Genetic Factors In Schizophrenia And Bipolar Disorder Research Paper

This aticle first considers diagnostic issues, including the most widely used diagnostic criteria for schizophrenia, bipolar disorder, and related disorders. Evidence for genetic factors in schizophrenia and bipolar disorder is reviewed next, including some complicating factors, such as the likely presence of etiologic heterogeneity and the interaction of genes with environmental stressors. Issues of genetic counseling are then considered. The chapter concludes with a discussion of evidence that genes for schizophrenia and bipolar disorder may have a ‘silver lining,’ in terms of increased creative potential.

1. Diagnostic Issues

1.1 Schizophrenia And Related Disorders

The criteria currently most widely used in diagnosing schizophrenia, bipolar disorder, and related disorders are those described in the most recent (4th) edition of the Diagnostic and Statistical Manual (DSM-IV) of the American Psychiatric Association (1994). Briefly summarized, the diagnostic criteria for schizophrenia listed in DSM-IV require two or more of the following five characteristic symptoms: hallucinations; delusions; disorganized speech; grossly disorganized behavior; and negative symptoms, such as flattened affect. In addition, the patient must have shown significant deterioration of functioning in areas such as occupational or interpersonal relations, and have had continuous signs of illness for at least six months, including at least one month with the characteristic symptoms.

DSM-IV also recognizes several related disorders that manifest milder forms of certain symptoms that are often seen in schizophrenia. Schizotypal personality disorder, for example, is used for persons who have several characteristic features (e.g., magical thinking, recurrent illusions, and peculiar behavior or speech). If a person does not have such schizotypal eccentricities but has shown several signs of marked disinterest in interpersonal relationships (e.g., extreme aloofness, absence of any close friends), then a diagnosis of schizoid personality disorder is given. In paranoid personality disorder, there is a broad pattern of extreme suspiciousness, as shown by several signs (e.g., irrational fears that others wish to harm one). Family and adoption studies suggest that these personality disorders are part of a ‘spectrum’ of disorders that are genetically related to schizophrenia proper.

1.2 Bipolar Disorder And Related Disorders

In DSM-IV, the diagnosis of ‘bipolar I’ disorder requires a history of at least one episode of mania, which is defined as a period in which a person’s mood is ‘abnormally and persistently’ expansive, elevated, or irritable. In addition, diagnosis of bipolar disorder requires that the mood change either involve psychotic features which last at least a week, or lead to hospitalization. The disturbance in mood must be severe enough to disrupt seriously social or occupational functioning, and/or to require hospitalization. The manic episode must also involve at least three (four, if only an irritable mood is present) of seven symptoms: (a) greatly inflated self-esteem, (b) increased activity or restlessness, (c) unusual talkativeness, (d) racing thoughts or flight of ideas, (e) decreased need for sleep, (f ) distractibility, and (g) extremely risky actions (such as buying sprees or reckless driving) whose potentially dangerous consequences are not appreciated. The criteria for a hypomanic episode are essentially the same as those for a manic one, except that the symptoms are neither psychotic nor so severe that they severely impair social functioning or require hospitalization. The term bipolar is potentially confusing, because it does not require a history of major depression, even though most persons with bipolar disorder will also have experienced episodes of major depression. (By contrast, in major depressive disorder, a person has experienced an episode of major depression, but not an episode of mania.)

As in the case of schizophrenia, there appears to be a ‘spectrum’ of affective disorders that are genetically related to bipolar disorder but that have symptoms which are milder than those found in frank bipolar disorder. Thus DSM-IV also includes (a) bipolar II disorder (in which a patient has experienced a hypomanic, rather than a manic episode, as well as an episode of major depression) and (b) cyclothymic disorder, which involves a history of multiple hypomanic episodes as well as multiple periods of depressive symptoms (but not major depression). Bipolar disorder is also often accompanied by other concurrent, or ‘co-morbid,’ disorders, such as anxiety disorders, or alcohol and substance abuse. Moreover, even when symptoms of mania and depression are in remission, patients with a history of bipolar disorder often still meet criteria for personality disorders, particularly those with narcissistic or histrionic features.

1.3 Differential Diagnosis

A diagnosis of schizophrenia requires that diagnoses of mood disorders and schizoaffective disorders be excluded. Diagnoses of schizophrenia, mood disorders, and schizoaffective disorders all require the exclusion of neurologic syndromes caused by factors such as substance abuse, medications, or general medical condition. Until rather recently there has been a tendency, particularly in the USA, for mania with psychotic features to be misdiagnosed as schizophrenia. Accurate differential diagnosis is crucial, because misdiagnosis (and resultant inappropriate treatment) may result, on the one hand, in patients being needlessly exposed to harmful side effects of medication or, on the other hand, being deprived of appropriate medications that may alleviate unnecessary suffering and save patients’ jobs, marriages, even their lives. Mania, for example, often responds well to medications (particularly lithium and certain anticonvulsants such as carbamazepine and valproate) that are less effective in treating schizophrenia. It is important, moreover, to diagnose and treat schizophrenia and bipolar disorder as early as possible in the course of the illness because there is increasing evidence that these illnesses (and their underlying brain pathologies) tend to worsen if left untreated (e.g., Wyatt 1991).

2. Evidence For Genetic Factors In Schizophrenia

There are several complementary lines of evidence for an important role of genetic factors in the etiology of schizophrenia. First, there are converging lines of evidence from family, twin, and adoption studies that the risk for schizophrenia is greatly increased among schizophrenics’ biological relatives. Second, investigators have, in recent years, increasingly marshalled techniques from molecular genetics to look for more direct evidence of genetic factors in schizophrenia.

2.1 Family, Twin, And Adoption Studies

A person’s risk of developing schizophrenia increases, on average, with his or her increasing degree of genetic relatedness to a schizophrenic patient (e.g., Matthysse and Kidd 1976, Holzman and Matthysse 1990, Torrey et al. 1994). The risk of developing schizophrenia over a person’s lifetime is about 0.8 percent for people in the general population, though there is a several-fold variation in the prevalence of schizophrenia across different populations that have been studied around the world (e.g., see Torrey 1987). By contrast, a person’s lifetime risk is 5–10 percent if the person has a first-degree relative with schizophrenia, and is much higher—nearly 50 percent in some studies—if a person is the monozygotic (genetically identical) twin of a schizophrenic patient. While these risk figures are consistent with genetic transmission, they are not conclusive, because the degree of genetic resemblance among relatives tends to parallel their level of exposure to similar environments.

Twin studies have rather consistently reported concordance rates for schizophrenia in monozygotic (MZ) twins that are several times higher than those for dizygotic (DZ) twins (e.g., Gottesman et al. 1987, Torrey et al. 1994). It is also noteworthy that the schizophrenia concordance rate for MZ twins reared apart is quite close to that for MZ twins reared together. On the other hand, the number of such MZ twins reared apart is rather small. Moreover, it is unclear whether these twins who were reared in different settings may still have had significant contact with each other after they were separated, so that the separation of shared genetic and environmental factors may not have been complete.

The most conclusive available evidence for genetic factors in schizophrenia, therefore, comes from adoption studies. For example, Heston (1966) studied 47 adult adoptees who had been born in the USA to a schizophrenic mother, but separated shortly after birth. Five of these ‘index’ adoptees were found to have subsequently developed a diagnosis of schizophrenia, vs. none of 50 matched control adoptees who had been born to demographically matched, but psychiatrically healthy, mothers.

More systematic adoption studies of schizophrenia have been carried out in Scandinavia. In Denmark, for example, Kety et al. (1994) were able to identify all individuals in the entire country who had been adopted away from their biological parents at an early age and subsequently were hospitalized with a diagnosis of schizophrenia. For each of these 74 schizophrenic adoptees, a ‘control’ adoptee was identified who was closely matched for age, gender, and the socioeconomic status of the adoptive home. The control adoptees’ biological parents had not been hospitalized for mental illness. Psychiatric diagnoses of over 1100 of the adoptees’ respective biological and adoptive relatives were made after careful review of psychiatric interviews and records. Significantly higher rates of schizophrenia were found in the biological (but not in the adoptive) relatives of schizophrenic adoptees than in the biological relatives of control adoptees (5.0 percent vs. 0.4 percent). The prevalence of schizotypal personality disorder was also significantly elevated among the schizophrenic adoptees’ biological relatives (Kendler et al. 1994). Moreover, rates of schizophrenia and related ‘spectrum’ disorders were significantly elevated even in the schizophrenic adoptees’ biological paternal half-siblings, who had not even shared the same womb as the schizophrenic adoptees.

2.2 Association And Linkage Studies

In recent decades, advances in molecular genetics have enabled researchers to look more directly for genetic factors in schizophrenia. One strategy is to investigate ‘candidate genes’ for which there is a theoretical reason to suspect a role in schizophrenia. Thus several groups of investigators have looked for genes that influence susceptibility to certain infectious agents, since exposure to these agents, particularly during pre or perinatal development, appears to increase risk for schizophrenia. For example, in a number of epidemiological studies, increased risk of exposure to influenza during the middle trimester of gestation has been found to be associated with increased risk of schizophrenic outcome. Individuals’ genotypes can powerfully affect their immunologic response to infections (and their mothers’ ability to combat infections while they are still in utero). Several studies have found that certain alleles of genes that play an important role in immune function, such as those in the HLA complex, are more prevalent in schizophrenia. McGuffin (1989) reviewed nine such studies and concluded that there was a highly significant association between the HLA A9 allele and risk for the paranoid subtype of schizophrenia. Murray et al. (1993) suggested that prenatal exposure to influenza may increase risk of schizophrenia in the offspring because, in genetically susceptible mothers, the flu virus stimulates production of maternal antibodies that cross the placenta and disrupt fetal brain development.

Even if one does not have a good ‘candidate’ gene for a disorder such as schizophrenia, however, it is still possible to apply the more indirect strategy of genetic ‘linkage.’ This strategy makes use of the fact that genes that are located very near one another on the same chromosome (and are therefore said to be closely ‘linked’) will tend to be inherited together. As the result of recent advances in molecular genetics, there are now thousands of identified genes that can be used to mark various regions of different chromosomes. One can then study large numbers of ‘multiplex’ families in which there are at least two family members with schizophrenia, in order to examine whether there is a significant tendency for schizophrenia and alleles for particular marker genes to be transmitted together within the same family. In principle, genetic linkage studies provide an elegant approach that makes it possible to identify disease genes whose role in a disease such as schizophrenia is a complete surprise to investigators (and thus would never have been chosen as ‘candidate’ genes for association studies).

There are, however, some practical difficulties with linkage studies. One difficulty is that, because hundreds of different marker genes can be examined, there is a rather high probability of obtaining a spurious, or ‘false positive,’ linkage finding by chance. In order to screen out such false positive findings, it is important to determine whether interesting linkage results can be confirmed in new, independent, studies involving large numbers of multiplex families. In fact, linkage findings implicating genes on a particular region of chromosome 6 as risk factors for schizophrenia have recently been reported by several different research groups (e.g., see review by Gershon et al. 1998).

2.3 Etiologic Heterogeneity And Genotype–Environment Interactions

The search for genetic factors is complicated by several factors. It is likely, for example, that disorders such as schizophrenia are etiologically heterogeneous. That is, it is probable that the syndrome of schizophrenia can be produced by a number of different combinations of genes and/or environmental factors. Thus, while each of a number of different genes may well increase risk for schizophrenia, it is likely that no single gene is necessary for the production of most cases. One approach to the problem of heterogeneity is to identify characteristics that distinguish specific, genetically more homogeneous, subtypes of schizophrenia or bipolar disorder. Maziade et al. (1994), for example, found evidence for linkage at a locus on chromosome 11 with schizophrenia in one, but not the others, of several large pedigrees that they examined. The schizophrenics in the extended family that did show linkage were distinguished from the families that did not by having a particularly severe and unremitting form of schizophrenia. If this finding can be confirmed in other pedigrees, it would provide a valuable example of the subtyping strategy.

A related problem is that, while a particular gene or genes may significantly increase one’s risk for developing schizophrenia or bipolar disorder, it will usually not be sufficient to produce the disorder; that is, most individuals who carry a susceptibility gene will not themselves become ill. (That is, there is ‘incomplete penetrance’ of the gene’s effects.) One strategy for dealing with this latter problem is to identify more sensitive, subclinical, phenotypes that indicate the presence of the gene even in people who do not develop the illness. For example, studies of schizophrenics’ families suggest that most schizophrenics carry a gene which leads to schizophrenia only 5–10 percent of the time, but causes abnormal smooth pursuit eye movements over 70 percent of the time. These eye movement dysfunctions should thus provide a better target for genetic linkage studies than schizophrenia itself (Holzman and Matthysse 1990).

A further complication is the likelihood of genotype–environment interactions. For example, there is evidence from dozens of studies that pre-and perinatal complications are significant risk factors for schizophrenia (e.g., Torrey et al. 1987). Such complications are, of course, hardly unique to schizophrenia; this suggests that pre-or perinatal insults to the developing brain may interact with genetic liability factors to produce schizophrenia. Kinney et al. (1999), for example, found that schizophrenics were much more likely than were either control subjects or the schizophrenics’ own non-schizophrenic siblings to have both a major perinated complication and eye tracking dysfunction. Moreover, these non-schizophrenic siblings tended to have either a history of perinatal complications or eye tracking dysfunction, but not both in the same sibling. This pattern of findings was consistent with a two-factor model in which perinatal brain injury and specific susceptibility genes often interact to produce schizophrenia.

3. Evidence For Genetic Factors In Bipolar Disorder

As in the case of schizophrenia, several converging lines of evidence strongly implicate genetic factors in the etiology of bipolar disorder.

3.1 Family, Twin, And Adoption Studies

There is a strong tendency for bipolar disorder to run in families, and the risk of bipolar disorder in a firstdegree relative of a manic-depressive is about 8 percent (vs. about 1 percent in the general population). Further evidence for a high heritability of bipolar disorder is provided by twin studies, particularly three twin studies conducted in Scandinavia in recent decades. The average concordance rate for bipolar disorder in these latter studies was 55 percent in MZ vs. only 5 percent in DZ twin pairs (see review by Vehmanen et al. 1995).

Complementary evidence for genetic factors in bipolar disorder is provided by adoption studies. For example, Mendlewicz and Ranier (1977) identified 29 adoptees with bipolar disorder, along with demographically matched controls who were either (a) psychiatrically normal or (b) had had polio during childhood. When the biological and adoptive parents of these adoptees were interviewed, significantly more cases of bipolar disorder, major depression, and schizoaffective disorder were found in the biological parents of bipolar adoptees than in the biological parents of either of the two control groups. The respective groups of adoptive parents, by contrast, did not differ significantly in the prevalence of these disorders.

3.2 Association And Linkage Studies

Although there have been dozens of reports of linkage between bipolar disorder and genetic loci on various chromosomes, few of these reports have subsequently been confirmed. Among these few are linkages to particular regions of chromosomes X, 18 and 21; each of these linkages has been confirmed by several independent groups. While this suggests that genes in these regions significantly influence susceptibility for bipolar disorder, the effects of these genes on susceptibility may be modest in size (for reviews, see Gershon et al. 1998 and Pekkarinen 1998).

A key difficulty in identifying genes in bipolar disorder is the likely presence of etiologic heterogeneity. One approach to overcoming this challenge is to search for markers of genetic subtypes of bipolar disorder. For example, MacKinnon et al. (1998), after identifying a strong tendency for a subtype of bipolar disorder with concomitant panic disorder to run in families, found strong evidence for linkage of the panic-disorder subtype to marker genes on a region of chromosome 18. There was no evidence of such linkage for bipolar disorder without panic disorder.

Other research has found that manic-depressive patients who respond well to lithium treatment represent a subtype that is etiologically more homogeneous, and has a stronger familial tendency, than non-lithium responders. Both linkage and association studies suggest that susceptibility to this lithium- responsive subtype is increased by certain alleles of the gene for phospholipidase C, an enzyme important in the phosphoinositol cycle that is thought to be a therapeutic target of lithium (Alda 1999).

4. Genetic Counseling

A recent study (Trippitelli et al. 1998) found that most patients with bipolar disorder and their unaffected spouses would be interested in receiving counseling about their own genetic risk, and that of their children. The majority of patients and spouses, for example, would take advantage of a test for susceptibility genes. Even when precise gene(s) involved in a particular case of schizophrenia or bipolar disorder are unknown, genetic counselors can provide a patient’s relatives with an estimated risk of developing the disorder, based on average figures from many different family studies. These risk figures, it should be noted, refer to a person’s lifetime risk—an important distinction (e.g., for schizophrenia, one’s risk has been cut roughly in half by age 30, and by age 50 it is extremely small). It is crucial that counseling be based on accurate differential diagnosis. Problems often encountered in counseling, such as counselees being confused by genetic information, or feeling fearful and embarrassed, may be heightened in families with bipolar disorder and schizophrenia, because these psychiatric disorders often carry a social stigma, and because many parents have (unfairly) been blamed for their child’s illness. For these reasons, Kessler (1980) emphasized the importance of (a) careful follow-up, to make sure that counseling was understood, and (b) the inclusion of professionals with good psychotherapeutic skills as part of the counseling team.

5. Creativity And Liability For Schizophrenia And Bipolar Disorder

There is increasing evidence, from converging lines of research, for the idea that genetic liability for schizophrenia and bipolar disorder is associated with unusual creative potential. This idea, which has long been the subject for theoretical speculation, has received empirical support from several complementary types of studies. For example, a number of studies involving non-clinical samples have reported that more creative subjects tend to score higher on personality test variables that are associated with liability for schizophrenia or bipolar disorder.

5.1 Creativity In Schizophrenics’ Biological Relatives

Of even greater interest are studies that have found unusual creativity in samples of the healthier bio- logical relatives of diagnosed schizophrenics. In an Icelandic sample, for example, Karlsson (1970) found that the biological relatives of schizophrenics were significantly more likely than people in the general population to be recognized in Who’s Who for their work in creative professions. In the adoption study noted earlier, Heston (1966) serendipitously discovered that, among the psychologically healthier ‘index’ adoptees (i.e., who had a biological mother with schizophrenia), there was a subgroup of psychologically healthy individuals who had more creative jobs and hobbies than the control adoptees.

Using a similar research design, Kinney et al. (2000) studied the adopted-away offspring of schizophrenic and control biological parents. The creativity of these adoptees’ actual vocational and avocational activities was rated by investigators who were blind to the adoptees’ personal and family histories of psychopathology. Real-life creativity was rated as significantly higher, on average, for subjects who, while not schizophrenic, did have signs of magical thinking, recurrent illusions or odd speech.

5.2 Creativity In Patients With Bipolar Disorder And Their Relatives

There is also evidence for increased creativity among subjects with major mood disorders and their biological relatives. Studies of eminent writers, artists, and composers carried out in the USA, the UK, and France, all found significantly higher rates of major mood disorders among these creators than among the general population (e.g., see Jamison 1990). Richards et al. (1988) extended this link by showing that measures of ‘everyday,’ or non-eminent, creativity were significantly higher, on average, in manic-depressive and cyclothymic subjects and their normal relatives than in control subjects who did not have a personal or family history of mood disorders.

Moreover, both creative artists and patients with mood disorders report that their creativity is significantly enhanced during periods of moderately elevated mood (e.g., Richards and Kinney 1990). These complementary findings suggest that the association between increased creative potential and genetic liability for bipolar disorder may extend not only to the millions of people with bipolar disorder, but also to tens of millions of others who, while not ill themselves, may carry genes for the disorder.

5.3 Implications Of Link Between Creativity And Genes For Schizophrenia And Bipolar Disorder

It is important to determine what maintains the high prevalence of genes for bipolar disorder in the population, despite the high rates of illness and death that are associated with this disorder. One interesting possibility is that genes which increase liability for bipolar disorder may also be associated with personally and socially beneficial effects, such as increased drive and creativity. The research findings suggesting an association between genes for bipolar disorder and increased creativity are also potentially of great significance in terms of how patients and their families view greater liability for bipolar disorder, as well as for combatting the social stigma that is still often attached to the disorder.

Parallel considerations apply in the case of schizophrenia, but perhaps with even greater force, because schizophrenia tends to be an even more chronic and disabling disease, have even lower fertility, and carry an even greater social stigma. As rapid advances in molecular biology and discovery of genetic markers make it possible to detect major genes for schizophrenia (and to identify individuals who carry these genes), it will become increasingly important to know whether such genes are associated with positive, as well as negative, behavioral phenotypes or outcomes—and to understand what genetic and/or environmental modifiers affect how these genes are expressed.

Bibliography:

1. Alda A 1999 Pharmacogenetics of lithium response in bipolar disorder. J. Psychiatry Neurosci. 24(2): 154–8
2. American Psychiatric Association 1994 Diagnostic and Statistical Manual of Mental Disorders (DSM-IV), 4th edn. American Psychiatric Association, Washington, DC
3. Gershon E S, Badner J A, Goldin L R, Sanders A R, Cravchik A, Detera-Wadleigh S D 1998 Closing in on genes for manicdepressive illness and schizophrenia. Neuropsychopharmacology 18: 233–42
4. Gottesman I I, McGuffin P, Farmer A E 1987 Clinical genetics as clues to the ‘real’ genetics of schizophrenia. Schizophrenia Bulletin 13: 23–48
5. Heston L L 1966 Psychiatric disorders in foster home reared children of schizophrenic mothers. British Journal of Psychiatry 112: 819–25
6. Holzman P S, Matthysse S 1990 The genetics of schizophrenia: A review. Psychological Science 1(5): 279–86
7. Jamison K R 1990 Manic-depressive illness and accomplishment: Creativity, leadership, and social class. In: Goodwin F, Jamison K R (eds.) Manic-depressive Illness. Oxford University Press, Oxford, UK
8. Karlsson J L 1970 Genetic association of giftedness and creativity with schizophrenia. Hereditas 66: 177–81
9. Kendler K S, Gruenberg A M, Kinney D K 1994 Independent diagnoses of adoptees and relatives, using DSM-III criteria, in the provincial and national samples of the Danish adoption study of schizophrenia. Archives of General Psychiatry 51: 456–68
10. Kessler S 1980 The genetics of schizophrenia: a review. Schizophrenia Bulletin 6: 404–16
11. Kety S S, Wender P H, Jacobsen B, Ingraham L J, Jansson L, Faber B, Kinney D K 1994 Mental illness in the biological and adoptive relatives of schizophrenic adoptees. Replication of the Copenhagen study in the rest of Denmark. Archives of General Psychiatry 51: 442–55
12. Kinney D K, Richards R L, Lowing P A, LeBlanc D, Zimbalist M A, Harlan P 2000 Creativity in offspring of schizophrenics and controls: An adoption study. Journal of Creativity Research 13(1): 17–25
13. Matthysse S, Kidd K K 1976 Estimating the genetic contribution to schizophrenia. American Journal of Psychiatry 133: 185–91
14. Maziade M, Martinez M, Cliche D, Fournier J P, Garneau Y, Merette C 1994 Linkage on the 11q21–22 region in a severe form of schizophrenia. American Psychiatric Association Annual Meeting, New Research Programs and Abstracts, p. 97
15. McGuffin P 1989 Genetic markers: an overview and future perspectives. In: Smeraldi E, Belloni L (eds.) A Genetic Perspecti e for Schizophrenic and Related Disorders. EdiErmes, Milan
16. Mendlewicz J, Ranier J D 1977 Adoption study supporting genetic transmission of manic-depressive illness. Journal of the American Medical Association 222: 1624–7
17. Murray R M, Takei N, Sham P, O’Callaghan E, Wright P 1993 Prenatal influenza, genetic susceptibility and schizophrenia. Schizophrenia Research 9(2,3): 137
18. Pekkarinen P 1998 Genetics of bipolar disorder. Psychiatria Fennica 29: 89–109
19. Richards R L, Kinney D K 1990 Mood swings and everyday creativity. Creativity Research Journal 3: 202–17
20. Torrey E F 1987 Prevalence studies in schizophrenia. British Journal of Psychiatry 150: 598–608
21. Torrey E F, Bowler A E, Taylor E H, Gottesman, I I 1994 Schizophrenia and Manic-depressi e Disorder. Basic Books, New York
22. Trippitelli C L, Jamison K R, Folstein M R, Bartko J J, DePaulo J P 1998 Pilot study on patients’ and spouses’ attitudes toward potential genetic testing for bipolar disorder. American Journal of Psychiatry 155(7): 899–904
23. Vehmanen L, Katrio J, Lonnqvist J 1995 Twin studies on concordance for bipolar disorder. Clinical Psychiatry and Psychopathology 26: 107–16
24. Wyatt R J 1991 Neuroleptics and the natural course of schizophrenia. Schizophrenia Bulletin 17: 325–51
25. Mackinnon D F, Xu J, McMahon F J, Simpson S G, Stine O C, McInnis M G, De Paulo J R 1998 Bipolar disorder and panic disorder in families: an analysis of chromosome 18 data. American Journal of Psychiatry 155(6): 829–31
26. Kinney D K, Yurgelun-Todd D A, Tramer S J, Holzman P S 1998 Inverse relationship of perinatal complications with eyetracking dysfunction in relatives of patients with schizophrenia: evidence for a two-factor model. American Journal of Psychiatry 155(7): 976–8
27. Richards R, Kinney D K, Lunde I, Benet M, Merzel A P C 1988 Creativity in manic-depressives, cyclothymes, their normal relatives and control subjects. Journal of Abnormal Psychology. 97(3): 1–8