Directory UMM :Data Elmu:jurnal:B:Biological Psichatry:Vol48.Issue3.2000:
Neuropsychologic Functioning among the Nonpsychotic
Relatives of Schizophrenic Patients:
The Effect of Genetic Loading
Stephen V. Faraone, Larry J. Seidman, William S. Kremen, Rosemary Toomey,
John R. Pepple, and Ming T. Tsuang
Background: We previously reported that the nonpsychotic relatives of schizophrenic patients exhibited disturbances in executive functioning, verbal and visual memory, auditory attention, mental control, and verbal ability.
In a 4-year follow-up, we showed that the discriminating
power of most of these tests was stable over time.
Methods: In this report we compare 41 nonpsychotic
persons who have only one schizophrenic first-degree
relative (simplex families) with 36 nonpsychotic persons
who have two schizophrenic first-degree relatives (multiplex families). Our goal was to test a hypothesis that
neuropsychologic deficits would be worse among the
latter.
Results: Relatives from multiplex families differed significantly from controls on estimated intelligence, immediate
and delayed logical memories, and immediate visual
reproductions. In contrast, in comparisons with controls,
relatives from simplex families only differed on immediate
logical memories. Comparisons between relatives from
multiplex and simplex families showed that the former
group had significantly worse scores for estimated intelligence, immediate and delayed logical memories, and
immediate visual reproductions. We also found group 3
gender interactions: the worse performance of the multiplex group was seen for females.
Conclusions: These results are consistent with the idea
that neuropsychologic deficits in relatives of schizophrenic patients reflect their degree of genetic predisposition to schizophrenia. They also suggest hypotheses
about gender differences in the familial transmission of
From the Harvard Medical School Department of Psychiatry at the Massachusetts
Mental Health Center and Brockton/West Roxbury VA Medical Center,
Brockton (SVF, LJS, RT, JRP, MTT), Psychiatry Service, Massachusetts
General Hospital, Boston (SVF, LJS, MTT), Harvard Institute of Psychiatric
Epidemiology and Genetics, Boston (SVF, LJS, RT, JRP, MTT), and the
Department of Epidemiology, Harvard School of Public Health, Boston (MTT),
Massachusetts, and the Department of Psychiatry, University of California,
Davis School of Medicine, Sacramento and UC Davis–Napa Psychiatric
Research Center, Napa State Hospital, Napa, California (WSK).
Address reprint requests to Stephen V. Faraone, PhD, Harvard Medical School,
Suite 255, 750 Washington Street, South Easton MA 02375.
Received June 16, 1999; revised September 16, 1999; accepted September 21,
1999.
© 2000 Society of Biological Psychiatry
the disorder. Biol Psychiatry 2000;48:120 –126 © 2000
Society of Biological Psychiatry
Key Words: Genetics, neuropsychology, endophenotypes, multiplex families
Introduction
W
e previously reported that the nonpsychotic relatives
of schizophrenic patients exhibited disturbances in
executive functioning, verbal and visual memory, auditory
attention, mental control, and verbal ability (Faraone et al
1995b; Toomey et al 1998) and that most of these findings
were stable over a 4-year period (Faraone et al 1999a).
These findings were consistent with the findings in relatives of schizophrenic patients reviewed by Kremen et al
(1994) and with other reports (Cannon et al 1994; Keefe et
al 1994). In our work, the ability of neuropsychologic tests
to discriminate controls from relatives was strongest
among female subjects (Kremen et al 1997) and for
research participants younger than 60 years of age (Faraone et al 1995b, 1996).
Neuropsychologic studies of relatives are valuable for
two reasons. First, as we have discussed elsewhere (Faraone et al 1995a; Tsuang et al 1993), finding markers of the
vulnerability to schizophrenia may provide phenotypes for
genetic studies. For example, studies that have incorporated eye-tracking (Arolt et al 1996) and sensory gating
measures (Coon et al 1993), have shown these phenotypes
to suggest genetic linkage where the clinical diagnosis did
not. Second, unlike studies of patients, studies of relatives
are not confounded by neuroleptic treatment, chronic
hospitalization, and the potential neurotoxic effects of
psychosis (Faraone et al 1999b).
For several reasons, many researchers have suggested
that schizophrenia is caused by a multifactorial process
comprised of several genes combined with adverse environmental factors (Gottesman 1991; Gottesman et al 1987;
Tsuang and Faraone 1994). Although the number of
0006-3223/00/$20.00
PII S0006-3223(99)00263-2
Neuropsychologic Functioning in Relatives of Schizophrenic Patients
schizophrenia genes is unknown, most researchers agree
that a single gene theory of schizophrenia is not viable,
even if that theory allows for many different single gene
variants (Gottesman and McGue 1990; Gottesman et al
1983; McGue et al 1983; McGuffin et al 1987). The
multifactorial model of schizophrenia has found some
support from segregation analysis studies (Faraone and
Tsuang 1985; Faraone et al 1988) and cannot be discounted as a viable model of the etiology of schizophrenia.
Although prior studies have concluded that neuropsychologic deficits in relatives of schizophrenic patients
reflect the effects of schizophrenia genes, they have not
examined the implications that the multifactorial model
has for the distribution of neuropsychologic deficits
among family members of schizophrenic patients. If a) the
multifactorial model is correct in positing the existence of
a graded genetic loading for schizophrenia and b) schizophrenia genes cause neuropsychologic impairment among
nonpsychotic relatives of schizophrenic patients, then the
degree of impairment in relatives should be associated
with their genetic loading. Because we have no gold
standard measure of genetic loading, we defined that
construct using family data. This approach assumes that
nonpsychotic relatives having only one first-degree relative with schizophrenia have a lower genetic loading than
those having two relatives with schizophrenia. Thus, to
test the hypothesis that the degree of neuropsychologic
impairment in relatives is associated with their genetic
loading for schizophrenia, we compared relatives from
families having one and two schizophrenic members.
Methods and Materials
Research participants in this study were first-degree, nonpsychotic relatives of schizophrenic patients and control subjects.
The patients were recruited from inpatient and outpatient units at
three Boston area hospitals specializing in the care of chronically
ill psychotic patients: the Brockton/West Roxbury Veterans
Affairs Medical Center, the Taunton State Hospital, and the
Massachusetts Mental Health Center. Inclusion criteria for patients were as follows: a diagnosis of schizophrenia based on the
Diagnostic and Statistical Manual of Mental Disorders, 3rd
edition revised (DSM-III-R; American Psychiatric Association
1987); age greater than 17 years; English as the primary
language; at least an eighth grade education; and available
first-degree relatives.
For probands, DSM-III-R diagnoses were derived from structured interviews using the Schedule for Affective Disorders and
Schizophrenia (Endicott and Spitzer 1978). Relatives were interviewed with the Structured Clinical Interview for DSM-III-R
(SCID; Spitzer et al 1987) for Axis I disorders and the Structured
Interview for DSM-III Personality Disorders (SIDP; Stangl and
Zimmerman 1983).
To be eligible for the study, relatives and controls had to be
between 18 and 59 years of age, have at least an eighth grade
BIOL PSYCHIATRY
2000;48:120 –126
121
education, have English as their first language, and be free of
psychosis during their lifetime. The exclusion criteria for both
controls and relatives also required absence of 1) substance abuse
within the past 6 months; 2) history of head injury with any
documented cognitive sequelae or with loss of consciousness
greater than 5 min; 3) neurologic disease or damage; 4) brain
surgery; 5) mental retardation; and 6) medical illnesses that may
significantly impair neurocognitive function. We recruited 41
relatives from families having only one schizophrenic relative
(simplex families, 29 male relatives, 12 female relatives) and 36
from families having two schizophrenic relatives (26 male
relatives, 10 female relatives). For simplex families, we ruled out
schizophrenia in first-degree relatives by direct interviews and in
second-degree relatives by the family history method. For
multiplex families, the two schizophrenic relatives and the
nonschizophrenic relatives were all first-degree relatives of one
another.
We recruited 100 control subjects through advertisements in
the catchment areas of the hospitals from which the probands had
been ascertained (58 male subjects, 42 female subjects). Control
subjects went through the same screening process as the relatives, with one exception. Instead of using a structured interview,
we screened control subjects for psychopathology using the short
form of the Minnesota Multiphasic Personality Inventory
(MMPI-168; Vincent et al 1984). Potential control subjects were
excluded if any MMPI scale, except Masculinity-Femininity,
was above 70.
This report focuses on the following tests, which our prior
studies implicated as risk indicators in schizophrenia families:
the Wisconsin Card Sorting Test (WCST), the immediate and
delayed recall conditions of the Wechsler Memory ScalesRevised (WMS-R) Logical Memory stories, the immediate and
delayed recall scores on the WMS-R Visual Reproductions, and
a dichotic (digits) listening task. We also estimated intelligence
from the vocabulary and block design subtests of the Wechsler
Adult Intelligence Scale-Revised (WAIS-R) and measured academic achievement with the reading, arithmetic, and spelling
subtests of the Wide Range Achievement Test-Revised
(WRAT-R).
The psychiatric interviews of relatives were blind to the
neuropsychologic test results from the relative. Diagnoses of
probands and relatives were blind to all other psychiatric and
neuropsychologic data.
We used logistic regression, linear regression, and ordinal
logistic regression to make comparisons among groups. Because
our prior work had demonstrated gender effects in differences
between relatives of schizophrenic patients and control subjects,
we also included gender and its interaction with relative group in
our statistical models. Our statistical analyses addressed the
non-independence of observations within families by adjusting
variance estimates with Huber’s formula (Huber 1967), a “theoretical bootstrap” that produces accurate statistical tests for
clustered data. The method, as implemented in STATA (Stata
Corporation 1999) works by entering the cluster scores (i.e., sum
of scores within families) into the formula for the estimate of
variance using the linearization method (Binder 1983; Kish and
Frankel 1974). All tests were two tailed and used the .05 level of
significance.
122
S.V. Faraone et al
BIOL PSYCHIATRY
2000;48:120 –126
Table 1. Group Comparisons on Neuropsychologic Test Scores
Test scores: mean (SD)
Multiplex (n 5 36)
Intelligence and achievement
Estimated IQ
WRAT-R reading
WRAT-R arithmetic
WRAT-R spelling
Abstraction
WCST categories
WCST total perseverations
Verbal memory
WMS-R logical memories, immediate
WMS-R logical memories, delayed
Auditory attention
Dichotic listening, digits detected
Visual memory
WMS-R visual reproductions, immediate
WMS-R visual reproductions, delayed
101 (15.9)a
97 (13.9)
92 (9.5)
100 (8.3)
Simplex (n 5 41)
102 (10)
98 (11.4)
95 (13.3)
98 (13.1)
Controls (n 5 100)
109 (14.3)
103 (12.4)
97 (13.6)
100 (13.5)
Statistic
p value
4.4
2.9
0.9
0.5
.01
.06
.38
.62
x2(2) 5 2.4
F(2,149) 5 1.0
.31
.36
29.8 (5.6)
26 (6.4)
F(2,150) 5 12.1
F(2,150) 5 7.7
,.001
,.001
5.6 (1.2)
9.5 (14.9)
F(2,148)
F(2,149)
F(2,113)
F(2,113)
5
5
5
5
5.1 (1.6)
14.9 (17.8)
5.2 (1.7)
10 (15)
22.6 (7.3)a,b
19.6 (7.7)a,b
26.6 (6.7)a
23.7 (7.6)
97.4 (23.5)
96 (22)
101.2 (21.3)
F(2,145) 5 0.9
.42
29.4 (6.1)a,b
26.9 (8.7)
33 (4.8)c
30.7 (6)
34.7 (4.5)
31.2 (5.9)
F(2,149) 5 7.8
F(2,148) 5 2.1
,.001
.13
a
p , .01 vs. control subjects.
p , .05 vs. simplex.
c
p , .05 vs. control subjects.
WRAT-R, Wide Range Achievement Test-Revised; WCST, Wisconsin Card Sorting Test; WMS-R, Wechsler Memory Scale-Revised.
b
Results
The three groups did not differ in parental education (x2[2]
5 0.1, p 5 .9), ethnicity (x2[10] 5 11.6, p 5 .3), gender
(x2[2] 5 3.5, p 5 .2) or age (F[2,150] 5 1.9, p 5 .2).
Table 1 shows the results of group comparisons on
neuropsychologic test scores. We found significant threeway group differences for estimated intelligence, WMS-R
immediate and delayed logical memories, and WMS-R
immediate visual reproductions.
Pairwise comparisons showed that the relatives from
multiplex families differed significantly from controls on
all of these variables, even after correcting for the group
difference in estimated intelligence. In contrast, in comparisons with control subjects, relatives from simplex
families differed only on one test score: WMS-R immediate logical memories. Comparisons between relatives
from multiplex and simplex families showed that the
former group had significantly worse scores for WMS-R
immediate and delayed logical memories, and WMS-R
immediate visual reproductions. The multiplex and simplex subjects did not differ in the sizes of their families of
origin (6.5 vs. 5.0, Wilcoxon z 5 1.5, p 5 .13). Thus,
the neuropsychologic differences between these groups
cannot be attributed to difference in family size. Notably,
age could not account for group differences given the
similar ages in our three groups: 40 6 9 for multiplex,
41 6 11 for simplex, and 38 6 11 for control subjects
(F[2,150] 5 1.9, p 5 .2).
When comparing relatives from simplex and multiplex
families, we found group 3 gender interactions for four
variables: auditory attention (t[51] 5 2.1, p 5 .04),
WMS-R delayed logical memories (t[51] 5 2.7, p 5
.01) and WMS-R immediate (t[51] 5 2.1, p 5 .04)
and delayed visual reproductions (t[51] 5 2.1, p 5
.04). Follow-up analyses by gender revealed significant
multiplex–simplex differences for female but not for male
subjects. For auditory attention, female subjects from
multiplex families were marginally more impaired than
female subjects from simplex families (90 6 21 vs. 105 6
12, t[19] 5 1.9, p 5 .07), but we found no such
difference for male subjects (100 6 24 vs. 92 6 24, t[19]
5 1.2, p 5 .24). The same pattern was seen for
immediate visual memory (female subjects: 29 6 6 vs.
36 6 3, t[19] 5 3.9, p 5 .001; male subjects: 30 6 6
vs. 32 6 5, t[19] 5 1.0, p 5 .33), delayed visual
memory (female subjects: 26 6 10 vs. 35 6 3, t[19] 5
3.0, p 5 .008; male subjects: 28 6 8 vs. 29 6 6, t[19]
5 0.5, p 5 .6) and delayed verbal memory (female
subjects: 17 6 9 vs. 29 6 7, t[19] 5 3.5, p 5 .002;
male subjects: 21 6 7 vs. 22 6 7, t[19] 5 0.4, p 5 .7).
Discussion
Our prior studies suggested that neuropsychologic impairments in relatives of schizophrenic patients are stable traits
caused by the set of genes that also increases the predisposition to schizophrenia (Faraone et al 1995b, 1996;
1999a; Kremen et al 1997; Lyons et al 1995; Toomey et al
1998). The present report extends that finding by showing
that relatives from families having two schizophrenic
members have more neuropsychologic impairment than
relatives from families having one schizophrenic member.
Neuropsychologic Functioning in Relatives of Schizophrenic Patients
This new finding is consistent with the multifactorial
model of schizophrenia (Gottesman 1991; Gottesman et al
1987; Tsuang and Faraone 1994). This model posits that
no one gene or environmental factor causes schizophrenia.
Instead, it is the sum of these genes and environmental
factors that leads to the disorder. If this is true, then there
must be a graded genetic predisposition to the disorder
such that the probability of developing schizophrenia (or
showing related neuropsychologic impairments) increases
as the degree of predisposition increases. Presumably,
multiplex families harbor more schizophrenia genes than
simplex families. Thus, our finding of greater impairments
in relatives in multiplex families is consistent with the
multifactorial model.
Our results, however, do not address the genetic versus
environmental causes of neuropsychologic deficits given
the inferential limitations of family studies that do not
include twin or adoptive relatives (Faraone and Santangelo
1992; Faraone and Tsuang 1995; Faraone et al 1999b).
Also, our data cannot determine why some relatives are
predisposed to schizophrenia but never developed the
disorder. Further work is needed to determine if they have
a low “dose” of genes or if they had not been exposed to
environmental agents that trigger the disorder.
Notably, the group differences we found were strongest
for the memory measures. Although we cannot draw
definitive conclusions about the nature of their memory
problems, we can suggest several points worthy of follow-up by future research. The memory dysfunction is
seen in both visual and verbal domains, and the memory
tasks used are typically considered measures of long-term
memory. We have, however, previously demonstrated
(Faraone et al 1995b) that, unlike patients with schizophrenia (Seidman et al 1998), the relatives do not have
abnormal rates of forgetting compared with control subjects (i.e., they do not have typical long-term memory
deficits). Thus, the memory deficits appear to be associated with defects in the acquisition and/or the retrieval of
information. Further research is needed to shed light on the
nature of the cognitive dysfunctions underlying impaired
task performance. Possible processes include defects in
working memory, sustained attention, and encoding
processes.
We cannot attribute our findings to group differences in
general intellectual functioning for two reasons. First, the
relatives from simplex and multiplex families did not
differ in estimated intelligence. Second, although the
relatives from multiplex families showed significantly
lower intelligence than control subjects, statistical corrections for this confound did not change our results.
We previously reported significant or near-significant
group 3 gender interactions in four neuropsychologic
functions: verbal memory, motor function, mental control/
BIOL PSYCHIATRY
2000;48:120 –126
123
encoding, and auditory attention (Faraone et al 1999a;
Kremen et al 1997). At the follow-up assessment (Faraone
et al 1999a), the group 3 gender interaction was significant for delayed and immediate verbal memory, and
delayed and immediate visual memory. At both assessments, the nature of the interactions was the same: the
greater impairment of relatives compared with control
subjects was more pronounced for female than for male
subjects. The present work extends these findings by
showing that the gender effect is found not only when
comparing relatives of schizophrenic patients with control
subjects, but also when comparing relatives from simplex
and multiplex schizophrenia families.
Although we have no definitive explanation for these
gender differences, they suggest that men carrying the
familial risk for schizophrenia have a lower threshold for
developing schizophrenia than women predisposed to
schizophrenia. If that were so, then male relatives with
neuropsychologic deficits would have been more likely to
have developed psychosis and have been excluded from
our study. In contrast, if females are able to sustain greater
impairment without developing psychosis, they would
have been more likely to remain in our sample. It is also
possible that men with neuropsychologic deficits are less
likely to participate than are women with these deficits.
The idea that females have a higher threshold for
expressing psychosis is consistent with epidemiologic
evidence suggesting that males might be at greater risk for
the disorder and that relatives of female schizophrenic
patients might be at greater risk for schizophrenia than
relatives of male patients (Goldstein 1995). Both of these
findings are consistent with the idea that women are less
likely to develop schizophrenia than men given the same
degree of familial predisposition. Yet, because epidemiologic studies have not consistently shown these gender
effects, this idea requires further confirmation. Moreover,
although the gender differences we found are intriguing,
our inferences are limited due to the small subsamples
created after stratifying the sample on gender and the fact
that this issue has not been addressed by other
investigators.
Notably, we did not find group differences for the
Wisconsin Card Sorting Test (WCST), despite a fairly
large literature showing that test to be impaired among
schizophrenic patients (e.g., Braff et al 1991; Koren et al
1998; Seidman et al 1997); however, the literature about
WCST deficits in relatives of schizophrenic patients is
contradictory. In two different samples, Pogue-Geile et al
1991, 1989) found that siblings of schizophrenic patients
did worse on the WCST than control subjects. Mirsky et al
(1992) found significant WCST deficits in two different
samples as well. Similar findings were reported from a
twin study by Goldberg et al (1997). Despite these positive
124
S.V. Faraone et al
BIOL PSYCHIATRY
2000;48:120 –126
findings, other studies have not found impaired WCST
performance among either first-degree (Condray and
Steinhauer 1992; Keefe et al 1994; Scarone et al 1997;
Stratta et al 1997) or monozygotic co-twin (Goldberg et al
1990) relatives of schizophrenic patients. Moreover, in a
4-year follow-up study, we showed that WCST scores in
relatives of schizophrenic patients were not stable over
time, raising questions about their utility as trait measures
(Faraone et al 1999a).
Taken together, these conflicting results suggest that the
WCST may be a weak measure of the genetic vulnerability
to schizophrenia. Notably, the WCST is relatively easy;
normal children achieve adult levels by age 10 –12 years
(Heaton 1981; Heaton et al 1993). Unlike more discriminating measures of the vulnerability to schizophrenia, such
as the identical pairs version of the continuous performance task (Cornblatt and Erlenmeyer-Kimling 1984;
Cornblatt and Keilp 1994), or the Wechsler Logical
Memories (Faraone et al 1999a), the WCST has no time
constraint in terms of its stimulus processing load or
response time. Thus, the WCST may simply be too easy to
pick up the subtle neuropsychologic deficts seen in most
relatives of schizophrenic patients.
Our work must be interpreted in the context of its
methodological limitations. Our probands came from centers that specialize in the treatment of chronically ill
psychotic patients. Thus, although they are likely to be
representative of chronic, relatively severe cases of schizophrenia, our results may not generalize to relatives ascertained through milder cases. Another potential problem is
that, although we used a broad neuropsychologic battery,
we did not test all aspects of neuropsychologic
functioning.
Also, as we have discussed elsewhere, defining simplex
and multiplex families is subject to error (Faraone et al
1999b). Thus, some of the simplex patients may not be
genetically different from the multiplex patients; but,
despite this uncertainty, as a group, the simplex patients
should have a lower genetic loading for schizophrenia than
the multiplex group. Moreover, misclassifying multiplex
families as simplex would make it more difficult to find
group differences and thus cannot account for our results.
We must also be cautious when inferring specific
neuropsychologic function deficits from test performance.
Although it is reasonable to view each test score as
primarily influenced by an underlying function, the tests
are complex and their scores multidetermined. Thus, to
understand more fully the nature of the deficit we have
documented among the relatives of schizophrenic patients,
future work should analyze these deficits in terms of more
fine-grained measures of information processing
components.
Despite these limitations, our results suggest that the
amount of neuropsychologic impairment in relatives of
schizophrenic patients increases with their genetic loading
for schizophrenia. This provides further, indirect, evidence
that one or more of the genes that cause schizophrenia also
can be expressed as neuropsychologic deficits in relatives.
Preparation of this article was supported in part by the National Institute
of Mental Health Grants 1 R01MH41874-01, 5 UO1 MH46318-02 and 1
R37MH43518-01 to Dr. Ming T. Tsuang and by grants from the Stanley
Foundation and the National Alliance for Research for Schizophrenia and
Affective Disorders to Larry J. Seidman.
We thank the following people for their contributions to the project:
Gwen Barnes, Mimi Braude, Deborha Catt, Jo-Ann Donatelli, Tova
Ferro, Lisa Gabel, David Goldfinger, Robin Green, Lynda Jacobs, Denise
Leville, Cathy Monaco, Anne Shore, Rob Trachtenberg, and Drs. Keith
Hawkins, Marlene Oscar-Berman and Theresa Pai. We also thank Drs.
Anna Mitchell, David Osser, and Larry Albert of the Taunton State
Hospital for helping us with the recruitment of research participants.
References
American Psychiatric Association (1987): Diagnostic and Statistical Manual of Mental Disorders (Third Edition, Revised).
Washington, DC: American Psychiatric Association.
Arolt V, Lencer R, Achim N, Muller-Myhsok B, Purmann S,
Schurmann M, et al (1996): Eye tracking dysfunction is a
putative phenotypic susceptibility marker of schizophrenia
and maps to a locus on chromosome 6p in families with
multiple occurrence of the disease. Am J Med Genet Neuropsychiatr Genet 67:560 –563.
Binder DA (1983): On the variances of asymptotically normal
estimators from complex surveys. Int Stat Rev 51:279 –292.
Braff DL, Heaton R, Kuck J, Cullum M, Moranville J, Grant I,
et al (1991): The generalized pattern of neuropsychological
deficits in outpatients with chronic schizophrenia with heterogeneous Wisconsin Card Sorting Test results. Arch Gen
Psychiatry 48:891– 898.
Cannon TD, Zorrilla LE, Shtasel D, Gur RE, Gur RC, Marco EJ,
et al (1994): Neuropsychological functioning in siblings
discordant for schizophrenia and healthy volunteers. Arch
Gen Psychiatry 51:651– 661.
Condray R, Steinhauer SR (1992): Schizotypal personality disorder in individuals with and without schizophrenic relatives:
Similarities and contrasts in neurocognitive and clinical
functioning. Schiz Res 7:33– 41.
Coon H, Plaetke R, Holik J, Hoff M, Myles-Worsley M, Waldo
M, et al (1993): Use of a neurophysiological trait in linkage
analysis of schizophrenia. Biol Psychiatry 34:277–289.
Cornblatt B, Erlenmeyer-Kimling L (1984): Early attentional
predictors of adolescent behavioral disturbances in children at
risk for schizophrenia. In: Watt NF, Anthony EJ, Wynne LC,
Rolf JE, editors. Children at Risk for Schizophrenia: A
Longitudinal Perspective. New York: Cambridge University
Press, 198 –211.
Cornblatt BA, Keilp JG (1994): Impaired attention, genetics, and
the pathophysiology of schizophrenia. Schiz Bull 20:31– 46.
Endicott J, Spitzer RL (1978): A diagnostic interview: The
Neuropsychologic Functioning in Relatives of Schizophrenic Patients
Schedule for Affective Disorders and Schizophrenia. Arch
Gen Psychiatry 35:837– 844.
Faraone SV, Kremen WS, Lyons MJ, Pepple JR, Seidman LJ,
Tsuang MT (1995a): Diagnostic accuracy and linkage analysis: How useful are schizophrenia spectrum phenotypes?
Am J Psychiatry 152:1286 –1290.
Faraone SV, Lyons MJ, Tsuang MT (1988): Schizophrenia:
Mathematical models of genetic transmission. In: Tsuang
MT, Simpson JC, editors. Nosology, Epidemiology and Genetics, Vol 3. Amsterdam: Elsevier Science, 501–530.
Faraone SV, Santangelo S (1992): Methods in genetic epidemiology. In: Fava M, Rosenbaum JF, editors. Research Designs
and Methods in Psychiatry. Amsterdam: Elsevier, 87–105.
Faraone SV, Seidman LJ, Kremen WS, Pepple JR, Lyons MJ,
Tsuang MT (1995b): Neuropsychological functioning among
the nonpsychotic relatives of schizophrenic patients: A diagnostic efficiency analysis. J Abnorm Psychol 104:286 –304.
Faraone SV, Seidman LJ, Kremen WS, Toomey R, Lyons MJ,
Tsuang MT (1996): Neuropsychological functioning among
the elderly nonpsychotic relatives of schizophrenic patients.
Schiz Res 21:27–31.
Faraone SV, Seidman LJ, Kremen WS, Toomey R, Pepple JR,
Tsuang MT (1999a): Neuropsychological functioning among
the nonpsychotic relatives of schizophrenic patients: A fouryear follow-up study. J Abnorm Psychol 108:176 –181.
Faraone SV, Tsuang D, Tsuang MT (1999b): Genetics of Mental
Disorders: A Guide for Students, Clinicians, and Researchers. New York: Guilford.
Faraone SV, Tsuang MT (1985): Quantitative models of the
genetic transmission of schizophrenia. Psychol Bull 98:41–
66.
Faraone SV, Tsuang MT (1995): Methods in psychiatric genetics. In: Tsuang MT, Tohen M, Zahner GEP, (editors).
Textbook in Psychiatric Epidemiology. New York: WileyLiss, 81–134.
Goldberg TE, Ragland D, Torrey EF, Gold JM, Bigelow LB,
Weinberger DR (1990): Neuropsychological assessment of
monozygotic twins discordant for schizophrenia. Arch Gen
Psychiatry 47:1066 –1072.
Goldberg TE, Torrey EF, Gold JM, Bigelow LB, Ragland RD,
Taylor E, et al (1997): Genetic risk of neuropsychological
impairment in schizophrenia: A study of monozygotic twins
discordant and concordant for the disorder. Schiz Res 17:77–
84.
Goldstein JM (1995): Gender and the familial transmission of
schizophrenia. In: Seeman MV, (editor). Gender and Psychopathology. Washington, DC: American Psychiatric Press,
201–226.
Gottesman II, McGuffin P, Farmer AE (1987): Clinical genetics
as clues to the “real” genetics of schizophrenia (a decade of
modest gains while playing for time). Schizophr Bull 13:23–
47.
Gottesman II (1991): Schizophrenia Genesis: The Origin of
Madness. New York: Freeman.
Gottesman II, Carey G, Hanson DR (1983): Pearls and perils in
epigenetic psychopathology. In: Guze SB, Earls FJ, Barrett
JE, editors. American Psychopathological Association Series.
Childhood psychopathology and development. New York:
Raven Press, 287–300.
BIOL PSYCHIATRY
2000;48:120 –126
125
Gottesman II, McGue M (1990): Mixed and mixed-up models for
the transmission of schizophrenia. In: Cicchetti D, Grove W,
editors. Thinking Clearly About Psychology: Essays in Honor
of Paul E. Meehl. Minneapolis: University of Minnesota
Press.
Heaton RK (1981): Wisconsin Card Sorting Test Manual.
Odessa, FL: Psychological Assessment Resources.
Heaton RK, Chelune GJ, Talley JL, Kay GG, Curtiss G (1993):
Wisconsin Card Sorting Test Manual–Revised. Odessa, FL:
Psychological Assessment Resources.
Huber PJ (1967): The behavior of maximum likelihood estimates
under non-standard conditions. Proc Fifth Berkeley Symposium on Mathematical Statistics and Probability 1:221–233.
Keefe RSE, Silverman JM, Roitman SEL, Harvey PD, Duncan
MA, Alroy D, et al (1994): Performance of nonpsychotic
relatives of schizophrenic patients on cognitive tests. Psychiatry Res 53:1–12.
Kish L, Frankel MR (1974): Inference from complex samples. J
Roy Stat Soc A:1–37.
Koren D, Seidman LJ, Harrison RH, Lyons MJ, Kremen WS,
Caplan B, et al (1998): Factor structure of the Wisconsin Card
Sorting Test: Dimensions of deficit in schizophrenia. Neuropsychology 12:289 –302.
Kremen WS, Goldstein JM, Seidman LJ, Toomey R, Lyons MJ,
Tsuang MT, et al (1997): Sex differences in neuropsychological function in nonpsychotic relatives of schizophrenic probands. Psychiatry Res 66:131–144.
Kremen WS, Seidman LJ, Pepple JR, Lyons MJ, Tsuang MT,
Faraone SV (1994): Neuropsychological risk indicators for
schizophrenia: A review of family studies. Schiz Bull 20:103–
119.
Lyons MJ, Toomey R, Seidman LJ, Kremen WS, Faraone SV,
Tsuang MT (1995): Verbal learning and memory in relatives
of schizophrenics: Preliminary findings. Biol Psychiatry 37:
750 –753.
McGue M, Gottesman II, Rao DC (1983): The transmission of
schizophrenia under a multifactorial threshold model. Am J
Hum Genet 35:1161–1178.
McGuffin P, Farmer A, Gottesman II (1987): Is there really a
split in schizophrenia? The genetic evidence. Br J Psychiatry
150:581–592.
Mirsky AF, Lochhead SJ, Jones BP, Kugelmass S, Walsh D,
Kendler KS (1992): On familial factors in the attentional
deficit in schizophrenia: A review and report of two new
subject samples. J Psychiatr Res 26:383– 403.
Pogue-Geile MF, Garrett AH, Brunke JJ, Hall JK (1991):
Neuropsychological impairments are increased in siblings of
schizophrenic patients. Schiz Res 4:390.
Pogue-Geile MF, Watson JR, Steinhauer SR, Goldstein G
(1989): Neuropsychological impairments among siblings of
schizophrenic probands. Schiz Res 2:70.
Scarone S, Abbruzzese M, Gambini O (1997): The Wisconsin
Card Sorting Test discriminates schizophrenic patients and
their siblings. Schiz Res 10:103–107.
Seidman LJ, Goldstein JM, Goodman JM, Koren D, Turner WM,
Faraone SV, et al (1997): Sex differences in olfactory
identification and Wisconsin Card Sorting performance in
schizophrenia: Relationship to attention and verbal ability.
Biol Psychiatry 42:104 –115.
126
BIOL PSYCHIATRY
2000;48:120 –126
Seidman LJ, Stone WS, Jones R, Harrison RH, Mirsky AF
(1998): Cognitive effects of schizophrenia and temporal
lobe epilepsy on memory. J Int Neuropsychol Soc 4:342–
352.
Spitzer RL, Williams JBW, Gibbon M (1987): Structured Clinical Interview for DSM-III-R Personality Disorders (SCIDII). New York: Biometrics Research Department, New York
State Psychiatric Institute.
Stangl D, Zimmerman M (1983): Structured Interview for the
DSM-III Personality Disorders, 2nd ed. Iowa City, IA: The
University of Iowa.
Stata Corporation (1999): Stata Reference Manual: Release 6.0.
College Station, TX: Stata Corporation.
Stratta P, Daneluzzo E, Mattei P, Bustini M, Casacchia M, Rossi
A (1997): No deficit in Wisconsin Card Sorting Test perfor-
S.V. Faraone et al
mance of schizophrenic patients’ first-degree relatives.
Schizophr Res 26:147–51.
Toomey R, Faraone SV, Seidman LJ, Kremen WS, Pepple JR,
Tsuang MT (1998): Association of vulnerability markers in
relatives of schizophrenic patients. Schiz Res 31:89 –98.
Tsuang MT, Faraone SV (1994): Epidemiology and behavioral
genetics of schizophrenia. In: Watson SJ, editor. Biology of
Schizophrenia and Affective Disease. New York: Raven
Press, 163–195.
Tsuang MT, Faraone SV, Lyons MJ (1993): Identification of the
phenotype in psychiatric genetics. Eur Arch Psychiatr Neurol
Sci 243:131–142.
Vincent KR, Castillo IM, Hauser RI, Zapata JA, Stuart HJ, Cohn
CK, et al (1984): MMPI-168 Codebook. Norwood, NJ: Ablex
Publishing Corporation.
Relatives of Schizophrenic Patients:
The Effect of Genetic Loading
Stephen V. Faraone, Larry J. Seidman, William S. Kremen, Rosemary Toomey,
John R. Pepple, and Ming T. Tsuang
Background: We previously reported that the nonpsychotic relatives of schizophrenic patients exhibited disturbances in executive functioning, verbal and visual memory, auditory attention, mental control, and verbal ability.
In a 4-year follow-up, we showed that the discriminating
power of most of these tests was stable over time.
Methods: In this report we compare 41 nonpsychotic
persons who have only one schizophrenic first-degree
relative (simplex families) with 36 nonpsychotic persons
who have two schizophrenic first-degree relatives (multiplex families). Our goal was to test a hypothesis that
neuropsychologic deficits would be worse among the
latter.
Results: Relatives from multiplex families differed significantly from controls on estimated intelligence, immediate
and delayed logical memories, and immediate visual
reproductions. In contrast, in comparisons with controls,
relatives from simplex families only differed on immediate
logical memories. Comparisons between relatives from
multiplex and simplex families showed that the former
group had significantly worse scores for estimated intelligence, immediate and delayed logical memories, and
immediate visual reproductions. We also found group 3
gender interactions: the worse performance of the multiplex group was seen for females.
Conclusions: These results are consistent with the idea
that neuropsychologic deficits in relatives of schizophrenic patients reflect their degree of genetic predisposition to schizophrenia. They also suggest hypotheses
about gender differences in the familial transmission of
From the Harvard Medical School Department of Psychiatry at the Massachusetts
Mental Health Center and Brockton/West Roxbury VA Medical Center,
Brockton (SVF, LJS, RT, JRP, MTT), Psychiatry Service, Massachusetts
General Hospital, Boston (SVF, LJS, MTT), Harvard Institute of Psychiatric
Epidemiology and Genetics, Boston (SVF, LJS, RT, JRP, MTT), and the
Department of Epidemiology, Harvard School of Public Health, Boston (MTT),
Massachusetts, and the Department of Psychiatry, University of California,
Davis School of Medicine, Sacramento and UC Davis–Napa Psychiatric
Research Center, Napa State Hospital, Napa, California (WSK).
Address reprint requests to Stephen V. Faraone, PhD, Harvard Medical School,
Suite 255, 750 Washington Street, South Easton MA 02375.
Received June 16, 1999; revised September 16, 1999; accepted September 21,
1999.
© 2000 Society of Biological Psychiatry
the disorder. Biol Psychiatry 2000;48:120 –126 © 2000
Society of Biological Psychiatry
Key Words: Genetics, neuropsychology, endophenotypes, multiplex families
Introduction
W
e previously reported that the nonpsychotic relatives
of schizophrenic patients exhibited disturbances in
executive functioning, verbal and visual memory, auditory
attention, mental control, and verbal ability (Faraone et al
1995b; Toomey et al 1998) and that most of these findings
were stable over a 4-year period (Faraone et al 1999a).
These findings were consistent with the findings in relatives of schizophrenic patients reviewed by Kremen et al
(1994) and with other reports (Cannon et al 1994; Keefe et
al 1994). In our work, the ability of neuropsychologic tests
to discriminate controls from relatives was strongest
among female subjects (Kremen et al 1997) and for
research participants younger than 60 years of age (Faraone et al 1995b, 1996).
Neuropsychologic studies of relatives are valuable for
two reasons. First, as we have discussed elsewhere (Faraone et al 1995a; Tsuang et al 1993), finding markers of the
vulnerability to schizophrenia may provide phenotypes for
genetic studies. For example, studies that have incorporated eye-tracking (Arolt et al 1996) and sensory gating
measures (Coon et al 1993), have shown these phenotypes
to suggest genetic linkage where the clinical diagnosis did
not. Second, unlike studies of patients, studies of relatives
are not confounded by neuroleptic treatment, chronic
hospitalization, and the potential neurotoxic effects of
psychosis (Faraone et al 1999b).
For several reasons, many researchers have suggested
that schizophrenia is caused by a multifactorial process
comprised of several genes combined with adverse environmental factors (Gottesman 1991; Gottesman et al 1987;
Tsuang and Faraone 1994). Although the number of
0006-3223/00/$20.00
PII S0006-3223(99)00263-2
Neuropsychologic Functioning in Relatives of Schizophrenic Patients
schizophrenia genes is unknown, most researchers agree
that a single gene theory of schizophrenia is not viable,
even if that theory allows for many different single gene
variants (Gottesman and McGue 1990; Gottesman et al
1983; McGue et al 1983; McGuffin et al 1987). The
multifactorial model of schizophrenia has found some
support from segregation analysis studies (Faraone and
Tsuang 1985; Faraone et al 1988) and cannot be discounted as a viable model of the etiology of schizophrenia.
Although prior studies have concluded that neuropsychologic deficits in relatives of schizophrenic patients
reflect the effects of schizophrenia genes, they have not
examined the implications that the multifactorial model
has for the distribution of neuropsychologic deficits
among family members of schizophrenic patients. If a) the
multifactorial model is correct in positing the existence of
a graded genetic loading for schizophrenia and b) schizophrenia genes cause neuropsychologic impairment among
nonpsychotic relatives of schizophrenic patients, then the
degree of impairment in relatives should be associated
with their genetic loading. Because we have no gold
standard measure of genetic loading, we defined that
construct using family data. This approach assumes that
nonpsychotic relatives having only one first-degree relative with schizophrenia have a lower genetic loading than
those having two relatives with schizophrenia. Thus, to
test the hypothesis that the degree of neuropsychologic
impairment in relatives is associated with their genetic
loading for schizophrenia, we compared relatives from
families having one and two schizophrenic members.
Methods and Materials
Research participants in this study were first-degree, nonpsychotic relatives of schizophrenic patients and control subjects.
The patients were recruited from inpatient and outpatient units at
three Boston area hospitals specializing in the care of chronically
ill psychotic patients: the Brockton/West Roxbury Veterans
Affairs Medical Center, the Taunton State Hospital, and the
Massachusetts Mental Health Center. Inclusion criteria for patients were as follows: a diagnosis of schizophrenia based on the
Diagnostic and Statistical Manual of Mental Disorders, 3rd
edition revised (DSM-III-R; American Psychiatric Association
1987); age greater than 17 years; English as the primary
language; at least an eighth grade education; and available
first-degree relatives.
For probands, DSM-III-R diagnoses were derived from structured interviews using the Schedule for Affective Disorders and
Schizophrenia (Endicott and Spitzer 1978). Relatives were interviewed with the Structured Clinical Interview for DSM-III-R
(SCID; Spitzer et al 1987) for Axis I disorders and the Structured
Interview for DSM-III Personality Disorders (SIDP; Stangl and
Zimmerman 1983).
To be eligible for the study, relatives and controls had to be
between 18 and 59 years of age, have at least an eighth grade
BIOL PSYCHIATRY
2000;48:120 –126
121
education, have English as their first language, and be free of
psychosis during their lifetime. The exclusion criteria for both
controls and relatives also required absence of 1) substance abuse
within the past 6 months; 2) history of head injury with any
documented cognitive sequelae or with loss of consciousness
greater than 5 min; 3) neurologic disease or damage; 4) brain
surgery; 5) mental retardation; and 6) medical illnesses that may
significantly impair neurocognitive function. We recruited 41
relatives from families having only one schizophrenic relative
(simplex families, 29 male relatives, 12 female relatives) and 36
from families having two schizophrenic relatives (26 male
relatives, 10 female relatives). For simplex families, we ruled out
schizophrenia in first-degree relatives by direct interviews and in
second-degree relatives by the family history method. For
multiplex families, the two schizophrenic relatives and the
nonschizophrenic relatives were all first-degree relatives of one
another.
We recruited 100 control subjects through advertisements in
the catchment areas of the hospitals from which the probands had
been ascertained (58 male subjects, 42 female subjects). Control
subjects went through the same screening process as the relatives, with one exception. Instead of using a structured interview,
we screened control subjects for psychopathology using the short
form of the Minnesota Multiphasic Personality Inventory
(MMPI-168; Vincent et al 1984). Potential control subjects were
excluded if any MMPI scale, except Masculinity-Femininity,
was above 70.
This report focuses on the following tests, which our prior
studies implicated as risk indicators in schizophrenia families:
the Wisconsin Card Sorting Test (WCST), the immediate and
delayed recall conditions of the Wechsler Memory ScalesRevised (WMS-R) Logical Memory stories, the immediate and
delayed recall scores on the WMS-R Visual Reproductions, and
a dichotic (digits) listening task. We also estimated intelligence
from the vocabulary and block design subtests of the Wechsler
Adult Intelligence Scale-Revised (WAIS-R) and measured academic achievement with the reading, arithmetic, and spelling
subtests of the Wide Range Achievement Test-Revised
(WRAT-R).
The psychiatric interviews of relatives were blind to the
neuropsychologic test results from the relative. Diagnoses of
probands and relatives were blind to all other psychiatric and
neuropsychologic data.
We used logistic regression, linear regression, and ordinal
logistic regression to make comparisons among groups. Because
our prior work had demonstrated gender effects in differences
between relatives of schizophrenic patients and control subjects,
we also included gender and its interaction with relative group in
our statistical models. Our statistical analyses addressed the
non-independence of observations within families by adjusting
variance estimates with Huber’s formula (Huber 1967), a “theoretical bootstrap” that produces accurate statistical tests for
clustered data. The method, as implemented in STATA (Stata
Corporation 1999) works by entering the cluster scores (i.e., sum
of scores within families) into the formula for the estimate of
variance using the linearization method (Binder 1983; Kish and
Frankel 1974). All tests were two tailed and used the .05 level of
significance.
122
S.V. Faraone et al
BIOL PSYCHIATRY
2000;48:120 –126
Table 1. Group Comparisons on Neuropsychologic Test Scores
Test scores: mean (SD)
Multiplex (n 5 36)
Intelligence and achievement
Estimated IQ
WRAT-R reading
WRAT-R arithmetic
WRAT-R spelling
Abstraction
WCST categories
WCST total perseverations
Verbal memory
WMS-R logical memories, immediate
WMS-R logical memories, delayed
Auditory attention
Dichotic listening, digits detected
Visual memory
WMS-R visual reproductions, immediate
WMS-R visual reproductions, delayed
101 (15.9)a
97 (13.9)
92 (9.5)
100 (8.3)
Simplex (n 5 41)
102 (10)
98 (11.4)
95 (13.3)
98 (13.1)
Controls (n 5 100)
109 (14.3)
103 (12.4)
97 (13.6)
100 (13.5)
Statistic
p value
4.4
2.9
0.9
0.5
.01
.06
.38
.62
x2(2) 5 2.4
F(2,149) 5 1.0
.31
.36
29.8 (5.6)
26 (6.4)
F(2,150) 5 12.1
F(2,150) 5 7.7
,.001
,.001
5.6 (1.2)
9.5 (14.9)
F(2,148)
F(2,149)
F(2,113)
F(2,113)
5
5
5
5
5.1 (1.6)
14.9 (17.8)
5.2 (1.7)
10 (15)
22.6 (7.3)a,b
19.6 (7.7)a,b
26.6 (6.7)a
23.7 (7.6)
97.4 (23.5)
96 (22)
101.2 (21.3)
F(2,145) 5 0.9
.42
29.4 (6.1)a,b
26.9 (8.7)
33 (4.8)c
30.7 (6)
34.7 (4.5)
31.2 (5.9)
F(2,149) 5 7.8
F(2,148) 5 2.1
,.001
.13
a
p , .01 vs. control subjects.
p , .05 vs. simplex.
c
p , .05 vs. control subjects.
WRAT-R, Wide Range Achievement Test-Revised; WCST, Wisconsin Card Sorting Test; WMS-R, Wechsler Memory Scale-Revised.
b
Results
The three groups did not differ in parental education (x2[2]
5 0.1, p 5 .9), ethnicity (x2[10] 5 11.6, p 5 .3), gender
(x2[2] 5 3.5, p 5 .2) or age (F[2,150] 5 1.9, p 5 .2).
Table 1 shows the results of group comparisons on
neuropsychologic test scores. We found significant threeway group differences for estimated intelligence, WMS-R
immediate and delayed logical memories, and WMS-R
immediate visual reproductions.
Pairwise comparisons showed that the relatives from
multiplex families differed significantly from controls on
all of these variables, even after correcting for the group
difference in estimated intelligence. In contrast, in comparisons with control subjects, relatives from simplex
families differed only on one test score: WMS-R immediate logical memories. Comparisons between relatives
from multiplex and simplex families showed that the
former group had significantly worse scores for WMS-R
immediate and delayed logical memories, and WMS-R
immediate visual reproductions. The multiplex and simplex subjects did not differ in the sizes of their families of
origin (6.5 vs. 5.0, Wilcoxon z 5 1.5, p 5 .13). Thus,
the neuropsychologic differences between these groups
cannot be attributed to difference in family size. Notably,
age could not account for group differences given the
similar ages in our three groups: 40 6 9 for multiplex,
41 6 11 for simplex, and 38 6 11 for control subjects
(F[2,150] 5 1.9, p 5 .2).
When comparing relatives from simplex and multiplex
families, we found group 3 gender interactions for four
variables: auditory attention (t[51] 5 2.1, p 5 .04),
WMS-R delayed logical memories (t[51] 5 2.7, p 5
.01) and WMS-R immediate (t[51] 5 2.1, p 5 .04)
and delayed visual reproductions (t[51] 5 2.1, p 5
.04). Follow-up analyses by gender revealed significant
multiplex–simplex differences for female but not for male
subjects. For auditory attention, female subjects from
multiplex families were marginally more impaired than
female subjects from simplex families (90 6 21 vs. 105 6
12, t[19] 5 1.9, p 5 .07), but we found no such
difference for male subjects (100 6 24 vs. 92 6 24, t[19]
5 1.2, p 5 .24). The same pattern was seen for
immediate visual memory (female subjects: 29 6 6 vs.
36 6 3, t[19] 5 3.9, p 5 .001; male subjects: 30 6 6
vs. 32 6 5, t[19] 5 1.0, p 5 .33), delayed visual
memory (female subjects: 26 6 10 vs. 35 6 3, t[19] 5
3.0, p 5 .008; male subjects: 28 6 8 vs. 29 6 6, t[19]
5 0.5, p 5 .6) and delayed verbal memory (female
subjects: 17 6 9 vs. 29 6 7, t[19] 5 3.5, p 5 .002;
male subjects: 21 6 7 vs. 22 6 7, t[19] 5 0.4, p 5 .7).
Discussion
Our prior studies suggested that neuropsychologic impairments in relatives of schizophrenic patients are stable traits
caused by the set of genes that also increases the predisposition to schizophrenia (Faraone et al 1995b, 1996;
1999a; Kremen et al 1997; Lyons et al 1995; Toomey et al
1998). The present report extends that finding by showing
that relatives from families having two schizophrenic
members have more neuropsychologic impairment than
relatives from families having one schizophrenic member.
Neuropsychologic Functioning in Relatives of Schizophrenic Patients
This new finding is consistent with the multifactorial
model of schizophrenia (Gottesman 1991; Gottesman et al
1987; Tsuang and Faraone 1994). This model posits that
no one gene or environmental factor causes schizophrenia.
Instead, it is the sum of these genes and environmental
factors that leads to the disorder. If this is true, then there
must be a graded genetic predisposition to the disorder
such that the probability of developing schizophrenia (or
showing related neuropsychologic impairments) increases
as the degree of predisposition increases. Presumably,
multiplex families harbor more schizophrenia genes than
simplex families. Thus, our finding of greater impairments
in relatives in multiplex families is consistent with the
multifactorial model.
Our results, however, do not address the genetic versus
environmental causes of neuropsychologic deficits given
the inferential limitations of family studies that do not
include twin or adoptive relatives (Faraone and Santangelo
1992; Faraone and Tsuang 1995; Faraone et al 1999b).
Also, our data cannot determine why some relatives are
predisposed to schizophrenia but never developed the
disorder. Further work is needed to determine if they have
a low “dose” of genes or if they had not been exposed to
environmental agents that trigger the disorder.
Notably, the group differences we found were strongest
for the memory measures. Although we cannot draw
definitive conclusions about the nature of their memory
problems, we can suggest several points worthy of follow-up by future research. The memory dysfunction is
seen in both visual and verbal domains, and the memory
tasks used are typically considered measures of long-term
memory. We have, however, previously demonstrated
(Faraone et al 1995b) that, unlike patients with schizophrenia (Seidman et al 1998), the relatives do not have
abnormal rates of forgetting compared with control subjects (i.e., they do not have typical long-term memory
deficits). Thus, the memory deficits appear to be associated with defects in the acquisition and/or the retrieval of
information. Further research is needed to shed light on the
nature of the cognitive dysfunctions underlying impaired
task performance. Possible processes include defects in
working memory, sustained attention, and encoding
processes.
We cannot attribute our findings to group differences in
general intellectual functioning for two reasons. First, the
relatives from simplex and multiplex families did not
differ in estimated intelligence. Second, although the
relatives from multiplex families showed significantly
lower intelligence than control subjects, statistical corrections for this confound did not change our results.
We previously reported significant or near-significant
group 3 gender interactions in four neuropsychologic
functions: verbal memory, motor function, mental control/
BIOL PSYCHIATRY
2000;48:120 –126
123
encoding, and auditory attention (Faraone et al 1999a;
Kremen et al 1997). At the follow-up assessment (Faraone
et al 1999a), the group 3 gender interaction was significant for delayed and immediate verbal memory, and
delayed and immediate visual memory. At both assessments, the nature of the interactions was the same: the
greater impairment of relatives compared with control
subjects was more pronounced for female than for male
subjects. The present work extends these findings by
showing that the gender effect is found not only when
comparing relatives of schizophrenic patients with control
subjects, but also when comparing relatives from simplex
and multiplex schizophrenia families.
Although we have no definitive explanation for these
gender differences, they suggest that men carrying the
familial risk for schizophrenia have a lower threshold for
developing schizophrenia than women predisposed to
schizophrenia. If that were so, then male relatives with
neuropsychologic deficits would have been more likely to
have developed psychosis and have been excluded from
our study. In contrast, if females are able to sustain greater
impairment without developing psychosis, they would
have been more likely to remain in our sample. It is also
possible that men with neuropsychologic deficits are less
likely to participate than are women with these deficits.
The idea that females have a higher threshold for
expressing psychosis is consistent with epidemiologic
evidence suggesting that males might be at greater risk for
the disorder and that relatives of female schizophrenic
patients might be at greater risk for schizophrenia than
relatives of male patients (Goldstein 1995). Both of these
findings are consistent with the idea that women are less
likely to develop schizophrenia than men given the same
degree of familial predisposition. Yet, because epidemiologic studies have not consistently shown these gender
effects, this idea requires further confirmation. Moreover,
although the gender differences we found are intriguing,
our inferences are limited due to the small subsamples
created after stratifying the sample on gender and the fact
that this issue has not been addressed by other
investigators.
Notably, we did not find group differences for the
Wisconsin Card Sorting Test (WCST), despite a fairly
large literature showing that test to be impaired among
schizophrenic patients (e.g., Braff et al 1991; Koren et al
1998; Seidman et al 1997); however, the literature about
WCST deficits in relatives of schizophrenic patients is
contradictory. In two different samples, Pogue-Geile et al
1991, 1989) found that siblings of schizophrenic patients
did worse on the WCST than control subjects. Mirsky et al
(1992) found significant WCST deficits in two different
samples as well. Similar findings were reported from a
twin study by Goldberg et al (1997). Despite these positive
124
S.V. Faraone et al
BIOL PSYCHIATRY
2000;48:120 –126
findings, other studies have not found impaired WCST
performance among either first-degree (Condray and
Steinhauer 1992; Keefe et al 1994; Scarone et al 1997;
Stratta et al 1997) or monozygotic co-twin (Goldberg et al
1990) relatives of schizophrenic patients. Moreover, in a
4-year follow-up study, we showed that WCST scores in
relatives of schizophrenic patients were not stable over
time, raising questions about their utility as trait measures
(Faraone et al 1999a).
Taken together, these conflicting results suggest that the
WCST may be a weak measure of the genetic vulnerability
to schizophrenia. Notably, the WCST is relatively easy;
normal children achieve adult levels by age 10 –12 years
(Heaton 1981; Heaton et al 1993). Unlike more discriminating measures of the vulnerability to schizophrenia, such
as the identical pairs version of the continuous performance task (Cornblatt and Erlenmeyer-Kimling 1984;
Cornblatt and Keilp 1994), or the Wechsler Logical
Memories (Faraone et al 1999a), the WCST has no time
constraint in terms of its stimulus processing load or
response time. Thus, the WCST may simply be too easy to
pick up the subtle neuropsychologic deficts seen in most
relatives of schizophrenic patients.
Our work must be interpreted in the context of its
methodological limitations. Our probands came from centers that specialize in the treatment of chronically ill
psychotic patients. Thus, although they are likely to be
representative of chronic, relatively severe cases of schizophrenia, our results may not generalize to relatives ascertained through milder cases. Another potential problem is
that, although we used a broad neuropsychologic battery,
we did not test all aspects of neuropsychologic
functioning.
Also, as we have discussed elsewhere, defining simplex
and multiplex families is subject to error (Faraone et al
1999b). Thus, some of the simplex patients may not be
genetically different from the multiplex patients; but,
despite this uncertainty, as a group, the simplex patients
should have a lower genetic loading for schizophrenia than
the multiplex group. Moreover, misclassifying multiplex
families as simplex would make it more difficult to find
group differences and thus cannot account for our results.
We must also be cautious when inferring specific
neuropsychologic function deficits from test performance.
Although it is reasonable to view each test score as
primarily influenced by an underlying function, the tests
are complex and their scores multidetermined. Thus, to
understand more fully the nature of the deficit we have
documented among the relatives of schizophrenic patients,
future work should analyze these deficits in terms of more
fine-grained measures of information processing
components.
Despite these limitations, our results suggest that the
amount of neuropsychologic impairment in relatives of
schizophrenic patients increases with their genetic loading
for schizophrenia. This provides further, indirect, evidence
that one or more of the genes that cause schizophrenia also
can be expressed as neuropsychologic deficits in relatives.
Preparation of this article was supported in part by the National Institute
of Mental Health Grants 1 R01MH41874-01, 5 UO1 MH46318-02 and 1
R37MH43518-01 to Dr. Ming T. Tsuang and by grants from the Stanley
Foundation and the National Alliance for Research for Schizophrenia and
Affective Disorders to Larry J. Seidman.
We thank the following people for their contributions to the project:
Gwen Barnes, Mimi Braude, Deborha Catt, Jo-Ann Donatelli, Tova
Ferro, Lisa Gabel, David Goldfinger, Robin Green, Lynda Jacobs, Denise
Leville, Cathy Monaco, Anne Shore, Rob Trachtenberg, and Drs. Keith
Hawkins, Marlene Oscar-Berman and Theresa Pai. We also thank Drs.
Anna Mitchell, David Osser, and Larry Albert of the Taunton State
Hospital for helping us with the recruitment of research participants.
References
American Psychiatric Association (1987): Diagnostic and Statistical Manual of Mental Disorders (Third Edition, Revised).
Washington, DC: American Psychiatric Association.
Arolt V, Lencer R, Achim N, Muller-Myhsok B, Purmann S,
Schurmann M, et al (1996): Eye tracking dysfunction is a
putative phenotypic susceptibility marker of schizophrenia
and maps to a locus on chromosome 6p in families with
multiple occurrence of the disease. Am J Med Genet Neuropsychiatr Genet 67:560 –563.
Binder DA (1983): On the variances of asymptotically normal
estimators from complex surveys. Int Stat Rev 51:279 –292.
Braff DL, Heaton R, Kuck J, Cullum M, Moranville J, Grant I,
et al (1991): The generalized pattern of neuropsychological
deficits in outpatients with chronic schizophrenia with heterogeneous Wisconsin Card Sorting Test results. Arch Gen
Psychiatry 48:891– 898.
Cannon TD, Zorrilla LE, Shtasel D, Gur RE, Gur RC, Marco EJ,
et al (1994): Neuropsychological functioning in siblings
discordant for schizophrenia and healthy volunteers. Arch
Gen Psychiatry 51:651– 661.
Condray R, Steinhauer SR (1992): Schizotypal personality disorder in individuals with and without schizophrenic relatives:
Similarities and contrasts in neurocognitive and clinical
functioning. Schiz Res 7:33– 41.
Coon H, Plaetke R, Holik J, Hoff M, Myles-Worsley M, Waldo
M, et al (1993): Use of a neurophysiological trait in linkage
analysis of schizophrenia. Biol Psychiatry 34:277–289.
Cornblatt B, Erlenmeyer-Kimling L (1984): Early attentional
predictors of adolescent behavioral disturbances in children at
risk for schizophrenia. In: Watt NF, Anthony EJ, Wynne LC,
Rolf JE, editors. Children at Risk for Schizophrenia: A
Longitudinal Perspective. New York: Cambridge University
Press, 198 –211.
Cornblatt BA, Keilp JG (1994): Impaired attention, genetics, and
the pathophysiology of schizophrenia. Schiz Bull 20:31– 46.
Endicott J, Spitzer RL (1978): A diagnostic interview: The
Neuropsychologic Functioning in Relatives of Schizophrenic Patients
Schedule for Affective Disorders and Schizophrenia. Arch
Gen Psychiatry 35:837– 844.
Faraone SV, Kremen WS, Lyons MJ, Pepple JR, Seidman LJ,
Tsuang MT (1995a): Diagnostic accuracy and linkage analysis: How useful are schizophrenia spectrum phenotypes?
Am J Psychiatry 152:1286 –1290.
Faraone SV, Lyons MJ, Tsuang MT (1988): Schizophrenia:
Mathematical models of genetic transmission. In: Tsuang
MT, Simpson JC, editors. Nosology, Epidemiology and Genetics, Vol 3. Amsterdam: Elsevier Science, 501–530.
Faraone SV, Santangelo S (1992): Methods in genetic epidemiology. In: Fava M, Rosenbaum JF, editors. Research Designs
and Methods in Psychiatry. Amsterdam: Elsevier, 87–105.
Faraone SV, Seidman LJ, Kremen WS, Pepple JR, Lyons MJ,
Tsuang MT (1995b): Neuropsychological functioning among
the nonpsychotic relatives of schizophrenic patients: A diagnostic efficiency analysis. J Abnorm Psychol 104:286 –304.
Faraone SV, Seidman LJ, Kremen WS, Toomey R, Lyons MJ,
Tsuang MT (1996): Neuropsychological functioning among
the elderly nonpsychotic relatives of schizophrenic patients.
Schiz Res 21:27–31.
Faraone SV, Seidman LJ, Kremen WS, Toomey R, Pepple JR,
Tsuang MT (1999a): Neuropsychological functioning among
the nonpsychotic relatives of schizophrenic patients: A fouryear follow-up study. J Abnorm Psychol 108:176 –181.
Faraone SV, Tsuang D, Tsuang MT (1999b): Genetics of Mental
Disorders: A Guide for Students, Clinicians, and Researchers. New York: Guilford.
Faraone SV, Tsuang MT (1985): Quantitative models of the
genetic transmission of schizophrenia. Psychol Bull 98:41–
66.
Faraone SV, Tsuang MT (1995): Methods in psychiatric genetics. In: Tsuang MT, Tohen M, Zahner GEP, (editors).
Textbook in Psychiatric Epidemiology. New York: WileyLiss, 81–134.
Goldberg TE, Ragland D, Torrey EF, Gold JM, Bigelow LB,
Weinberger DR (1990): Neuropsychological assessment of
monozygotic twins discordant for schizophrenia. Arch Gen
Psychiatry 47:1066 –1072.
Goldberg TE, Torrey EF, Gold JM, Bigelow LB, Ragland RD,
Taylor E, et al (1997): Genetic risk of neuropsychological
impairment in schizophrenia: A study of monozygotic twins
discordant and concordant for the disorder. Schiz Res 17:77–
84.
Goldstein JM (1995): Gender and the familial transmission of
schizophrenia. In: Seeman MV, (editor). Gender and Psychopathology. Washington, DC: American Psychiatric Press,
201–226.
Gottesman II, McGuffin P, Farmer AE (1987): Clinical genetics
as clues to the “real” genetics of schizophrenia (a decade of
modest gains while playing for time). Schizophr Bull 13:23–
47.
Gottesman II (1991): Schizophrenia Genesis: The Origin of
Madness. New York: Freeman.
Gottesman II, Carey G, Hanson DR (1983): Pearls and perils in
epigenetic psychopathology. In: Guze SB, Earls FJ, Barrett
JE, editors. American Psychopathological Association Series.
Childhood psychopathology and development. New York:
Raven Press, 287–300.
BIOL PSYCHIATRY
2000;48:120 –126
125
Gottesman II, McGue M (1990): Mixed and mixed-up models for
the transmission of schizophrenia. In: Cicchetti D, Grove W,
editors. Thinking Clearly About Psychology: Essays in Honor
of Paul E. Meehl. Minneapolis: University of Minnesota
Press.
Heaton RK (1981): Wisconsin Card Sorting Test Manual.
Odessa, FL: Psychological Assessment Resources.
Heaton RK, Chelune GJ, Talley JL, Kay GG, Curtiss G (1993):
Wisconsin Card Sorting Test Manual–Revised. Odessa, FL:
Psychological Assessment Resources.
Huber PJ (1967): The behavior of maximum likelihood estimates
under non-standard conditions. Proc Fifth Berkeley Symposium on Mathematical Statistics and Probability 1:221–233.
Keefe RSE, Silverman JM, Roitman SEL, Harvey PD, Duncan
MA, Alroy D, et al (1994): Performance of nonpsychotic
relatives of schizophrenic patients on cognitive tests. Psychiatry Res 53:1–12.
Kish L, Frankel MR (1974): Inference from complex samples. J
Roy Stat Soc A:1–37.
Koren D, Seidman LJ, Harrison RH, Lyons MJ, Kremen WS,
Caplan B, et al (1998): Factor structure of the Wisconsin Card
Sorting Test: Dimensions of deficit in schizophrenia. Neuropsychology 12:289 –302.
Kremen WS, Goldstein JM, Seidman LJ, Toomey R, Lyons MJ,
Tsuang MT, et al (1997): Sex differences in neuropsychological function in nonpsychotic relatives of schizophrenic probands. Psychiatry Res 66:131–144.
Kremen WS, Seidman LJ, Pepple JR, Lyons MJ, Tsuang MT,
Faraone SV (1994): Neuropsychological risk indicators for
schizophrenia: A review of family studies. Schiz Bull 20:103–
119.
Lyons MJ, Toomey R, Seidman LJ, Kremen WS, Faraone SV,
Tsuang MT (1995): Verbal learning and memory in relatives
of schizophrenics: Preliminary findings. Biol Psychiatry 37:
750 –753.
McGue M, Gottesman II, Rao DC (1983): The transmission of
schizophrenia under a multifactorial threshold model. Am J
Hum Genet 35:1161–1178.
McGuffin P, Farmer A, Gottesman II (1987): Is there really a
split in schizophrenia? The genetic evidence. Br J Psychiatry
150:581–592.
Mirsky AF, Lochhead SJ, Jones BP, Kugelmass S, Walsh D,
Kendler KS (1992): On familial factors in the attentional
deficit in schizophrenia: A review and report of two new
subject samples. J Psychiatr Res 26:383– 403.
Pogue-Geile MF, Garrett AH, Brunke JJ, Hall JK (1991):
Neuropsychological impairments are increased in siblings of
schizophrenic patients. Schiz Res 4:390.
Pogue-Geile MF, Watson JR, Steinhauer SR, Goldstein G
(1989): Neuropsychological impairments among siblings of
schizophrenic probands. Schiz Res 2:70.
Scarone S, Abbruzzese M, Gambini O (1997): The Wisconsin
Card Sorting Test discriminates schizophrenic patients and
their siblings. Schiz Res 10:103–107.
Seidman LJ, Goldstein JM, Goodman JM, Koren D, Turner WM,
Faraone SV, et al (1997): Sex differences in olfactory
identification and Wisconsin Card Sorting performance in
schizophrenia: Relationship to attention and verbal ability.
Biol Psychiatry 42:104 –115.
126
BIOL PSYCHIATRY
2000;48:120 –126
Seidman LJ, Stone WS, Jones R, Harrison RH, Mirsky AF
(1998): Cognitive effects of schizophrenia and temporal
lobe epilepsy on memory. J Int Neuropsychol Soc 4:342–
352.
Spitzer RL, Williams JBW, Gibbon M (1987): Structured Clinical Interview for DSM-III-R Personality Disorders (SCIDII). New York: Biometrics Research Department, New York
State Psychiatric Institute.
Stangl D, Zimmerman M (1983): Structured Interview for the
DSM-III Personality Disorders, 2nd ed. Iowa City, IA: The
University of Iowa.
Stata Corporation (1999): Stata Reference Manual: Release 6.0.
College Station, TX: Stata Corporation.
Stratta P, Daneluzzo E, Mattei P, Bustini M, Casacchia M, Rossi
A (1997): No deficit in Wisconsin Card Sorting Test perfor-
S.V. Faraone et al
mance of schizophrenic patients’ first-degree relatives.
Schizophr Res 26:147–51.
Toomey R, Faraone SV, Seidman LJ, Kremen WS, Pepple JR,
Tsuang MT (1998): Association of vulnerability markers in
relatives of schizophrenic patients. Schiz Res 31:89 –98.
Tsuang MT, Faraone SV (1994): Epidemiology and behavioral
genetics of schizophrenia. In: Watson SJ, editor. Biology of
Schizophrenia and Affective Disease. New York: Raven
Press, 163–195.
Tsuang MT, Faraone SV, Lyons MJ (1993): Identification of the
phenotype in psychiatric genetics. Eur Arch Psychiatr Neurol
Sci 243:131–142.
Vincent KR, Castillo IM, Hauser RI, Zapata JA, Stuart HJ, Cohn
CK, et al (1984): MMPI-168 Codebook. Norwood, NJ: Ablex
Publishing Corporation.