DAVID 1962b Problems of Sampling and Diagnosis in the Major Twin Studies of Schizophrenia. Journal of Psychiatric Research 1: 16‑34.

ROSENTHAL, DAVID (editor) 1963 The Genain Quadrup­lets: A Case Study and Theoretical Analysis of Hered­ity and Environment in Schizophrenia. New York: Basic Books.

RÜDIN, EDITH 1953 Ein Beitrag zur Frage der Zwangs­krankheit, insbesondere ihrer hereditären Beziehungen. Archiv für Psychiatrie und Nervenkrankheiten 191: 14‑54.

SHIELDS, JAMES 1954 Personality Differences and Neu­rotic Traits in Normal Twin Schoolchildren: A Study in Psychiatric Genetics. Eugenics Review 45:213‑245.

SHIELDS, JAMES; and SLATER, ELIOT (1960) 1961 Hered­ity and Psychological Abnormality. Pages 298‑343 in Hans J. Eysenck (editor), Handbook of Abnormal Psychology: An Experimental Approach. New York: Basic Books.

SJÖGREN, TORSTEN 1948 Genetic‑Statistical and Psychi­atric Investigations of a West Swedish Population. Acta psychiatrica et neurologica scandinavica Supple­ment 52.

SJÖGREN, TORSTEN; SJÖGREN, HAKON; and LINDGREN, ARE G. H. 1952 Morbus Alzheimer and Morbus Pick:

A Genetic, Clinical and Patho‑anatomical Study. Acta psychiatrica et neurologica scandinavica Supplement 82.

SLATER, ELIOT 1938 Zur Erbpathologie des manisch­depressiven Irreseins: Die Eltern und Kinder von Manisch‑Depressiven. Zeitschrift für die gesamte Neurologie und Psychiatrie 163:1‑47.

SLATER, ELIOT 1953 Psychotic and Neurotic Illnesses in Twins. Medical Research Council Special Report No. 278. London: H.M. Stationery Office.

SLATER, ELIOT 1958 The Monogenic Theory of Schizo­phrenia. Acta genetica et statistica medica (Basel) 8:50‑56.

SLATER, ELIOT 1961 The Thirty‑fifth Maudsley Lecture: Hysteria 311. Journal of Mental Science 107:359‑381.

STENSTEDT, AKE 1952 A Study in Manic‑Depressive Psychosis. Acta psychiatrica et neurologica scandinavica Supplement 79.

STRÖMGREN, ERIK 1938 Beiträge zur psychiatrischen Erblehre. Acta psychiatrica et neurologica scandinavica Supplement 19.

STUMPFL, FRIEDRICH 1936 Die Ursprünge des Verbrechens dargestellt am Lebenslauf von Zwillingen. Leipzig: Thieme.

SYMONDS, C. P. 1943 The Human Response to Flying Stress. British Medical Journal [1943]:703‑706, 740­‑744 Lecture 1: "Neurosis in Flying Personnel." Lecture 2: "The Foundations of Confidence."

TIENARI, P. 1963 Psychiatric Illnesses in Identical Twins. Acta psychiatrica scandinavica 39 (Supplement 171). ‑ The entire issue is devoted to Tienari's study.

WENDT, G. G.; LANDZETTEL, I.; and SOLTH, K. 1960 Krankheitsdauer und Lebenserwartung bei der Huntingtonschen Chorea. Archiv für Psychiatrie und Nervenkrankheiten 201:298‑312.