Neurocognitive functioning and schizophrenia spectrum disorders can be independent expressions of familial liability for schizophrenia in community control children: the UCLA family study

Schizophrenia Research 54 (2002) 111 – 120 www.elsevier.com/locate/schres

Neurocognitive functioning and schizophrenia spectrum disorders can be independent expressions of familial liability for schizophrenia in community control children: the UCLA family study Robert F. Asarnow a,b,c,*, Keith H. Nuechterlein a,c, Joy Asamen a,d, David Fogelson a, Kenneth L. Subotnik a, Kenneth Zaucha a,d, Donald Guthrie a,b,e a

UCLA Neuropsychiatric Institute and Department of Psychiatry, Los Angeles, CA, USA b UCLA Mental Retardation Research Center, Los Angeles, CA, USA c UCLA Department of Psychology, Los Angeles, CA, USA d Department of Psychology, Pepperdine University, Malibu, CA, USA e UCLA Department of Biostatistics, Los Angeles, CA, USA Received 1 October 2001; accepted 4 October 2001

Abstract This study provided a further test of the hypothesis that certain neuromotor, language and verbal memory dysfunctions reflect genetic predisposition to schizophrenia, by examining the effects of family loading for schizophrenia (FLS) in normal controls without personal histories of schizophrenia or attention deficit hyperactivity disorder. In a case control design, 11 community controls (CC) with FLS were compared to 47 CC without FLS on tests of expressive and receptive language, visual motor coordination, full scale intelligence and verbal memory. In this study, FLS primarily reflects the incidence of schizophrenia spectrum diagnoses in the second-degree relatives of CC probands. CC probands with FLS had significantly poorer general intelligence, expressive and receptive vocabulary abilities, visual motor coordination and slower motor speed than CC probands without FLS. The variance in neurocognitive functioning associated with FLS is not due to the presence of any psychiatric disorders in CC probands, nor the presence of schizophrenia spectrum disorders in their parents. The relation between FLS and neurocognitive and neuromotor functioning in CC probands was moderated by the parent’s cognitive functioning. The results of the present study indicate that familial liability to schizophrenia can be transmitted across two generations, independent of the presence of schizophrenia spectrum disorders in either the parent or proband, and account for significant variance in proband neurocognitive and neuromotor functioning. These findings suggest the neurocognitive and neuromotor functioning and schizophrenia spectrum disorders can be relatively independent expressions of familial liability to schizophrenia. D 2002 Elsevier Science B.V. All rights reserved. Keywords: Schizophrenia; Liability to schizophrenia; Schizotypy; Language; Neuromotor; Neurocognitive; Verbal memory; Genetic predisposition to schizophrenia; Schizophrenia spectrum disorders

*

Corresponding author. UCLA Neuropsychiatric Institute and Department of Psychiatry and Biobehavioral Science, NPI 48-240C, 760 Westwood Plaza, Los Angeles, CA, 90024-1759, USA. E-mail address: [email protected] (R.F. Asarnow). 0920-9964/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 0 - 9 9 6 4 ( 0 1 ) 0 0 3 5 8 - 9

112

R.F. Asarnow et al. / Schizophrenia Research 54 (2002) 111–120

1. Introduction Meehl’s (1962) seminal framework, for explicating the interplay of genetic predisposition and environmental factors in the etiology of schizophrenia, conceptualized schizotypy as a personality organization that reflected the imposition of social learning history upon individuals with a genetic liability for schizophrenia. Genetic liability for schizophrenia was hypothesized to produce a ‘‘neural integrative defect’’. This framework focused interest on elucidating the behavioral manifestations of the neurointegrative deficit. Led by Peter Venables, his students and collaborators, studies were conducted to identify the psychobiological processes that reflected the neural integrative deficit. One research strategy used to identify the manifestations of genetic liability to schizophrenia are studies of the first-degree relatives of patients with schizophrenia. Subtle neuromotor and neurocognitive abnormalities, including language and short-term memory, are thought to be behavioral manifestations of genetic liability to schizophrenia because they aggregate in the non-psychotic first-degree relatives of patients with schizophrenia. (See reviews by Asarnow, 1988; Erlenmeyer-Kimling et al., 2000; Kremen et al., 1994; Moldin and Erlenmeyer-Kimling, 1994). For example, subtle neuromotor (Erlenmeyer-Kimling, 2000; Fish, 1984; Hans et al., 1999; Marcus et al., 1987; McNeil et al., 1993; Walker et al., 1994) verbal and non-verbal memory (Cannon et al., 1994; Faraone et al., 1995) and language impairments (Faraone et al., 1995; Jones et al., 1994) are present in non-psychotic children of patients with schizophrenia. A related research strategy is comparing schizophrenia probands with and without family loading for schizophrenia (FLS). Impairments in schizophrenia probands with FLS compared to those without FLS are presumed to reflect the genetic liability to schizophrenia. Methodological problems result in ‘‘few robust differences between patients with familial and sporadic schizophrenia’’ (Roy and Crowe, 1994). The current study extended the traditional family loading design to provide a further test of the hypothesis that certain neuromotor, language, and short-term memory dysfunctions reflect genetic predisposition to schizophrenia by examining the effect of FLS in normal controls, who are free of personal histories of schiz-

ophrenia spectrum disorders. If a neurocognitive task taps familial liability to schizophrenia, normal controls with FLS should perform more poorly than normal controls without FLS. Normal controls with FLS are free of the confounding effects of medications and life experiences associated with the diagnosis of schizophrenia that may obscure the effects of FLS in patients with schizophrenia. The present study differed from prior studies of the non-psychotic relatives of patients with schizophrenia in three ways. First, prior studies examined subjects with a first-degree relative with schizophrenia (e.g., children of patients with schizophrenia). In contrast, in the current study, FLS subjects had either first- or second-degree relatives with schizophrenia spectrum disorder diagnoses. Second, prior studies demonstrate that neurocognitive and neuromotor impairments are present in the absence of schizophrenia in one generation (e.g., non-psychotic children of patients with schizophrenia). The present study provided a further test of the hypothesis that schizophrenic disorders and neurocognitive and neuromotor impairments can be independent manifestations of liability to schizophrenia by examining the relation between neurocognitive and neuromotor functioning and schizophrenia spectrum disorders (a wider and more subtle range of disorders than examined in most prior studies) over multiple generations (child proband, parent, and grandparents/aunts and uncles). Finally, the present study determined if there was familial transmission across generations of neurocognitive and neuromotor functioning in children with FLS. With very few exceptions (primarily genetic linkage studies, e.g., Freedman et al., 1997; Holzman et al., 1988), prior studies have evaluated neurocognitive functioning in only one generation. We examined the relation between neurocognitive functioning in children and their parents to determine if familial transmission of neurocognitive and neuromotor functioning was moderated by FLS.

2. Methods 2.1. Ascertainment of probands and diagnosis of probands’ relatives The UCLA Family Study consists of parallel studies of relatives of three child (Childhood Onset

R.F. Asarnow et al. / Schizophrenia Research 54 (2002) 111–120

Schizophrenia, Community Controls, and Attention Deficit Hyperactivity Disorder) and three adult proband groups. This paper presents data on the child community control (CC) subjects from the UCLA Family Study. Lists of potential CC probands living in the same zip codes as childhood onset schizophrenia probands were obtained from a scientific survey research film (Survey Sampling). Parents of potential CCs were contacted by telephone to explain the purpose of the study and screen out probands with schizophrenia and ADHD or who were placed in special classes for either exceptionally bright or learning disabled children. Children with learning disabilities were excluded because learning disabilities can result in poor performance on some of the neurocognitive tasks used in this study. Exceptionally gifted children were excluded to make the CC probands as comparable as possible to the Attention Deficit Hyperactivity and Schizophrenia probands. Parents of all participants in the study provided a written informed consent and child probands assented to participate in the study. An experienced child psychologist used the KSADS (Orvaschel and Puig-Antich, 1987) to interview the parent about the CC proband’s psychiatric symptoms and then conducted a psychiatric interview of the proband. The psychologist then reviewed medical and school records and K-SADS interviews with an experienced child psychiatrist to reach a consensus lifetime diagnosis (Russell et al., 1989). Children were excluded from the community control group for a personal diagnosis of either schizophrenia or ADHD and conduct disorder, but not for any other psychiatric diagnoses. More than 31% of the 58 community control probands received at least one lifetime DSM-III-R diagnosis. The most frequent diagnoses were: separation anxiety disorder (5), adjustment disorder (4), oppositional defiant disorder (3), overanxious disorder (3), and simple phobia (3). Children were excluded if they were less than 9 years old, had a full scale I.Q. less than 70, or had a CNS disease or had taken psychoactive drugs that could produce psychotic episodes. Parents of CC probands were given structured diagnostic interviews, the Diagnostic Interview Schedule (Robins et al., 1981; Regier et al., 1984) supplemented with the psychosis section of the Present State Examination (Wing et al., 1974) for

113

Axis I disorders, and the SCID-II (Spitzer and Williams, 1986) for Axis II disorders, by a set of interviewers that was entirely separate (housed in separate buildings) from the proband diagnostic team. Interviewers of parents were blind to the proband’s diagnosis. Family history information was collected on all first- and second-degree relatives, usually from two informants, using the Relative Family History Interview (Gershon, 1985). Family history diagnoses of schizotypal or paranoid personality disorders have relatively low sensitivity but adequate specificity (Li, et al., 1977; Asarnow et al., 2001; Fogelson et al., 2001). Direct interviews, family history, and information from medical records were reviewed to make consensus lifetime DSM-III R diagnoses for parents. Consensus lifetime DSM-III R diagnoses for seconddegree relatives were based on family history and, when available, medical records. For all first- and second-degree relatives, the consensus diagnoses were reviewed by senior clinicians (DLF, KLS, KHN, and RFA), who were always blind to the diagnoses of probands and other family members. Our prior reports (Asarnow et al., 2001; Fogelson et al., 1991) describe the background and training of family interviewers, inter-rater reliability on key diagnoses, and the diagnostic process in more detail. CC probands were considered to have a FLS if any first- or second-degree relative received an FH-RDC or DSM-III R diagnosis of schizophrenia, schizoaffective disorder-mainly depressed, or DSM-III R diagnoses of paranoid or schizotypal personality disorders. 2.2. Neurocognitive tasks Three domains of neurocognitive functioning were examined in CC probands: short-term verbal memory, language, and neuromotor functioning. All but one of the neurocognitive tests, the California Verbal Learning Test for Children (CVLT-C; Delis et al., 1986), yielded age-corrected standard scores as dependent variables. Short-term verbal memory was assessed by the CVLT-C. A list of 15 words (five from each of three semantic categories) is read to the participants over five trials. In the present study, the dependent variable was the number of words recalled on the first trial. Receptive language functioning was assessed by the Peabody Picture Vocabulary Test-Revised (PPVT-

114

R.F. Asarnow et al. / Schizophrenia Research 54 (2002) 111–120

R; Dunn and Dunn, 1981). The PPVT-R requires subjects to point to one of the four pictures corresponding to a word read to the subject by the examiner. Expressive language was assessed by the vocabulary subtest of the Wechsler Intelligence Scale for Children-Revised (WISC-R; Wechsler, 1974). The WISC-R vocabulary subtest requires subjects to define words that are read to them by the examiner. Neuromotor functioning was assessed by the Visual-Motor Coordination subtests of the WISC-R (block design, object assembly, coding, and mazes). These subtests make extensive demands on motor speed and visual-motor and visual-spatial integration (Lutey, 1977). In addition, Full Scale I.Q. of the WISC-R was used as an index of general intellectual functioning. Research assistants who were blind to the diagnosis of the relatives administered the CVLT, WISC-R AND PPVT-R to CC probands. In addition to the measures administered to CC probands, the Wechsler Adult Intelligence ScaleRevised (WAIS-R; Wechsler, 1981) vocabulary subtest was administered to parents of CC probands by research assistants who were blind to proband diagnoses. It was be used as an index of parental cognitive functioning.

3. Results 3.1. Subject characteristics There were 11 CC probands with FLS and 47 CC probands without FLS. The CC probands with FLS did not significantly differ in age (M = 11.7, S.D. = 1.9) from the CC probands without FLS (M = 12.3, S.D. = 2.5). Nonetheless, for the WISC-R and PPVT, age-adjusted standard scores were used in the analyses. The groups did not differ in the proportion of probands who were females (CC with FLS = 63.6%, CC without FLS = 46.8%). In the present study, FLS primarily reflects the incidence of schizophrenia spectrum diagnoses in second-degree relatives. Only 4.5% of the parents of CC probands with FLS have a schizophrenia spectrum disorder (one parent with a diagnosis of paranoid personality disorder). In contrast, more than 15% (Table 1) of the second-degree relatives of CC pro-

Table 1 Incidence of schizophrenia spectrum diagnoses in relatives of Community Control probands with family loading for schizophrenia

Schizophrenia Schizotypal personality disorder Paranoid personality disorder

Parents (n = 22) (%)

Second degree relatives (n = 57) (%)

0 0

10.5 1.75

4.5

3.5

bands with FLS had schizophrenia spectrum diagnoses. 3.2. Comparison of CC probands with and without FLS As noted above, 31% of CC probands received personal DSM-III R diagnoses. Consequently, we used mixed model analyses of variance to evaluate the effect of FLS, a personal DSM-III R diagnosis, and the interaction of FLS and personal DSM-III R diagnoses on each of the five neurocognitive measures. There was a significant main effect of FLS on four of the five measures (Table 2). CC probands with FLS performed significantly more poorly (Figs. 1 and 2) than CC probands without FLS on WISC-R full scale I.Q., WISC-R Vocabulary and Visual-Motor Coordination subtests, and PPVT-R. The effect sizes (Cohen’s d) for these differences were large (0.74 –1.19). The only test that did not differentiate between CC with and without FLS was CVLT trial 1 (effect size = 0.12). The main effects of personal DSM-III R diagnosis on neurocognitive functioning were very small (effect sizes < 0.10) and not statistically significant for any measure (Table 2). The FLS by personal DSM-III-R diagnoses interactions were also quite small, and not statistically significant. 3.3. Parental cognitive functioning Parents of CC probands with FLS obtained significantly lower (Table 2; effect size = 0.84) scaled scores on the WAIS-R vocabulary subtest than parents of CC probands without FLS (Fig. 2). There were statistically significant correlations between the mean parental WAIS-R vocabulary score

R.F. Asarnow et al. / Schizophrenia Research 54 (2002) 111–120

115

Table 2 Effects of family loading for schizophrenia on neurocognitive functioning in Community Control probands with and without FLS Proband

CCs with FLS (N = 11) CCs without FLS (N = 47)

Effect size for contrast between CCs with and without family loading FLS X Personal DSM III-R diagnosis ANOVA Effect of FLS Effect of personal DSM-III-R diagnosis FLS X Parental Vocabulary ANOVA Effect of FLS Effect of parental WAIS-R vocabulary

Parent

WISC-R F.S.I.Q.

WISC-R vocabulary

WISC-R visual/motor scale

PPVT-R

CVLT Trial 1

WAIS-R vocabulary

M = 103.73 S.D. = 13.18 M = 114.87 S.D. = 10.85

M = 10.18 S.D. = 1.94 M = 12.0 S.D. = 1.94

M = 10.57 S.D. = 2.25 M = 12.03 S.D. = 1.91

M = 97.3 S.D. = 13.92 M = 112.72 S.D. = 12.78

M = 8.14 S.D. = 1.07 M = 8.35 S.D. = 1.92

M = 11 S.D. = 1.76 M = 12.56 S.D. = 1.88

0.98

0.93

0.74

0.12

0.84

F1,54 = 6.74 p = 0.01 F1,54 = 0.14 p = 0.70

F1,54 = 6.02 p = 0.03 F1,54 = 0.24 p = 0.62

F1,54 = 4.47 p = 0.03 F1,54 = 0.21 p = 0.62

F1,52 = 7.15 p = 0.01 F1,52 = 0.66 p = 0.42

F1,36 = 0.02 p = 0.87 F1,36 = 0.03 p = 0.59

F1,54 = 7.10 p = 0.01

F1,54 = 0.39 p < 0.53 F1,54 = 18.25 p < 0.0001

F1,54 = 1.18 p < 0.28 F1,54 = 9.82 p < 0.003

F1,54 = 0.17 p < 0.68 F1,54 = 10.30 p < 0.002

F1,52 = 2.75 p < 0.11 F1,54 = 11.74 p < 0.001

F1,36 = 0.01 p < 0.94 F1,54 = 0.03 p < 0.86

and the scores of the CC children with FLS on all four tasks that differentiated CC children with and without FLS (Table 3). The correlations between

1.19

mean parental vocabulary score and these four proband neurocognitive and neuromotor scores were somewhat higher in CC probands with FLS than in

Fig. 1. Performance on WISC-R and WAIS-R subtests of CC probands and parents with and without FLS.

116

R.F. Asarnow et al. / Schizophrenia Research 54 (2002) 111–120

Fig. 2. Full Scale I.Q. and PPVT-R performance of CC probands with and without FLS.

CC probands without FLS (Table 3). This pattern of results suggests that FLS may moderate the relationship between proband and parental neurocognitive functioning. To evaluate formally whether the effect of FLS on proband neurocognitive functioning is moderated by the parental cognitive functioning (as measured by the parent WAIS-R vocabulary score), we conducted mixed model analyses of variance to determine the effects of FLS, parental vocabulary score, and the interaction between FLS and parental vocabulary score on each of the neurocognitive measures. If parental vocabulary moderates the effect of FLS and proband neurocognitive functioning, adding parental vocabulary to the ANOVA should result in a greatly diminished effect of FLS on proband neurocognitive functioning. There were significant main effects of parental cognitive functioning on all four measures that differentiated CC probands with and without FLS (Table 2). When parental vocabulary scores were added to the model, the main effects of FLS were no longer significant on these four measures (bottom of Table 2). The FLS X parental vocabulary score interaction was not significant for any of these measures. The correlations between mean parent vocabulary score and the four neurocognitive tasks that differentiate CC probands with and without FLS ranged

between 0.66 and 0.85 (Table 3). This suggests that the four tasks may tap a single, common factor. To determine the extent to which the proband’s scores on the four tasks that differentiated CC probands with and without FLS are tapping the same underlying processes, we carried out a principal components analysis. The first principal factor yielded an eigenvalue of 2.49, with factor loadings range from 0.68 (vocabulary scaled score and PPVT) to 0.94 (full scale I.Q.). The large and highly significant correlation (r = 0.84; Table 3) between parental vocabulary, and this factor score in CC probands with FLS suggests that the common variance shared by the four tasks is tied to parental cognitive functioning.

Table 3 Correlation between mean parent vocabulary scores and neurocognitive functioning in CC with and without FLS Proband Neurocognitive Score

Mean Parent CC with FLS

Vocabulary Score CC Without FLS

WISC-R Full Scale I.Q. WISC-R Vocabulary WISC-R Visual/Motor PPVT-R CVLT Trial 1 Factor 1

0.85** 0.66 * 0.66 * 0.66 * 0.12 0.84**

0.33 0.36 0.31 0.43 0.01 0.47

* p < 0.05. ** p < 0.001.

R.F. Asarnow et al. / Schizophrenia Research 54 (2002) 111–120

4. Conclusions In CC probands, a substantial portion of the variance on neurocognitive tasks demonstrated in prior studies to be sensitive to liability to schizophrenia is accounted for by FLS. While both CC probands with and without FLS obtained scores in the average range, CC probands with FLS had significantly lower general intelligence, expressive and receptive vocabulary abilities, and visual motor coordination and slower motor speed than the CC probands without FLS. The effect sizes for these differences were all quite large. These results are consistent with prior studies of first-degree relatives of patients with schizophrenia. Deficient neuromotor integration (Fish, 1984; Hans et al., 1999; McNeil et al., 1993; Walker et al., 1994) expressive and receptive language (Faraone et al., 1995; Jones et al., 1994) and decreased general intellectual functioning (Faraone et al., 1995) are found in the nonpsychotic biological relatives of patients with schizophrenia. Prior studies (Cannon et al., 1994; Seidman et al., 1998) have detected verbal declarative memory impairments in relatives of patients with schizophrenia, while this study did not find verbal memory (CVLT Trial 1) performance to be associated with FLS in CC probands. Some recent evidence (Faraone et al., 2000) suggests that the degree of FLS is related to the amount of verbal declarative memory impairment in relatives of patients with schizophrenia. Perhaps in the current study, the degree of FLS in CC probands with FLS was less than in the prior studies in which probands had one or more first-degree relatives with diagnoses of schizophrenia. Alternatively, declarative memory problems in patients with schizophrenia may reflect the interaction of genetic liability to schizophrenia with reduced volume of mesial temporal/hippocampal structures secondary to obstetric complications that cause hypoxic insults (Cannon et al., submitted for publication). Very few of the CC probands suffered this form of obstetric complication. This study extends prior findings that neuromotor and language impairments and reductions in general intellectual functioning are present in the non-psychotic first-degree relatives of patients with schizophrenia. In the current study, CC probands with FLS are free of both schizophrenia spectrum disorders and ADHD spectrum disorders, including conduct disor-

117

der and learning disabilities. In addition, none of the variance on the neurocognitive tasks that differentiates CC probands with and without FLS can be accounted for by any DSM-III R diagnoses in CC probands. Thus, the effect of FLS on neurocognitive functioning cannot be attributed to the presence of psychiatric disorders in the proband. The variance in neurocognitive functioning associated with FLS is also not due to the presence of schizophrenia spectrum disorders in the parents. The parents of CC probands with FLS are, with one exception, free of schizophrenia spectrum disorders. Indeed, the evidence of familial liability for schizophrenia in this group comes from second-degree relatives. The relationship between FLS and neurocognitive and neuromotor functioning of CC probands was moderated by the parent’s cognitive functioning (see Fig. 3). Parental cognitive functioning, as indexed by WAIS-R vocabulary, was not a result of the presence of schizophrenia spectrum disorders in the parents, since only one parent has a diagnosis of schizophrenia spectrum disorder. The results of the present study are consistent with the view that familial liability to schizophrenia can be transmitted across two generations, independent of the presence of schizophrenia spectrum disorders in either the parent or proband, and account for significant variance in proband neurocognitive and neuromotor functioning. These results provide a dramatic demonstration that, at least in some situations, familial liability to schizophrenia can account for the substantial portions of the variance in cognitive functioning in normal control subjects who are free of personal schizophrenia or ADHD diagnoses. Factor analyses of data from relatives of schizophrenia patients provide converging evidence that positive and negative schizotypy dimensions are relatively independent of neurocognitive functioning in the same relatives, although neurocognitive deficits were associated with odd and eccentric behavior (Nuechterlein et al., 2000). The association between neurocognitive functioning and FLS in CC probands without schizophrenia spectrum diagnoses in the present study does not mean that there is never a relationship between these alternative manifestations of liability to schizophrenia. Rather, it does demonstrate that neurocognitive functioning is not isomorphic with schizophrenia spectrum disorders, and

118

R.F. Asarnow et al. / Schizophrenia Research 54 (2002) 111–120

Fig. 3. Summary of relations between neurocognitive functioning and schizophrenia spectrum diagnoses over three generations.

suggests that neurocognitive functioning and schizophrenia spectrum disorders can be relatively separate manifestations of liability to schizophrenia. These results are reminiscent of the relation between smooth pursuit eye movement abnormalities and schizophrenia spectrum disorders that led Holzman et al. (1988) to posit a latent trait for schizophrenia. Performance on the WAIS-R vocabulary subtest was intended to provide an estimate of the parent’s general level of intellectual functioning in this study, since vocabulary tests have very high correlations with full scale I.Q. The WAIS-R Vocabulary test, however, is a measure of expressive vocabulary. As noted above, both patients with schizophrenia and their first-degree relatives have impaired language functioning. As a consequence, it is not clear whether the relationship between proband and parental neurocognitive functioning reflects the effect of FLS on the general level of intellectual functioning or language skills of parents of CC probands. The results of the principal components analysis suggest the transmission of general cognitive abilities across generations. To rigorously determine whether FLS is associated with the transmission of general cognitive abilities or specific cognitive abilities superimposed on, general cognitive abilities require a much more extensive examination of parallel cognitive measures in probands and their relatives than was undertaken in the current study.

This study builds upon the pioneering efforts of Peter Venables and his collaborators to refine the characterization of the psychobiological processes that reflect liability to schizophrenia. Of course, the results of this study need to be replicated with larger samples. The findings justify an effort at the replication given their implications for understanding the genetic liability to schizophrenia. The results of this study raise two competing hypotheses about the relationship between certain neurocognitive processes and schizophrenia spectrum disorders. There may be different susceptibility genes for schizophrenia spectrum disorders and certain neurocognitive impairments. Alternatively, the association between schizophrenia spectrum disorders in second-degree relatives and neurocognitive functioning in CC probands may reflect a common set of susceptibility genes. These hypotheses can be addressed in the future, more refined genetic analyses that include information on putative susceptibility genes for schizophrenia.

Acknowledgements This research was supported by research grants from NIMH (MH45112, MH49716, MH46981, MH30911 and MH14584) and by the Della Martin Foundation.

R.F. Asarnow et al. / Schizophrenia Research 54 (2002) 111–120

We acknowledge the invaluable assistance of Carol, Giannini, Tampsen Thorpe, Heidi Kuppinger, Richard Torquato, Diana Payne, Nancy Elliot, Lisa Thomrongsith and Kim Hughes for the assistance in data collection and Christie Ong and Beth Tang for assisting in the manuscript preparation.

References Asarnow, J.R., 1988. Children at risk for schizophrenia: converging lines of evidence. Schizophr. Bull. 14, 613 – 631. Asarnow, R.F., Nuechertein, K.H., Fogelson, D.L., Subotnik, K.L., Payne, D.A., Russell, A.T., Asamen, J., Kuppinger, H., Kendler, K.S., 2001. Schizophrenia and schizophrenia spectrum personality disorders in the first-degree relatives of children with schizophrenia: the UCLA family study. Arch. Gen. Psychiatry 58, 581 – 588. Cannon, T.D., Zorrilla, L.E., Shtasel, D., Gur, R.E., Gur, R.C., Marco, E.J., Moberg, P., Price, R.A., 1994. Neuropsychological functioning in siblings discordant for schizophrenia and healthy volunteers. Arch. Gen. Psychiatry 51, 651 – 661. Cannon, T.D., van Erp, T.G.M., Huttunen, M.J.L., Salonen, O., Valanne, L., Poutanen, V.P., Standertskjold-Nordenstam, C.G., 2000. Fetal hypoxia and regional brain volumes in schizophrenic patients, their siblings, and controls. Manuscript submitted for publication. Delis, D., Kramer, J., Kaplan, E., Ober, B., 1986. The California Verbal Learning Test: Children’s Version. The Psychological Corporation, San Antonio, TX. Dunn, L.M., Dunn, L.M., 1981. Peabody Picture Vocabulary TestRevised. American Guidance Service, Circle Pines, MN. Erlenmeyer-Kimling, L., Rock, D., Roberts, S.A., Janal, M., Kestenbaum, C., Cornblatt, B., Adamo, U.H., Gottesman, I.I., 2000. Attention, memory, and motor skills as childhood predictors of schizophrenia-related psychoses: the New York high-risk project. Am. J. Psychiatry 157, 1416 – 1422. Faraone, S.V., Seidman, L.J., Kremen, W.S., Pepple, J.R., Lyons, M.J., Tsuang, M.T., 1995. Neuropsychological functioning among the nonpsychotic relatives of schizophrenic patients: a diagnostic efficiency analysis. J. Abnorm. Psychol. 104, 286 – 304. Faraone, S.V., Seidman, L.J., Kremen, W.S., Toomey, R., Pepple, J.R., Tsuang, M.T., 2000. Neuropsychologic functioning among the nonpsychotic relatives of schizophrenic patients: the effect of genetic loading. Biol. Psychiatry 48, 120 – 126. Fish, B., 1984. Characteristics and sequelae of the neurointegrative disorder in infants at risk for schizophrenia: 1952 – 1982. In: Watt, N.F., Anthony, E.J., Wynne, L.C., Rolf, J. (Eds.), Children at Risk for Schizophrenia: A Longitudinal Perspective. Cambridge Univ. Press, New York, pp. 423 – 439. Fogelson, D.L., Nuechterlein, K.H., Asarnow, R.F., Subotnik, K.L., Talovic, S.A., 1991. Interrater reliability of the structured clinical interview for DSM-III-R, Axis II: schizophrenia spectrum and affective spectrum disorders. Psychiatry Res. 39, 55 – 63.

119

Fogelson, D.L., Nuechterlein, K., Asarnow, R., Payne, D., Subotnik, K., Kuppinger, H., 2001. Validity of the family method for diagnosing schizophrenia and schizophrenia related psychoses in first-degree relatives of schizophrenic probands. Paper presented at the International Congress of Schizophrenia Research, Whistler, British Columbia. Freedman, R., Coon, H., Myles-Worsley, M., Orr-Urtreger, A., Olincy, A., Davis, A., Polymeropoulos, M., Holik, J., Hopkins, J., Hoff, M., Rosenthal, J., Waldo, M.C., Reimherr, F., Wender, P., Yaw, J., Young, D.A., Breese, C.R., Adams, C., Patterson, D., Adler, L.E., Kruglyak, L., Leonard, S., Byerley, W., 1997. Linkage of a neurophysiological deficit in schizophrenia to a chromosome 15 locus. Proc. Natl. Acad. Sci. U. S. A. 94, 587 – 592. Gershon, E.S., 1985. Relative Psychiatric History (RPH) Symptom Checklist Interview. National Institute of Mental Health, Bethesda, MD. Hans, S.L., Marcus, J., Nuechterlein, K.H., Asarnow, R.F., Styr, B., Auerbach, J.G., 1999. Neurobehavioral deficits at adolescence in children at risk for schizophrenia: the Jerusalem infant development study. Arch. Gen. Psychiatry 56, 741 – 748. Holzman, P.S., Kringlen, E., Matthysse, S., Flanagan, S.D., Lipton, R.B., Cramer, G., Levin, S., Lange, K., Levy, D.L., 1988. A single dominant gene can account for eye tracking dysfunctions and schizophrenia in offspring of discordant twins [see comments] . Arch. Gen. Psychiatry 45, 641 – 647. Jones, P., Rodgers, B., Murray, R., Marmot, M., 1994. Child development risk factors for adult schizophrenia in the British 1946 birth cohort. Lancet 344, 1398 – 1402. Kremen, W.S., Seidman, L.J., Pepple, J.R., Lyons, M.J., Tsuang, M.T., Faraone, S.V., 1994. Neuropsychological risk indicators for schizophrenia: a review of family studies. Schizophr. Bull. 20, 103 – 119. Li, G., Aryan, M., Silverman, J.M., Haroutunian, V., Perl, D.P., Birstein, S., Lantz, M., Marin, D.B., Mohs, R.C., Davis, K.L., 1997. The validity of the family history method for identifying Alzheimer disease. Arch. Neurol. 54, 634 – 640. Lutey, C., 1977. Individual Intelligence Testing: A Manual and Sourcebook. Lutey Publishing, Greeley, Colo (2nd and enlarged edn.). Marcus, J., Hans, S.L., Nagler, S., Auerbach, J.G., Mirsky, A.F., Aubrey, A., 1987. Review of the NIMH Israeli Kibbutz—city study and the Jerusalem infant development study. Schizophr. Bull. 13, 425 – 438. McNeil, T.F., Harty, B., Blennow, G., Cantor-Graae, E., 1993. Neuromotor deviation in offspring of psychotic mothers: a selective developmental deficiency in two groups of children at heightened psychiatric risk? J. Psychiatr. Res. 27, 39 – 54. Meehl, P., 1962. Schizotaxia, schizotypy, schizophrenia. Am. J. Psychiatry 17, 827 – 838. Moldin, S.O., Erlenmeyer-Kimling, L., 1994. Measuring liability to schizophrenia: progress report 1994: editors’ introduction. Schizophr. Bull. 20, 25 – 29. Nuechterlein, K.H., Asarnow, J.R., Subotnik, K.L., Fogelson, D.L., Payne, D.L., Kendler, K.S., Neale, M.C., Jackson, K.C., Mintz, J., 2002. The structure of schizotypy: relationships between neurocognitive and personality disorder features in relatives of

120

R.F. Asarnow et al. / Schizophrenia Research 54 (2002) 111–120

schizophrenic patients in the UCLA Family Study. Schizophr. Res. 54, 121 – 130. Orvaschel, E., Puig-Antich, J., 1987. Schedule for Affective Disorder and Schizophrenia for School Aged Children—Epidemiologic Version: (K-SADS-E). Medical College of Pennsylvania, Philadelphia. Regier, D.A., Myers, J.K., Kramer, M., Robins, L.N., Blazer, D.G., Hough, R.L., Eaton, W.W., Locke, B.Z., 1984. The NIMH epidemiologic catchment area program. Historical context, major objectives, and study population characteristics. Arch. Gen. Psychiatry 41, 934 – 941. Robins, L.N., Helzer, J.E., Croughan, J., Ratcliff, K.S., 1981. National institute of mental health diagnostic interview schedule. Its history, characteristics, and validity. Arch. Gen. Psychiatry 38, 381 – 389. Roy, M.A., Crowe, R.R., 1994. Validity of the familial and sporadic subtypes of schizophrenia. Am. J. Psychiatry 151, 805 – 814. Russell, A.T., Bott, I., Sammons, C., 1989. The phenomenology of

schizophrenia occurring in childhood. J. Child Adolesc. Psychiatry 3, 339 – 407. Seidman, L.J., Stone, W.S., Jones, R., Harrison, R.H., Mirsky, A.F., 1998. Comparative effects of schizophrenia and temporal lobe epilepsy on memory. J. Int. Neuropsychol. Soc. 4, 342 – 352. Spitzer, R.L., Williams, J.B.W., 1986. Structured Clinical Interview for DSM-III-R Personality Disorders, in Biometrics Research. New York State Psychiatric Institute, New York. Walker, E.F., Savoie, T., Davis, D., 1994. Neuromotor precursors of schizophrenia. Schizophr. Bull. 20, 441 – 451. Wechsler, D., 1974. Wechsler Intelligence Scale for Children-Revised Manual. Psychological, New York, NY. Wechsler, D., 1981. WAIS-R Manual: Wechsler Adult Intelligence Scale-Revised. Psychological, New York, NY. Wing, J.K., Sartorius, N., Cooper, J.E., 1974. The Measurement and Classification of Psychiatric Symptoms: An Instruction Manual for the PSE and CATEGO Programs. Cambridge Univ. Press, London.