University of Southern California Dissertations and Theses

Heritability of schizotypal traits during adolescence

(USC Thesis Other)

Heritability of schizotypal traits during adolescence

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

HERITABILITY OF SCHIZOTYPAL TRAITS DURING ADOLESCENCE

by

Marissa Ericson

A Thesis Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF ARTS
(PSYCHOLOGY)

December 2008

Copyright 2008 Marissa Leigh Ericson

ii

TABLE OF CONTENTS

List of Tables iii.

List of Figures iv.

Abstract v.

Chapter 1: Introduction 1
1.1 Schizotaxia, Schizotypy, Schizotypal Personality Disorder 1
1.2 Schizotypal Personality Traits 5
1.3 Measurement of Schizotypal personality 7
1.4 Genetics of schizotypy 9
1.5 Current study aims and hypotheses 12

Chapter 2: Methods
2.1 Participants 14
2.2 Measures 14
2.3 Statistical Method 17

Chapter 3: Results
3.1 Factor Analysis 27
3.2 Descriptive Statistics 30
3.3 Genetic Model Fitting 31

Chapter 4: Discussion
4.4 General Discussion 40
4.5 Strengths and limitations 43
4.6 Future Directions 46

References 47

iii

List of Tables

TABLE 1. Summary of twin studies of schizotypal traits 12
TABLE 2. Loadings from Exploratory Factor Analyses for three extracted 29
sub-factors cognitive-perceptual, interpersonal-affective, and disorganized.

TABLE 3.Means, standard deviations and number of participants 31
(N=individuals) for SPQ-C total score and three sub-factors,
age 11-13 years, by zygosity

TABLE 4. Intraclass (MZ/DZ on diagonal) and cross-trait twin correlations 32
(MZ below diagonal, DZ above diagonal) for SPQ-C and subfactors

TABLE 5. Univariate genetic model fit values and parameter estimates for 34
SPQ-C total score and subscales
TABLE 6. Multivariate genetic model fit values for SPQ subscales 37
TABLE 7. Breakdown of Common and Unique Effects 39

iv

List of Figures

FIGURE 1. Univariate biometrical (ACE) model. 21
FIGURE 2. Multivariate genetic Cholesky factor model for three 24
SPQ-C subscales

FIGURE 3. Independent pathway model for three SPQ-C subscales 25

FIGURE 4. Common Pathways model for three SPQ-C subscales 26
FIGURE 5. Scree plot for eigenvalues in factor analysis of SPQ-C items 28
FIGURE 6. Standardized path coefficients for best-fitting multivariate 38
AE Common Pathways model for SPQ-C subscales

v

Abstract

The study described in the present paper attempted to clarify further the genetic
and environmental etiology of schizotypal personality traits through biometric model
fitting of data from a sample of MZ (n=91 twin pairs) and DZ (n=87 twin pairs)
adolescent twins (age 11-13 years old) drawn from the general population. Univariate
genetic analyses found that schizotypal traits are modestly heritable (additive genetic
effects ranging from 35 to 49%). Multivariate genetic model fitting results indicated that
additive genetic and unique environmental influences acted through a single common
latent pathway for cognitive-perceptual, interpersonal-affective and disorganization
symptom dimensions of schizotypal personality. The covariation among the three
schizotypy sub-factors could be accounted for by a common ‘schizotypy’ latent factor
which was significantly heritable, with additive genetic factors explaining 60% of the
latent factor variance.

1

Chapter 1: Introduction

Schizophrenia affects approximately 1% of the general population and is a
complex, often catastrophic illness of neurodevelopmental origin (Cannon et al., 2003).
Although past research using twin and adoption data has predominantly concentrated on
relatives of persons with schizophrenia, this has proven to be time-consuming, expensive,
and unreliable, as it is widely known that such a sampling strategy excludes the majority
of persons who ultimately develop the disorder (Fusar-Poli et al., 2007; Gottesman &
Erlenmeyer, 2002; McDonald et al., 2001). New strategies are needed to accurately and
cost-effectively identify liable individuals prior to the manifestation of clinical
symptoms. This study of a non-selected sample of twins will add to the literature on
individuals at risk for schizophrenia spectrum disorders, such as schizotypal personality
disorder, by examining the genetic underpinnings of schizotypal personality traits during
adolescence.

1.1 Schizotaxia, Schizotypy, Schizotypal Personality Disorder

Schizophrenia is increasingly being regarded as a neurodevelopmental disorder.
It has been shown that children who later develop schizophrenia differ from controls in
cognitive, neuro-motor, interpersonal, and behavioral functioning many years before the
onset of the illness (Crow et al., 1995, Gruzelier and Kaiser 1996, Isohanni et al., 2000
and Isohanni et al., 2001). Therefore, focus must be redirected onto the emerging
psychophysiological, behavioral, and clinical signs early in development.

2
Schizotaxia is widely regarded as the genetic predisposition or the unexpressed
liability to schizophrenia spectrum disorders (SSDs) (Gottesman, 2003; Meehl, 1962,
1989). It has been proposed that schizotaxia—the inheritance of a central nervous system
integrative defect derived from a genetic aberration in neuronal synaptic selectivity—
ultimately results in either a moderate outcome [schizotypal personality structure,
including schizotypal personality disorder (SPD)], the most severe and debilitating
outcome (schizophrenia), or anywhere else along the spectrum (i.e. schizoid,
schizoaffective, etc) (Tsuang et al., 2003; Linney et al., 2003). While some researchers
(Faraone et al 1995, 2001; Tsuang et al 2000, 2002; Lencz et al 2005) agree with the
basic conceptualization of schizotaxia, the majority offers several collective variations:
(a) while the etiology has a genetic component, biological consequences of adverse
environmental factors (such as pregnancy, prenatal environment, postnatal environment,
delivery complications, etc) also play a factor; (b) schizotaxia reflects a multifactorial
polygenic etiology, and not just the expression of a single gene, (as Meehl believed).
Broadly speaking, this latent CNS defect is thought to have distinct effects on
physiological responses and neurological soft signs; in addition, it is postulated that this
defect directly affects the acquisition and activation of diverse psychological functions
(namely perceptual-cognitive, semantic, motivational, and affective (Meehl, 1989, 1990,
1993).
Conceptualized both as a latent personality organization that harbors the
vulnerability for SSDs (Meehl et al., 1962, 1989; Linney et al., 2003) and as a non-

3
clinical manifestation of the same underlying psychophysiological, cognitive and
neurobiological processes that predispose an individual to SSDs (Claridge et al., 1994;
Claridge and Beech, 1995), schizotypal personality (also labeled schizotypy) provides a
viable framework for the detection of core features of liability to schizophrenia spectrum
disorders prior to decomposition into illness (Lenzenweger, 2006; Linney et al., 2003). It
has been argued that schizotypal personality traits are qualitatively similar to the
characteristic symptoms of schizophrenia, albeit quantitatively less severe. Schizotypal
personality has been shown to have a highly analogous dimensional structure in the
general population, with latent factors resembling the classical schizophrenia positive,
negative, and disorganization symptom dimensions (Raine et al., 1994; Vollema and Van
den Bosch, 1995; Claridge et al., 1996). Evidence supports three symptom clusters: 1).
Cognitive-Perceptual deficits (comprised of Ideas of Reference, Magical Thinking,
Unusual Perceptual Experiences, and Paranoid Ideation), 2) Interpersonal-affective
deficits (Social Anxiety, No Close Friends, Blunted Affect, Paranoid Ideation), and 3)
Disorganized behavior (Odd Behavior and Odd Speech) (Mason et al., 2005; Raine,
1991).
As a part of the schizophrenia spectrum and a disorder which reflects a
constellation of cognitive-perceptual, and interpersonal disturbances, as well as
characteristic disorganized behavior and speech patterns, SPD is one of the potential
phenotypic expressions of the familial-genetic liability to schizophrenia (Raine et al.,
2006; Venables, 1961). SPD is widely regarded as an attenuated form of schizophrenia
that represents a prodromal stage of the disease (Raine 2006; Cadenhead, 2004, 2007).

4
While the typical onset of schizophrenia is in late adolescence or early adulthood
schizotypal symptoms can emerge as early as 10-13 years of age (Myles-Worsley et al.,
2004). Though arguably controversial, it has also been suggested that SPD can exist
during childhood (Meijer and Treffers, 1991; Wolff et al., 1991). One recent study of an
unselected, community-based sample of children aged 11-13 years, found that those who
report either probable experience of more than one psychotic-like symptom or definite
experience of at least one psychotic-like symptom (such as believing that their thoughts
can be read, that that they were sent special messages through the television, that they
were being followed or that they ever heard voices that other people could not hear) are
16 times more likely to have SSD diagnosis 15 years later than children who report no
psychotic-like symptoms (Laurens et al., 2007). Moreover, studies indicate an increase in
the rate of psychiatric symptoms and adjustment problems beginning in early adolescence
(Erlenmeyer-Kimling, 2001, Walker et al., 1996) and suggest that the initial
manifestation of poor adjustment generally follows the onset of puberty (Walker and
Bollini, 2002). As compared to the positive symptoms which can emerge during
childhood or early in development, one study found that the negative symptoms in
particular increase dramatically as an individual passes through adolescence (Galdos and
van Os, 1995).
Hallucinatory experiences and delusional beliefs reported at age 11 during a
structured diagnostic interview have been shown to be strong and specific indices of an
SSD diagnosis by age 26 (Poulton et al., 2000). In addition, in a more recent study
comparing the schizotypal dimensions, as measured by the Schizotypal Personality

5
Questionnaire (SPQ) factors, in different diagnostic outcomes (schizophrenic, bipolar,
depressive and OCD patients), Rossi et al., (2001) demonstrated that the SPQ-total,
cognitive, and interpersonal subscale scores were able to discriminate among these
clinical groups. Depressive patients reported lower scores on all subscales than the OCD,
bipolar, and schizophrenic patients and schizophrenic patients scored highest on all
subscales, including the disorganization symptom factor. Few studies have examined the
discriminant validity between self-report measures of SPD or schizotypal traits and a
heightened vulnerability for the development of psychotic disorders in general. While
some researchers (Chapman et al., 1994) theorize that the positive symptoms (magical
ideations, perceptual aberrations) are more specific and reliable indices of later
development of SSDs, others posit that it is the negative symptoms (interpersonal
difficulties, blunted affective responses, social isolation, physical and social anhedonia)
that more fully capture the fundamental liability to the spectrum. That is, some
researchers argue that schizotaxia can be operationalized based specifically on psychiatric
signs (mild negative symptoms) and neuropsychological performance (Claridge et al
1997; Meehl, 1990). Psychotic-like experiences or schizotypal personality traits in
concert with other neuropsychological and physiological dysfunctions are thus likely and
important precursor signs in identifying children at risk of later developing an SSD.

1.2 Schizotypal Personality Traits
In recent years, early clinical intervention in schizophrenia has become a principal
aim of mental health services, and further research on the early stage of the disease may

6
ultimately provide significant clues to the precise mechanisms which underpin
schizophrenia, as well as SSDs (Lenzenweger, 2006). Thus, it is important from both
clinical and research perspectives to identify and quantify a set of traits which represent
this liability state to SSDs (Tsuang et al., 2002).
The construct of schizotypy plays a significant role in clinical psychiatry and has
demonstrated considerable heterogeneity in definition and application. The early onset of
schizotypal symptoms and the fact that schizotypal individuals do not present with the
same confounds as schizophrenia patients (anti-psychotic medication, acutely psychotic,
hospitalization) make schizotypal personality organization a prime target for effective
prevention and treatment research.
Prospective investigations of the genetic basis of schizotypal personality are
needed to identify precursor signs to later development of an SSD, particularly
schizotypal personality disorder. Despite the fact that self-report screening has been
found to aid in the detection of individuals at risk for developing an SSD (Claridge, 1994,
1997; Vollema & van den Bosch, 1995), the psychometric measurement of schizotypal
personality during adolescence continues to receive little attention in the literature. The
liability to the schizophrenia spectrum is assumed to have a continuous distribution in the
population and it has been well established that it is possible to recognize and measure
‘psychotic-like’ traits in otherwise substantially high-functioning or healthy individuals
(Eysenck et al., 1985; Crow et al., 1995).
Schizotypal personality has been proposed to be a normal personality dimension
that is not necessarily pathological (Claridge et al., 1985). According to a dimensional

7
view of abnormality, a continuum of variation exists that describes this underlying
predisposition to the decompensation into psychosis (Claridge, 1987, 1995). Moreover,
this approach to schizophrenia spectrum symptomatology has demonstrated that
schizotypy can be characterized as the concurrent variation of unique but inter-correlated
symptom dimensions (Stefanis et al, 2002), such as the positive, negative, and
disorganization factors (Liddle et al., 1987 and Peralta et al., 1992).
Overall, the schizotypy approach aims to identify normal individuals with
schizophrenia latent liability, which essentially increases their vulnerability to develop
the illness (Claridge, 1987; Meehl, 1989). The validity of this approach is suggested by
studies demonstrating an increase in liability for categorically defined SPD and
schizotypal personality in the relatives of patients with schizophrenia (Kendler et al.,
1985), as well as elevated probability of psychotic episodes in individuals who present
with high levels of schizotypy or psychosis-proneness (Chapman et al., 1994).
Schizotypal personality is widely considered an endophenotype for schizophrenia, in that
it shares some neurobiological, psychophysiological, and neurobehavioral characteristics
associated with disorders of the schizophrenia spectrum (Lenzenweger, 2006; Linney et
al., 2003; Gottesman et al., 2003)

1.3 Measurement of schizotypal personality
Though the past decade has witnessed increasing interest in the construction of
self-report measures for schizotypal traits among the general population, there is
currently no screening instrument for schizotypy during pre-pubertal development. One

8
of the most widely used methods for assessing schizotypy has been self-report
questionnaires. While there are a number of scales designed to assess psychosis-
proneness in general, such as the Physical and Social Anhedonia scales (Chapman,
Chapman, and Raulin, 1976), the STA scale (Claridge & Broks, 1984), the Magical
Ideation scale (Eckblad & Chapman, 1983) and the Perceptual Aberration scale (Claridge
& Broks, 1984), these scales do not assess the nine features of schizotypal personality
disorder which are outlined in the DSM-IV (APA, 1984). The Schizotypal Personality
Questionnaire (SPQ) was the first instrument designed as a screening instrument
specifically for this purpose (Raine, 1991).
Factor analytic studies have reported positive, negative,

and disorganization
factors in schizotypy that resemble but represent attenuated forms of classical
schizophrenia symptom dimensions (Mason et al., 2005; Raine, 1991). Broadly
speaking, theses studies have most frequently supported two or three components,
depending on the number and item content of scales included in the analyses (Vollema
and Postma, 2002; Claridge et al., 1996). These symptom dimensions are thought to
mirror attenuated versions of the positive, negative, and disorganized symptoms in DSM-
IV- defined schizophrenia (Mason et al., 2005; Raine, 1991). In two separate factor
analytic studies Mata et al., found a three-factor model fitted the data better than single or
two-factor models and that the three schizotypal factors appear to parallel the three
analogous factors that have been reported for schizophrenic symptomatology (Mata et al.,
2003, 2000). In addition, Rossi et al., (2002) demonstrated that a three-factor structure
solution comprised of cognitive-perceptual, interpersonal, and disorganization

9
dimensions underlies the differences in both schizophrenic and normal samples using the
SPQ. Rossi and colleagues reported this to be the first known study in which a
disorganized three-factor model of DSM-IV schizotypy was generalizable to a group of
schizophrenic patients (2002).

1.4 Genetics of schizotypy
Twin and adoption studies have reported strong genetic influences in liability for
development of an SSD and path analytic polygenic models estimate the overall genetic
contribution to be 60-80% (Cannon et al., 1998, 2005). It has been estimated that first
degree relatives of schizophrenic individuals are at ten times greater risk for development
of schizophrenia or an SSD, suggesting a substantial genetic component to the illness
(Turetsky, 2007). Moreover, proband concordance rates for schizophrenia range from
(40-55%) in monozygotic (MZ) twins, compared to (15-20%) in dizygotic (DZ) twins
(Cannon et al., 2003; Gottesman and Shields, 1982).
While researchers estimate that as much as 10% of the general population is
genetically liable to schizophrenia (Meehl, 1990; MacDonald et al., 2001), it is
acknowledged that the majority of this projected percentage may never in fact manifest
clinical schizophrenia. In approximately 40% of MZ twins in which one twin is
diagnosed with schizophrenia, the co-twin never meets diagnostic criteria for illness.
This discordance rate highlights the fact that environmental factors clearly also play a
critical role in the etiology of the disorder. Moreover, this discordance rate also brings to
light the possibility that some individuals are sub-threshold; while these individuals do

10
not meet criteria for clinical diagnosis, they may in fact be carriers of schizotaxia and
could have had unique experiences which buffered them from decompensation into
psychosis (Meehl, 1990). Theoretically, these sub-clinical individuals are thus more
likely to have schizotypal traits and be diagnosed with schizotypal personality disorder.
In recent years, twin investigations of the genetic basis of schizotypal personality
and the construct of schizotypy in adulthood have reported inconsistent results. While
some have found various components of schizotypal personality to be influenced by
additive genetic effects and to be moderately to significantly heritable, others have
reported only marginal or weak heritability (Linney et al., 2003). Studying twins from
the general population, Kendler & Hewitt (1992) found that the variance in most self-
report schizotypy scales (except for perceptual aberration,) involved substantial genetic
influences. Moreover, although Kendler & Hewitt (1992) and MacDonald (2001) found
common environmental influences on specific positive symptoms, Miller (1993) found
these influences on only the negative dimension of schizotypy. Vollema et al., (2002)
found that scores solely on the positive symptom dimension of a schizotypy questionnaire
administered to relatives of patients with psychotic disorders related most strongly to
their genetic risk of psychosis. MacDonald et al., (2001) reported one latent common
schizotypy factor (mainly explained by positive symptoms (perceptual aberration,
magical ideation, etc)), which was influenced by genetic (15%), shared-environmental
(45%) and unique environmental effects (40%). In another study of female twins by
Linney et al., (2003), two independent latent factors were uncovered, such that one factor
was comprised of the positive symptoms (cognitive disorganization, unusual experiences,

11
and delusional ideation) and that second was essentially a negative dimension (anhedonia
and social anxiety). The results from this study suggested that two distinct etiological
mechanisms exist for the various symptom dimensions of schizotypal personality.
However, this finding has yet to be replicated in an adult, adolescent, or child sample.
TABLE 1 briefly summarizes the twin and family studies that have been conducted to
date on schizotypal personality and its components. Taken together, these findings
provide evidence that both the positive and negative dimensions of schizotypy are
influenced by genetic effects in community-based samples and families unselected for
psychiatric disease.

12

TABLE 1. Twin studies of schizotypy and psychosis-proneness in non-clinical,
unselected samples
Study N
Age Range
Scales Used
Heritability
estimates
Clardige &
Hewitt
(1987)
108 MZ/ 102
DZ
Not reported STA h
2
approx
53%
Kendler et
al., (1991)
8 MZ/DZ
Pilot study
18-50 years
(Mean=
MZ: 33.4/
DZ: 14.3)
STA; Structured
Interview for
Schizotypy
h
2
> 60%
Kendler &
Hewitt
(1992)
70 MZ/ 63 DZ 18-50 years SA, PA, NonCon, PA,
STA- ParID + UE; P-
scale
Univariate:
h
2
range
29%-67%
McDonald et
al (2001)
98 MZ/59 DZ
(same sex
only)
18-27 years
(not reported)
PA, MI, SA; RISC h
2
range
27%-41%
Hay et al.,
2001
614 MZ/ 720
DZ
18-25 years
(not reported)
UE, SA h
2
range
28%- 50%
Linney et al.,
2003
148 MZ/ 338
DZ
17-84 years
(Mean= 47) O-LIFE & PDI
h
2
approx
50%; 37%
for PDI

Jang et al.,
2005
102 MZ/90 DZ
32-38 years

O-LIFE
h
2
range
48%-62%
*STA (Claridge & Broks, 1984); Social Anhedonia (SA) (Mishlove & Chapman, 1985); Physical
Anhedonia (PA); Rust Inventory of Schizotypal Cognitions (Rust, 1987) Oxford Liverpool
Inventory of Feelings and Experiences (O-LIFE; Mason et al., 1995); Peters Delusions Inventory
(PDI; Peters et al., 1996, 2003). Perceptual Aberration Scale (PAS; Chapman, Chapman, &
Raulin, 1978); NonCon ParID + UE (Mishlove & Chapman, 1985); Psychoticism Scale (P-scale;
Eysenck, H.J., Eysenck, S.B.G. & Barrett, P. (1985))

1.5 Current Study Aims and Hypotheses
Overall, the present study attempts to clarify further the genetic etiology of
schizotypy through biometric model fitting of data from a sample of MZ and DZ
adolescent twins drawn from the general population. Maximum likelihood model-fitting
approaches were applied to address the following questions: 1) What is the nature of the
phenotypic factor structure of questionnaire-defined schizotypal personality during
adolescence? 2) What are the relative contributions of genetic and environmental effects

13
to psychometrically measured schizotypal traits during adolescence? 3) Is there a
common latent factor that accounts for the genetic covariation among the three symptom
dimensions of schizotypal traits?

14

Chapter 2: Method

2.1 Participants
This study was conducted in the context of the Southern California Twin Project,
an ongoing longitudinal twin study assessing the risk factors for aggressive and antisocial
behaviors (see Baker, Barton, et al. 2006; Baker, Jacobson et al. 2007), for a detailed
description). The total sample included 605 sets of twins and triplets (n=1219
adolescents) and their primary caregivers, with assessments completed to date on two
different occasions (Waves 1 and 2) and two more assessments currently under way
(Waves 3 and 4). Twins were recruited from local schools and advertisements in the Los
Angeles community. The sample includes both male and female MZ and DZ pairs, and
opposite-sex DZ pairs and the ethnic composition of the entire sample is representative of
Southern California: 37% Hispanic, 27% Caucasian, 14% Black, 4% Asian, and the
remaining 17% of mixed/other (Baker et al., 2006).

2.2 Measures
The participating twins were asked to complete questionnaires, including the
SPQ-C. Their caregivers completed demographic surveys, which included information
about the ethnic and socioeconomic background of the twins, and an additional survey
about the similarity of the twins, which was used for purposes of zygosity determination.
Although additional and extensive data are available for these subjects, only the SPQ-C,
demographic, and twin similarity questionnaires were used in the present study.

15
Zygosity determination
Through DNA microsatellite analysis, which specifies that seven or more
concordant DNA markers constitutes an MZ status and one or more discordant markers
constitutes a DZ status, zygosity of the same-sex twin pairs was determined for the
majority of pairs (398/458 = 87%). For the remaining 60 twin pairs for whom adequate
DNA samples or results were not available, the Twin Similarity Questionnaire (Lykken,
1978) was employed to assign zygosity. For those cases in which both questionnaire and
DNA results were available, there was 90% agreement between the two.

Schizotypy Measures
In order to effectively fill a gap in the literature regarding the measurement of
schizotypal traits during childhood and adolescence and in community samples, the child
version of the Schizotypal Personality Questionnaire, entitled the ‘SPQ-C’ was devised
by Dr. Adrian Raine, in collaboration with Dr. Laura Baker and the Southern California
Twin Project. The SPQ-C, a self report questionnaire adapted from the Schizotypal
Personality Questionnaire- Brief (SPQ-B) scale developed by Raine and Benishav
(1995), was tailored to assess presence of traits found in DSM-IV schizotypal personality
disorder in community, adolescent samples. Consisting of 35 statements in a
dichotomous response format, this new instrument was based on questions from the brief
scale, which were in turn selected from the longer 74-item Schizotypal Personality
Questionnaire (Raine, 1991). Questions that appeared most reliable and valid from the
full SPQ were chosen during the development of the SPQ-B; subsequently, these same

16
questions were chosen during the construction of the SPQ-C. Some items were modified
to be easier understood by children. 20 of the original items on the brief SPQ-B scale for
adults remained intact on the SPQ-C child version, and 15 were either extracted from
other questions for clarity or were completely new items.
Key goals of the SPQ-B were to produce an instrument which (1) was short, (2)
highly correlated with longer self-report and clinical assessment measures, (3) assessed
the three main factors of schizotypal personality (4) had internal reliability, (5) showed
test-retest reliability, and (6) evidenced criterion validity by significant relationships with
clinical measures of SPD (Raine and Benishay, 1995). In turn, the SPQ-C should meet
these same criteria, although they have yet to be established.
For the SPQ-B, both exploratory and confirmatory factor analysis uncovered three
factors: Cognitive–Perceptual Deficits (likened to the positive symptoms of
schizophrenia), Interpersonal-affective Deficits (negative symptoms), and Disorganized
(Raine, 1994). The Cognitive–Perceptual Deficits factor captures constructs of referential
ideation, odd belief or magical thinking, unusual perceptual experience, and paranoid
ideation. The Interpersonal-affective deficits factor captures constructs of excessive
social anxiety, lack of close friends, inappropriate or constricted affect, and paranoid
ideation. Lastly, the Disorganization factor captures constructs of odd behavior and
speech patterns (Raine, 1991). In a study of four independent subject samples of
undergraduates, the three factors and total score from the SPQ-B were found to have
internal reliabilities ranging from .72 to .80, correlations with the full 74-item SPQ
ranging from .89 to .94, scale scores correlated with independent clinical ratings of

17
DSM-III R schizotypal traits (average r=.62), and test–retest reliabilities across a 2 month
interval between .86 to .95 (Axelrod et al., 2001; Raine and Benishay, 1995). In addition,
Confirmatory factor analysis has repeatedly demonstrated the three factor solution in
samples from many countries, such as England (Gruzelier et al. 1995; Gruzelier, 1996),
Taiwan (Chen et al. 1997), Mauritius (Reynolds et al. 2000), and France (Doumas et al.
2000). It has been replicated in at least fourteen independent samples (Bedwell et al.,
2005; Reynolds et al. 2000). Vollema and Hoijtink (2000) uncovered the same three-
factor structure in both schizophrenic inpatients and outpatients using analysis of
individual items. In this study, the SPQ-C total score was computed as the summation of
the three sub-factors and included all items.

2.3 Statistical Methods
Descriptive statistics
All raw data were examined through Statistical Package for the Social Sciences
(SPSS) EXPLORE for accuracy of data entry, missing values, and fit between their
distributions and the assumptions of multivariate analysis. Although there were
incomplete data within twin pairs, all available data was used for model fitting. Missing
data appeared to follow no consistent pattern and all raw data were used. To improve
pairwise linearity and to reduce extreme skewness and kurtosis, SPQ-total was rank
transformed and normal scores were used for all analyses.

18
Exploratory and Confirmatory Factor Analysis
In order to examine the latent structure of schizotypal dimensions, exploratory
factor analysis with varimax oblique rotation was first performed on SPQ-C items for
data collected from twins during Wave 2 of data collection (N=356), using SPSS
statistical analysis software. Taking into account the correlated nature of twins within a
twin pair, Twin A was used for exploratory analyses and Twin B was used for the
confirmatory analyses. As it was suspected that the factors would be inter-correlated,
principal component analysis with Promax (oblique) rotation was then employed. The
scree plot of eigenvalues was used to determine the number of factors to extract; this
method is often used to give a best first estimate of the underlying factor solution and is
widely supported in the literature (Rawlings et al., 2001).
Confirmatory factor analysis of the SPQ-C items was then performed using EQS
structural equation modeling software (Bentler et al, 1995). Few studies have investigated
the factor structure of schizotypy during childhood or adolescence. The adult literature
widely supports a three-factor solution (Mason et al., 2005; Vollema and Postma, 2002;
Clardidge et al., 1996; Raine et al., 1991), though some champion a two-factor model
(Venables, 2002; Kelley et al., 2002). Thus, both two and three-factor solutions were
examined in this sample. Consistent with the lion’s share of the literature, it was posited
that items would load onto a cognitive perceptual factor, an interpersonal-affective factor,
and a disorganization factor. Goodness of fit was determined by the Goodness of Fit
Index (GFI), which is often cited as the index of choice (Bentler et al., 1995). A value of
.9 or greater is suggested to be indicative of a good fit.

19
Genetic Analyses
The twin design was used to investigate the relative contributions of genetic and
environmental components of variance on the SPQ-C. Monozygotic (MZ) pairs share
100% of their genes and dizygotic (DZ) twin pairs share only 50% of their segregating
genes on average, and the effects of heredity and environment can be assessed through
comparison of MZ and DZ twin correlations. Evidence for genetic influence is suggested
when MZ twins show greater resemblance than DZ twins. Heritability is estimated as the
proportion of total phenotypic variation of a given trait that is found to be genetically
influenced, and is calculated as the squared value of the standardized genetic parameter
estimate (h
2
). Univariate and multivariate behavioral genetic models must make certain
assumptions about the nature of the processes being estimated. For example, twin models
must assume that random mating occurs in the parent generation. Moreover, it is assumed
that there an equal environment for MZ and DZ twins, and that there is not an interaction
between genes and environment (Plomin et al., 2001).
The classical twin design is one of the most powerful methods for assessing the
genetic and environmental influences on a given phenotype. As a way to measure the
similarity between twins on a given trait or behavior, intraclass correlations also provide
the proportion of variance in an observable trait attributable to between pair (or between
family) differences. In addition, the cross-twin cross-trait correlations provide a first
indication of the potential covariation of genetic and environmental influences on a set of
measured traits (Neale & Cardon, 1992).

20
Biometrical Model Fitting
Structural equation modeling, a method in which parameters are estimated
between latent (unobserved) and observed variables, was employed to investigate
estimates of the proportions of variance accounted for by genetic (A), shared
environmental (C), and unique environmental effects (E). Structural equation models are
often presented in path diagrams, such that the latent factors (A, C, E) are depicted in
circles and observed variables are depicted using squares (see Figure 1). The genetic
correlation (r
a
) is set to 1.0 for MZ twins, as they are genetically identical, and .5 for DZ
as they on average share 50% of their segregating genes. Shared environmental factors
refer to the experiences that twins have in common, such as shared family environment,
that contribute to the similarity within pairs of twins. Shared environmental contributions
are assumed by the model to contribute equally to the similarity in MZ and DZ pairs (i.e
the equal environment assumption (EEA)). Thus, the shared environment correlation (r
c
)
is set to 1.0 for both MZ and DZ twins in the model. Non-shared environmental
contributions are those factors or experiences that make twins or siblings dissimilar
(unique peer group or specific relationship with a family member, etc.). As there is no
correlation for the unique environment, this parameter includes measurement error. The
influence of A, C, and E on the observed variation in a given trait is denoted by
parameters a, c, and e; squaring the loadings of these parameters yields the variance
explained by each component (Neale & Cardon, 1992).

21
FIGURE 1. Univariate biometrical model.
a
A E C
Phenotype
Twin 1
A E C
Phenotype
Twin 2
ac
e
c e
ra = MZ = 1.0 / DZ = 0.5 rc = MZ=1.0 (1.0) / DZ=1.0 (0.25)

Note: A=additive genetic,C=common environment, E=non-shared environment;
a, c, e: factor loadings

Based on raw data for SPQ-C scores for each twin pair, structural equation
modeling was used in Mx genetic modeling software (Neale, 1993) to estimate the
contributions to individual differences in each variable from additive genetic effects,
shared environment, and unique environment. Raw maximum-likelihood estimation
procedures were employed, which yielded fit values (-2 times the log-likelihood for the
data given the model, or -2LL) for each model fit.
The most saturated model (Model 0) freely estimated the variances and
covariances between twins within each zygosity group, with both means and variances
allowed to differ across the two twins and across zygosity groups. This was followed by
another model (Model 0a) in which means were constrained to be equal across co-twins
and zygosity groups, and this model served as a basis for comparison of all subsequent
models. More highly constrained models were then fit to the data for each variable, in

22
which the A, C, and E components were estimated. These subsequent models were then
compared to Model 0a. The difference between -2LL for two nested models (log-
likelihood-ratio test statistic (LRT; Neale & Cardon, 1992)) follows a chi-square ( χ
2
)
distribution, with degrees of freedom (df) equal to the difference in df for the two models.
This χ
2
is one way to assess the overall fit of the model in comparison to the full,
saturated model. A non-significant χ
2
is interpreted in such a way that the model provides
not a statistically significant worse fit to the data; on the other hand, a significant χ
2
is
indicative of poor fit because it fits the data better than the comparison model. By looking
at the χ
2
relative to the degrees of freedom (df), one can delve further into testing nested
models to investigate that which is considered the most parsimonious.
The ∆ χ
2
, or the change in chi-square, the df, and the p-value are then used to
evaluate whether further constraints on a model, such as dropping a parameter, still fit the
data. Calculated as the LRT minus 2x the difference in df, the Akaike’s Information
Criteria (AIC) provides both a measure of goodness of fit and parsimony. The AIC is a
widely used index of fit, such that the lower the AIC, the better the balance between
goodness of fit and parsimony. The AIC was evaluated in all models according to this
rationale in order to identify the best fitting model. Furthermore, through this comparison
model testing, the significance of the individual parameters can be assessed. Univariate
models were first fit to the raw data, including the SPQ-C total score and each subscale,
in order to get a first glimpse at the underlying A, C, and E components in each score.
Multivariate genetic models were then fit to the raw SPQ-C subscales using Mx
software (Neale et al., 2003) to explore the genetic and environmental basis of

23
schizotypal traits during adolescence. Multivariate twin models can be employed to
explore genetic and environmental basis of the covariation among measures. To
investigate how genetic and environmental factors influence the development of
schizotypal traits (study question #2), and to explore the covariation between the three
sub-factors (study question #3), three multivariate twin models were fit to the data: a
Cholesky decomposition factor model, an independent pathway model, and a one-factor
Common Pathways model.
In a Cholesky factor decomposition of the genetic and environmental influences
among the three measured variables (FIGURE 2), additive genetic effects (A), shared
environmental effects (C), and non-shared environmental effects (E) are partitioned into
three sets of factors. While A1, C1 and E1 influence all three variables, A2, C2 and E2
influence only the second and third variables, and A3, C3 and E3 influence the third
variable only. A Cholesky decomposition is a full model to which other models are
compared, because variance-covariance parameter estimates are guaranteed to be
positive, semi-definite. As the Cholesky offers the fullest potential explanation of the
data, more constrained models are thus compared to it.

24

FIGURE 2 Multivariate genetic Cholesky factor model for three SPQ-C subscales
A1
P1
A2
P2
A3
P3
C1 C2 C3
E1 E2 E3
a11 a21 a31
a22 a32
a33

Note: A=additive genetic, C=common environment, E=non-shared environment; a, c, e:
factor loadings; P=phenotype

The independent pathway model (FIGURE 3) is a more highly constrained model, as
compared to a Cholesky decomposition. In addition to estimating the general genetic (A),
shared environment (C), and non-shared environmental (E) effects that load directly onto
each measured variable, this model also measures the specific or independent genetic and
environmental influences on each individual observed phenotype.

25

FIGURE 3. Independent pathway model for three SPQ-C subscales

Cognitive-
perceptual
c a e
A C E
Interpersonal-
Affective
c a e
Disorganization
c
a
e

Note: Common parameters: A=additive genetic, C=common environment, E=non-shared
environment; Specific effects: a, c, e:

In a Common Pathways model, common genetic and environmental factors influence all
observed variables via a single latent liability, or underlying psychometric factor. This is
considered a more stringent model than the independent pathway model, in that it
specifies that a common latent phenotypic factor is determined by the covariation among
the scales. Instead of considering the effects as direct loadings on the measured variables,
the common parameters are mediated through an underlying factor that encompasses the
variance shared among the variables. In addition to the genetic and environmental effects
of the latent factor, parameters specific to each measure (a, c, e) are also estimated.

26

FIGURE 4. Common Pathways model for three SPQ-C subscales

A E
A E
C
C A E A E C C
Cognitive-
perceptual
Disorganization

Interpersonal-
Affective
Schizotypy
Note: Common parameters: A=additive genetic, C=common environment, E=non-shared
environment; Unique factor loadings: a, c, e

27
Chapter 3: Results
3.1 Factor Analysis
In order to examine the latent structure of schizotypal dimensions, exploratory
factor analyses with both varimax and promax oblique rotations were performed on
SPQ-C items for twins during Wave 2 of data collection (N=356), using SPSS. It was
suspected that the factors would be inter-correlated and Principal Component analysis
with Promax (oblique) rotation was conducted on 35 items. The scree plot of eigenvalues
was used to determine the number of factors to extract. An elbow was detected in the plot
at the third factor, and thus a three-factor solution with varimax rotation and promax was
tested. The PCA yielded three large components that accounted for 67.64% of the total
variance. Factor loadings of the SPQ-C items from the three-factor solution with promax
rotation are shown in TABLE 2. The criterion cutoff for significance was set at +/- 0.3
for the item loadings. Each item, with the exception of two, loaded most highly on one
distinct factor, using this criterion. For instance, ‘I sometime hear or see things that aren’t
there’ had the highest loading of .75 on factor 1- cognitive and perceptual aberrations. On
the other hand, though the question ‘I don’t say much’ loaded on both factors 2
(disorganization) with -.37 and factor 3 (interpersonal-affective deficits) with .55, the
loading was higher on factor 3. Thus, it was considered in the summation of the
interpersonal-affective factor.
Confirmatory factor analysis was then performed using EQS Structural Equation
Modeling Software (Bentler, 1995). Few studies have investigated the factor structure of
schizotypy during childhood and adolescence. A three-factor model fitted the data

28
according to the goodness of fit index (GFI=.92). Though a two-factor model is
supported in the literature (often considered as a positive schizotypy and negative
schizotypy uncorrelated factors), a two-factor model did not fit the data well according to
the fit criterion (GFI=.67). As the GFI results yielded a 3-factor solution that is consistent
with what has been found in adult samples, the items that loaded onto each factor were
then computed using the factor loadings to create 3 subscales: Cognitive-Perceptual,
Interpersonal-affective, and Disorganization. Descriptive and genetic analyses were then
completed on these subscales as well as for the total score.

FIGURE 5. Scree plot for eigenvalues in factor analysis of SPQ-C items
Component Number
35 34 33 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1
Eigenvalue
10
8
6
4
2
0
Scree Plot

29
TABLE 2 Loadings from Exploratory Factor Analyses for three extracted subfactors:
cognitive-perceptual, interpersonal-affective, and disorganized

PROMAX Rotated factors
Factor 1: Factor 2: Factor 3:
I sometimes hear or see things that aren't there. .750 .056 -.179
I sometimes feel there is a person or spirit around me, even
though there is no one there.
.665 -.044 .009
I sometimes feel distracted by far-off sounds that I'm not
normally aware of.
.633 .085 -.070
I often get the feeling that other people are watching me when
I am out playing or shopping.
.588 .107 .062
I feel I have to be on my guard even with friends. .579 .020 .074
I have had experiences like seeing flying saucers, or knowing
something will happen before it does.
.575 .109 -.306
I often have to stop people from taking advantage of me. .563 -.075 .130
I find it best not to let other people know too much about me. .516 .115 .159
I often think people are talking about me. .502 .086 .121
I often think people are out to get me. .394 .109 .195
I sometimes think that common events or objects have a
special meaning for just me.
.391 .255 -.046
I often feel anxious or nervous. .314 -.088 .244
I sometimes think that people can read my mind. .302 .162 -.051
I am an odd, unusual person. -.088 .671 .181
I am a strange person. -.109 .663 .151
I act a bit crazy at times. .087 .635 -.126
I sometimes ramble on when I am talking. .004 .624 -.013
I talk a lot. -.028 .605 -.248
I sometimes act oddly. .242 .577 -.070
I have some strange ideas about things. .206 .555 -.042
I sometimes use words in unusual ways. .250 .541 -.070
I find it hard getting people to understand what I am saying. .228 .323 .250
I dress a lot differently than others. -.011 .316 .309
I don't have many friends. -.217 .299 .624
I find it hard to make close friends. -.054 .114 .581
I don't say much. .243 -.370 .554
I often keep in the background on social situations. .217 -.028 .504
I tend to keep my feelings to myself. .188 -.153 .485
I am fairly quiet and shy. -.004 -.174 .457
I feel very uncomfortable in social situations involving people
I do not know.
.153 -.044 .465
I feel very uneasy talking to people I do not know well. 280 -.007 .439.
I don't laugh or smile much. .151 -.164 .388
I get easily hurt by what people say. .153 .121 .312
I often pick up on hidden threats or put-downs from what
people say or do.
.174 .195 .310
I am a bit unfriendly and distant. .043 .197 .308

30

3. 2 Descriptive Statistics
Means, variances, and twin correlations
Descriptive statistics by zygosity and sex for SPQ-C total and subscales are
presented in TABLE 3. SPQ-C total score was rank transformed and normal scores were
used for all analyses. Due to sample size, all subsequent analyses utilized combined
males and females into two groups comprised of (1) all male and female MZ pairs (N =91
pairs); and (2) all male, female, and opposite sex DZ pairs (N=87 pairs). Overall, there
were no significant differences among means (t=.63, p=.53) across the sexes for any of
the SPQ subscales or for the SPQ total score. Means and variances of individual
schizotypy measures were then recalculated for MZ-combined and DZ-combined groups,
as reported in TABLE 3. Using Mx, mean and standard deviation differences between
zygosity groups were also investigated. Within these combined groups there were also no
significant differences uncovered between MZ and DZ twins for mean scores or standard
deviations (t=.99, p=.32) on the SPQ total score, Cognitive-Perceptual factor,
Interpersonal-Affective factor, and Disorganization factor. According to criteria set out
by the authors of the original SPQ scale, Dr. Raine and colleagues, it is anticipated that
subjects who will meet DSM-IV criteria for SPD will fall in the top 10% of the
distribution of scores (a score of 30 or higher) (Raine et al., 1995). Following this criteria,
out of the 356 total subjects, 6 subjects were found to be in clinical range. On the
cognitive-perceptual subscale, 9.5% of subjects endorsed 7 or more (out of 9 items.)
9.2% of subjects endorsed 11 or more (out of 15 items) on the interpersonal-affective

31
subscale and 13% of subjects reported 7 or more items (of 13 items) on the
disorganization subscale.
TABLE 3 Means, standard deviations and number of participants (N=individuals)
for SPQ-C total score and three sub-factors, age 11-13 years, by zygosity

MZ DZ
Means (Standard Deviations)
SPQ-C Total 10.35 (7.26)
N=182
9.60 (6.93)
N=174
Cognitive-Perceptual
Factor

2.50 (2.52)
N=182
2.46 (2.47)
N=174
Interpersonal-
Affective
Factor
4.78 (3.28)
N=182
4.20 (3.12)
N=174
Disorganized
Factor

3.06 (2.67)
N=182
2.95 (2.71)
N=174

Note: MZ = monozygotic; DZ = dizygotic; N = number of participants

3.3 Genetic Model Fitting
Intraclass correlations and univariate model-fitting
Twin models were based on pair-wise raw SPQ-C scores and included all
available data. All subscales were found to significantly correlate with each other;
correlations between factors were: 1) cognitive-perceptual and interpersonal-affective
(.62, p<.05); 2) cognitive-perceptual and disorganization (.62, p<.05) and,
3) interpersonal-affective and disorganization (.49, p<.05). MZ pairs were more highly
correlated than DZ pairs for all variables indicating genetic influences. Intraclass
correlations were significant for each subscale and the total SPQ within each zygosity

32
group, except for the interpersonal-affective DZ correlation and are provided in
TABLE 4.
TABLE 4. Intraclass (MZ/DZ on diagonal) and cross-trait twin correlations (MZ below
diagonal, DZ above diagonal) for SPQ-C and subfactors

Cognitive-
perceptual
Interpersonal-
affective
Disorganized SPQ-C
total
Cognitive-
perceptual
.48*/.32* .12 . 27* ----
Interpersonal-
affective
.37* .37*/.14 .08 ----
Disorganized .31* .22* .46*/.29* ----
SPQ-C total ---- ---- ---- .45*/.28*
Note: *(p<.05)

Univariate genetic model-fitting results for the SPQ-C total score, in addition to the sub-
factor scores are presented in TABLE 5. TABLE 5 provides fit indices and sequential
∆ χ
2
tests for each univariate nested model, as well as the genetic, shared environmental,
and individual specific estimates and corresponding confidence intervals for the full and
best fitting models. First, univariate analyses were completed for the SPQ-C total score
and then for each sub-scale. A fully saturated model in which the means were
unconstrained was tested first, and then a saturated model with means constrained was fit
to the data. This constrained model provided a more parsimonious fit to the data as
indicated by the AIC and χ
2
statistic (p=0.51). Thus, this model was considered the
baseline model for comparisons with subsequent models. Next, a 2 group ACE model (1),
was found to fit the data ( χ
2
= 1.65, df= 3, p=0.65). As this full ACE model was
parsimonious as indicated by the AIC and ∆ χ
2
statistic, this became the model to which
sub-models were compared. Next, the common environment parameter (C) was dropped

33
from the model. This AE means constrained model (2a) provided a significantly better fit
compared to model 1 ( ∆ χ
2
= 0.45, df=1, p=0.50). Lastly, in order to evaluate the power of
this sample to reject alternative hypotheses, models that dropped the genetic path, and
both the genetic and shared environmental paths were fit to the data. Both the CE and E
models (2b and 2c, respectively) fit poorly according to the fit statistics. The best fitting
model included genetic and non-shared environment. Based on the parameter estimates
from the best fitting model for SPQ-C total score, 45% of the variation in schizotypal
traits was attributable to additive genetic factors, and 55% was due to unique
environment and measurement error. These aforementioned steps were repeated for each
sub-factor (TABLE 5). For all subscales, an AE model fit the data best, according to the
AIC, ∆ χ
2
and corresponding p-value.

TABLE 5 Univariate genetic model fit values parameter estimates for SPQ-C total score and subscales
34
Overall Fit Change in Fit Parameter Estimates
Model -2LL(df) AIC χ
2
(df)
p
∆ χ
2
( ∆df)
p

Comparison A C E
SPQ Total
0. Saturated
(means free)
2373.96
(346)
1681.96 ---- ---- ---- --- --- ---
0a. Saturated
(means
constrained)
2376.26
(349)
1678.26 2.30 (3)
0.51
---- ---- --- --- ---
1. ACE 2377.90
(352)
1673.90 1.65 (3)
0.65
---- ---- 0.30 0.14 0.56
2a. C=0 2378.35
(353)
1672.35 2.09 (4)
0.72
0.45 (1)
0.50
Model 1 .45
(.20, .61)
-- .55
(.1, .72)
2b. A=0 2379.25
(353)
1673.25 2.99 (4)
0.56
1.34 (1)
0.25
Model 1 --- 0.37 0.63
2c. A=C=0 2405.18
(354)
1697.18 28.91 (5)
<.01
27.27 (2)
<.01
Model 1 --- ---- 1.0
Cognitive subscale
0. Sat (means free) 1621.32
(345)
931.32 ---- ---- ---- --- --- ---
0a. Sat (means
constrained)
1623.44
(348)
927.44 2.12 (3)
.55
---- ---- --- --- ---
1. ACE 1623.74
(351)
921.74 0.30 (3)
.97
---- ---- 0.28 0.18 0.54
2a. C=0 1624.50
(352)
920.50 1.06 (4)
.89
0.76 (1)
.38
Model 1 .49
(.3, .71)
-- .51
(.40, .90)
2b. A=0 1625.14
(352)
921.14 1.71 (4)
.78
1.40 (1)
.24
Model 1 --- 0.40 0.60
2c. A=C=0 1655.75
(353)
949.75 32.31 (5)
<.01
32.01 (2)
<.01
Model 1 --- --- 1.0

35

TABLE 5 (continued) Univariate genetic model fit values parameter estimates for SPQ-C total score and subscales
Overall Fit Change in Fit Parameter Estimates
Model -2LL(df) AIC χ
2
(df) p ∆ χ
2
( ∆df) p

Comparison A C E
Interpersonal
subscale

0. Sat (means
free)
1812.89 (345) 1122.89 ---- ---- ---- --- --- ---
0a. Sat (means
constrained)
1816.04 (348) 1120.04 3.15 (3) .4 ---- ---- --- --- ---
1. ACE 1819.69 (351) 1117.69 3.65 (3) .30 ---- ---- .35 .00 .65
2a. C=0 1819.91(352) 1115.69 3.87 (4) .42 .22 (1) .64 Model 1 .35
(.28, .56)
.00 .65
(.45, .91)
2b. A=0 1821.86 (352) 1117.86 5.81 (4) .21 2.16 (1) .14 Model 1 --- .27 .73
2c. A=C=0 1834.79 (353) 1128.79 18.75 (5)
<.01
15.09 (2)
<.01
Model 1 --- --- 1.0
Disorganization
subscale

0. Sat (means
free)
1678.07 (345) 988.07 ---- ---- ---- ---- ---- ----
0a. Sat (means
constrained)
1679.29 (348) 983.29 1.22 (3) .75 ---- ---- ---- ---- ----
1. ACE 1680.69 (351) 978.69 1.39 (3) .70 ----- ---- 0.36 0.10 0.54
2a. C=0 1680.93 (352) 976.93 1.64 (4) .8 0.24 (1) .62 Model 1 .47
(.21, .62)
--- .53
(.34, .64)
2b. A=0 1682.85 (352) 978.85 3.56 (4) .47 2.16 (1) .14 Model 1 --- .37 .63
2c. A=C=0 1708.92 (353) 1002.92 29.63 (5)
<.01
28.24 (2)
<.01
Model 1 --- ---- 1.0

36

Cross-twin cross trait correlations and multivariate genetic analyses
Cross-twin cross trait correlations are provided in TABLE 5 (see above). The DZ
cross-twin cross trait correlations were consistently less than the MZ cross-twin cross
trait correlations, suggesting that genetic factors contribute to the covariation amongst the
three schizotypal symptom dimensions.

Cholesky decomposition, Independent pathway, and Common Pathways models
were fit to the SPQ-C subscales in order to investigate genetic and environmental
relationships among the three schizotypal personality symptom dimensions. TABLE 6
provides a summary of the goodness of fit indices and sequential ∆ χ
2
test statistics for
each model. A full Cholesky ACE model fit the data (Model 1 in Table 6), and thus was
used as the model to which all submodels were compared. To investigate further the
genetic architecture of schizotypal traits, both common and independent pathways were
fitted to raw data.
The full Common Pathways model further provided a better fit to the data than the
saturated model (model 0a) according to the χ
2
and AIC. The AIC (329.83) statistic was
the lowest for the Common Pathways model, indicating that the model that fitted the data
best was such that allowed for genetic and unique environmental influences on
schizotypal traits in addition to the genetic and environmental influences operating
through the latent variable. When compared to the full saturated model, this model also
showed a significantly worse fit, indicating that the estimated variance and covariance
from this model was significantly different from the observed variance and covariance,

37
but at a much lower probability value than the other two models. Finally, a model that
dropped the common environment parameter yielded the lowest AIC statistic (329.83).
This CP-1 factor AE model fit the data best, according to the AIC value. Dropping either
the shared genetic or unique environmental path coefficients created a significant
deterioration in the overall fit of the model according to change in χ
2
, and corresponding
p-value. Therefore, the AE model was found to be the most parsimonious of all the
models fit, as it had the lowest AIC value.

TABLE 6 Multivariate genetic model fit values for SPQ subscales

Overall Fit

Change in Fit
Model -2LL(df) AIC χ
2
(df) p ∆ χ
2
( ∆df) p

Comparison

0. Sat (means free) 2379.21
(1010)
359.21 ----- ---- ----
0a. Sat (means
constrained)
2390.72
(1016)
358.72 11.50 (6)
.08
----- ----
1. CHOLESKY
ACE Model
2432.30
(1043)
346.298 41.58 (27)
.06
---- Model 0a
2a. 1 FACTOR
Independent Pathway
2425.56
(1043)
339.56 34.840
(27) .14
---- Model 1
2b. 1 FACTOR
Common Pathways
2430.29
(1047)
336.29 39.57
(30).11
---- Model 1
3a. 1 FACTOR CP
C=0
2431.83
(1051)
329.83 41.11 (35)
.22
1.54 (4)
.81
Model 2b
3b. 1 FACTOR CP
A=0
2480.22
(1051)
332.33 89.50 (35)
<.01
48.92
(4) <.01
Model 2b
3c. 1 FACTOR CP
A=C=0
2512. 91
(1052)
342.43 122.19
(36) <.01
81.61
(5) <.01
Model2b

38

As shown in FIGURE 6, the common schizotypy factor underlying all three symptom
dimensions was primarily explained by genetic influences, with a heritability of 0.60 in
addition to the influence of non-shared twin environment, 0.40. In order to calculate
estimates for proportions of variation, each standardized parameter estimate shown in
FIGURE 6 is squared (i.e. h2 of schizotypal personality latent factor= 0.77
2
.)

FIGURE 6. Parameter estimates from the best-fitting one common- factor model.

Cognitive Interpersonal
Disorganization
Schizotypal personality
AE
A E
A E A E
.77*
.63*
.94*
.
.84* .83*
.41* .39* .41* .49* .14 .26*

Note: Variability in each of the three subscales, Cognitive, Interpersonal, and
Disorganization is explained by specific genetic (a) and unique environmental (e)
influences, as well common genetic (A) and unique environmental (E) influences that act
through a latent common factor. Paths marked with an asterisk are significantly different
from zero.

39
ABLE 7. Breakdown of Common and Unique Effects
Common A Specific A Common E Specific E
T
Cognitive .51 .01 .34 .16
Interpsonal
Disorganization
er .41 .16 .26 .17
.41 .24 .27 .08

he Com Pathways m l, as seen in RE 6, demonstrated that
%
n

7
2

e for the
c
as the 95% CI for the genetic variance contained zero.
The results from t mon ode FIGU
one latent factor sufficiently explained the covariation among the cognitive-perceptual,
interpersonal-affective, and disorganized sub-factor symptom dimensions. As seen in
FIGURE 6, the common factor was significantly influenced by genes [0.77
2
= 0.60 (95
CI: .43, .72)] and unique environment [0.63
2
= 0.40 (95% CI: 28, .57)]. Factor loadings
are used in order to estimate the total heritability that is due to the common factor by
multiplying the squared loading with the squared parameter estimate for the common
genetic factor, A. The effects of non-shared environmental influences were estimated i
the same way. The heritability which is not due to the common factor is estimated by the
squared value of the unique genetic influence, a, in FIGURE 6. The summation of these
genetic components equals the total heritability of each variable. For example, for the
cognitive subscale, the common genetic factor accounted for .51 of its variance (.51=.7
x .94
2
) and no significant genetic variance is unique to this variable (.14
2
=.01 (95% CI:
.00, .10)). TABLE 7 provides further breakdown of both common and unique
contributions to each variable, calculated in a similar fashion. Specific varianc
interpersonal-affective and disorganized subscales could be accounted for by both geneti
and non-shared environmental effects (see TABLE 7); unique variance for the cognitive-
perceptual subscale could be accounted for by only non-shared environmental influences

40
Chapter 4: Discussion
4.1 General Discussion
This study sought to ad ap in the literature related to
aits during adolescence, by implementing a newly created scale
designe
13 year olds, using this brief, self-report screening
instrum
cal symptoms, could be particularly informative in its ability to
iden
s

dress a remarkable g
schizotypal personality tr
d to assess all nine DSM-IV criteria for schizotypal personality disorder. Not only
was this the first study to date to assess schizotypal personality traits using a community-
based sample, but it was the first to confirm a three-factor solution in an adolescent
sample of twins, using the SPQ-C.
A phenotypic three-factor structure similar to that which is often found in adults
emerged in our sample of 365 11-
ent, suggesting that it is possible to assess individuals as early as 11 years old for
schizotypal traits. Future studies are needed to further test the psychometric properties of
the SPQ-C, including the investigation of its clinical relevance and criterion validity
through clinical interviews for SPD and other diagnostic measures of schizophrenia
spectrum disorders.
The study of schizotypal traits during adolescence, years before the characteristic
manifestation of clini
tify individuals at risk to all disorders genetically related to schizophrenia. Research
on SPD and its precursor behavioral signs, such as schizotypal traits, is important in it
own right, as there is a 3% base rate in the general population for SPD. Psychometrically
measured schizotypy has been found to be elevated among schizophrenia patients
(Chapman et al., 1978) and relatives (Chen et al., 1998) and those with higher schizotypal

41
udies have shown schizotypal personality disorder is genetically related to
a viable
disorders, it also has been said
that the
o limited power. This study,
therefo

scores have more relatives with schizophrenia (Chapman et al., 1994). Moreover, twin
st
schizophrenia and therefore may reflect some facets of genetic predisposition to
schizophrenia (Chang et al., 2002; Kendler et al., 1993).
While it has been postulated that schizotypal personality may function as
marker for the predisposition to schizophrenia spectrum
construct can be regarded as part of normal personality variation. A phenotypic
three-factor solution fitted the data best, and the summated subscales were moderately
inter-correlated, suggesting potential genetic covariation among them. Genetic analyses
then revealed that the three identified subscales, in fact, share substantial genetic
etiology; thus, in our sample, the three subscales underlying schizotypal traits were
found to be neither phenotypically nor genetically distinct.
There have been relatively few studies to successfully use multivariate model
fitting to identify a ‘schizotypy’ latent factor, mainly due t
re, sought to elucidate further the genetic etiology of psychometrically measured
schizotypal traits through both univariate and multivariate genetic model fitting. The
results of the univariate analyses demonstrated a best-fitting model comprising additive
genetic and unique environmental components for schizotypal personality traits. These
findings are quite consistent with past twin studies of schizotypal personality traits in
adults (Kendler et al., 1991; Kendler & Hewitt, 1992; Miller, 1994; Linney et al., 2003;
Mac Donald et al. 2001). Our finding that one-factor Common Pathways-AE models
were the best-fitted model for all schizotypal subscales provides evidence for genetic but

4
not common environmental contribution to schizotypy in adolescents. Multivariate
2
ructural equation model fitting revealed a best-fitting model in which additive genetic
ne common latent
factor.

st
and unique environmental influences act through a single common latent pathway for
cognitive-perceptual, interpersonal-affective and disorganization symptom dimensions of
schizotypal personality. These results indicate that the components underlying
schizotypal personality are genetically related and that one underlying core dimension
can account for a significant amount of the covariation between them.
Overall, the additive genetic component estimated from the best-fitting models
was approximately 60% for the SPQ and its three factors through o
Based on these analyses, it could be concluded that schizotypal traits in
adolescents measured using self-report questionnaires have a modestly heritable
component and a substantially similar genetic etiology. These findings add to the
growing behavioral genetics literature indicating that adolescent schizotypy, a potential
endophenotype for schizophrenia spectrum disorders, is influenced by both genetic and
unique environmental factors and not shared environmental factors. Although it has been
demonstrated that schizotypal personality traits are quantitatively similar to the
schizophrenia symptoms, to the author’s knowledge, this is the first study to examine the
factors

underlying the relationship between the various known symptom dimensions to
schizotypal personality in a community-based, adolescent sample. The fact that
schizotypy was found to be influenced solely by additive genetic and unique
environmental influences is consistent both with twin and adoption studies of

43
schizophrenia and schizotypal personality traits in adult samples, which have reported
milar findings (Cardno, et al., 1999).
While common genetics effects were found to underlie the three aspects of
overed. For the three factors, approximately 2%
st two
latent f
4.2 Strengths and Limitations
There are several strengths to the current study. This is the first study, to the
author’s knowledge, to investigate psychometrically based schizotypal personality during
odeling techniques. The classic twin design is a powerful
ethod
si

schizotypal personality, specific genetic influences on the interpersonal-affective and
disorganized dimensions were also unc
and 4% of the genetic variance contributing to the interpersonal factor and
disorganization factor, respectively, was accounted for by specific genetic factors.
In addition to being the first investigation of schizotypal traits during adolescence
using biometric model fitting, this study was the first to uncover one latent typology. For
example, in an adult, unselected sample, Linney et al., (2003) proposed that at lea
actors are needed to account for the covariation among the symptom dimensions,
namely a positive symptom dimension and a negative symptom dimension. This
discrepancy gives rise to the notion that perhaps there is something quantitatively
different about schizotypal traits during adolescence as compared to during adulthood.
Future longitudinal studies are therefore needed to investigate the developmental
trajectory of schizotypal traits.
adolescence, using biometric m
m for estimating the degree to which genetic and environmental influences
contribute to schizotypal traits and the relationship between the various symptoms

44
dimensions within the construct. Another major strength of this study is that it employs a
community sample that is representative of Los Angeles County. Past research on
schizotypy has focused on undergraduates, adults, or unaffected relatives of
hizophrenic patients. In contrast, this is the first study to examine the genetic etiology
ize would undoubtedly increase power to detect the effect of common
environ
elated to an overarching, conceptual difficulty intrinsic to community-
based
sc
of schizotypal traits in an unaffected sample of adolescent twins. This is particularly
informative because schizophrenia research, in any respect, is characteristically rife with
potential confounds, such as anti-psychotic medications, acute psychosis, or
hospitalization.
However, the strengths of the study and our results must be interpreted in light of
the limitations of our sample and general concerns regarding twin methodology. First,
though our sample size was moderate in comparison to other twin studies on schizotypy,
a larger sample s
ment (C).
Second, subjects were unselected for any particular psychopathology (though they
were not excluded for any specific criteria) and one must take the population-based
nature of this sample into consideration when interpreting any results. Thus, another
limitation is one r
methodology; this study makes no attempt to relate the assessment and
identification of schizotypal traits to clinical diagnosis of schizophrenia, per se. Rather,
this study sought to clarify the phenotypic factorial structure of schizotypal traits in a
sample that had yet to be studied and to investigate the etiological processes underlying
this personality structure at a time in development when behaviors are known to emerge.

4
New strategies are needed to accurately uncover individuals who are at putative risk to
schizophrenia spectrum disorders, prior to the manifestation of clinical symptoms. Future
studies of schizotypal traits both in unaffected children and adolescents and in individuals
5
ith prodromal symptoms (or identified as prodromes) are needed to further investigate
rs, this would clearly deflate the heritability
estimat
al-affective (negative symptom factor) (Raine, 1992). However, this was not
the cas
w
the emerging clinical, behavioral, and physiological processes at play during
development. Studies are also needed to validate the newly created measure, the SPQ-C,
as it has now shown promise in its assessment of the widely agreed upon subcomponents
to schizotypal personality or schizotypy.
Another limitation is one that is unfortunately inherent to the very nature of the
twin design; several assumptions were made. For example, it must be assumed in twin
models that random mating occurs in the parent generation. As assortative mating
increases the similarity between DZ pai
es and inflate the shared environmental estimates. In addition, twin models must
assume that shared environmental effects contribute equally to the similarity in MZ and
DZ pairs (the equal environment assumption (EEA)), though this is arguably not always
the case.
Past research has often reported sex differences in schizotypal traits assessed in
college-aged and adult samples, namely that females tend to endorse more of the
cognitive-perceptual items (positive symptom factor) and men tend to score higher on the
interperson
e in our adolescent sample, suggesting that there is potentially a developmental
process that our sample or design failed to capture.

4
4.3 Future Directions
A next step for behavioral-genetic research into schizotypal traits is to examine
putative endophenotypes, both phenotypically and then at the biom
6
etric level.
Conceptually, endophenotypes are tools which hold the unique potential to reveal genetic
and neurobiological underpinnings of psychopathology. The endophenotype strategy is
an important tool in neuropsychiatric research, as it holds the potential to uncover the
genetic architecture of schizophrenia (Braff et al., 2007; Gottesman & Shields, 1972).
Psychophysiological deficits in schizotypal personality and schizophrenic patients
have been found to extend from the earliest pre-attentive stages of information processing
to higher cortical processes (Hall et al., 2006; van Beijsterveldt et al., 2002).
Schizophrenia spectrum endophenotypes, such as p50, MMN, and p300 components of
the auditory evoked potential (AEP) have been found to be highly heritable, reliable, and
informative indices of neurobiological dysfunction in adults (for reviews see Turetsky
2007; Hall et al., 2007). However, few studies have examined the relationship between
ERP dysfunction and schizotypal traits. Moreover, no studies, to date, have investigated
this during childhood and adolescence or have looked at the genetic covariation amongst
these variables during development. These unanswered yet potentially informative
questions seem like viable next steps in research on schizotypal personality and the
potential role it plays in the schizophrenia spectrum liability puzzle.

47
References
Akaike, A. C. (1987). Factor analysis and AIC. Psychometrika 52, 317-32
American Psychiatric Association. (1994). Diagnostic and statistical manual of mental
disorders (4
th
ed.). Washington, DC: Author.

Andreasen NC. (1982). Negative symptoms in schizophrenia: definition and reliability.
Arch Gen Psychiatry. 39:784-788.

Baker LA, Barton M, Lozano DI, Raine A, Fowler JH. (2006). The Southern California
Twin Register at the University of Southern California: II. Twin Res Hum Genet.
9(6):933-940.

Baker LA, Jacobson KC, Raine A, Lozano DI, Bezdjian S. (2007). Genetic and
environmental bases of childhood antisocial behavior: A multi-informant twin
study. Journal of Abnormal Psychology. 116(2): 219-235.
Baker, L.A. (2006). Methods for understanding genetic and environmental influences in
normal and abnormal personality. In Strack, S. (Ed.) Differentiating Normal and
Abnormal Personality-Second Edition.
Ballmaier, M., Toga, A. W., Siddarth, P., Blanton, R. E., Levitt, J. G., & Lee, M. (2004).
Thought disorder and nucleus accumbens in childhood: A structural MRI study.
Psychiatry Research: Neuroimaging, 130(1), 43-55.
Barrantes-Vidal, N., Kwapil, T.R., Caparrós, B., Obiols, J.E., (submitted for publication.)
Four-year stability of psychometrically defined positive and negative schizotypy
in an adolescent community sample.
Benes, F. M. & Tamminga, C. A. (2002). Neurobiology of schizophrenia. New York,
NY, US: John Wiley & Sons

Berenbaum, H., and McGrew, J. (1993) Familial resemblance of schizotypic traits.
Psychological Medicine, 23:327-333,

Braff, DL., Freedman, R.,Schork, NJ., Gottesman, I (2007) Deconstructing
Schizophrenia: An Overview of the Use of Endophenotypes in Order to
Understand a Complex Disorder Schizophrenia Bulletin Advance Access
Schizophr Bull 33: 21-32.
Braff, D.L., (1993) Information processing and attention dysfunctions in schizophrenia.
Schizophrenia Bulletin 19, pp. 233–259.

48

Cancro RC, Lehmann HE. Schizophrenia: clinical features. Comprehensive Textbook of
Psychiatry. Baltimore, Md: Williams & Wilkins:1169-1199.
Cannon, TD (2005b) The inheritance of intermediate phenotypes for schizophrenia. Curr

annon TD, Kaprio J, Lonnqvist J, Huttunen M, Koskenvuo M. (1998) The genetic
y.55:67-74.
Contributions of genes, environment, and their interactions.
Schizophrenia bulletin, 29(4), 653-669.
Chapm l
Chapman, L.J., Chapman, J.P., Kwapil, T.R., Eckblad, M. and Zinser, M.C., (1994)
Chen, W.J., Hsiao, C.K. and Lin, C.C.H., (1997). Schizotypy in community samples:
Claridge, G., (1997). Theoretical background and issues. In: Claridge, G., Editor,
k.
laridge, G., McCreery, C., Mason, O., Bentall, R., Boyle, G., Slade, P. and Popplewell,
y.
.
schizotypy.
ty,

Claridge, C. McCreery, O. Mason, R. Bentall, G. Boyle, P. Slade and D. Popplewell

Opin Psychiatry; 18:135–140
C
epidemiology of schizophrenia in a Finnish twin cohort: a population-based
modeling study.Arch Gen Psychiatr
Cannon, T. D., G. M. van Erp,Theo, Bearden, C. E., Loewy, R., Thompson, P., & Toga,
A. W. (2003). Early and late neurodevelopmental influences in the prodrome to
schizophrenia:
an, L.J., Chapman, J.P. and Raulin, M.L., (1976) Scales for physical and socia
anhedonia. J. Abnorm. Psychology 85, pp. 374–382.
Chapman, L.J., Chapman, J.P. and Raulin, M.L., (1978) Body-image aberration in
schizophrenia. J. Abnorm. Psychology 87, pp. 399–407.
Putatively psychosis-prone subjects 10 years later. J. Abnorm. Psychology 103,
pp. 171–183.
the three-factor structure and correlation with sustained attention. J. Abnorm.
Psychology 106, pp. 649–654.
Schizotypy. Implications for Illness and Health, Oxford Univ. Press, New Yor
C
D., (1996) The factor structure of ‘schizotypal’ traits: a large replication stud
Br. J. Clin. Psychol. 35, pp. 103–115
Claridge and T. Beech (1995) Fully and quasi-dimensional constructions of
In: A. Raine, T. Lencz and S.A. Mednick, Editors, Schizotypal Personali
Cambridge Univ. Press, Cambridge, NY, pp. 192–216.
(1996) The factor structure of ‘schizotypal’ traits: a large replication study. Br. J.
Clin. Psychol. 35, pp. 103–115.

49

ophrenia: A meta-analysis. Psychiatry Research:
Neuroimaging, 122(2), 69-87.
DiDuca
673–682.
Elliott, psychology of schizophrenia: relations
with clinical and neurobiological dimensions. Psyhol. Med. 25, pp. 581–594
Eysenc version of the
psychoticism scale. Personality and Individual Differences, 6, pp. 21-29.
Faraon , N., & Caviness, V.
S. (2003). Structural brain abnormalities among relatives of patients with
3),
Franke, P., Maier, W., Hardt, J. and Hain, C., (1993) Cognitive functioning and
.
Fusar-P orgwardt, S., Placentino, A., & Caverzasi, E.
(2007). Neurofunctional correlates of vulnerability to psychosis: A systematic
-

Golden, R. R. and P. E. Meehl (1979). Detection of the schizoid taxon with MMPI
Goodin slippage in schizotypic
individuals. J. Nerv. Ment. Dis. 189, pp. 750–756.
Gottesm
s. Am J Psychiatry; 160:636–645.

Crow ,T.J (1995) A continuum of psychosis, one human gene and not much else: the case
for homogeneity. Schizophr. Res. 17 (1995), pp. 135–145.

Davidson, L. L., & Heinrichs, R. W. (2003). Quantification of frontal and temporal lobe
brain-imaging findings in schiz
, D. and Joseph, S., (1999) Assessing schizotypal traits in 13–18 year olds:
revising the JSS. Pers. Individ. Differ. 27, pp.
Diforio, D., Walker, E.F. and Kestler, L.P., (2000) Executive functions in adolescents
with schizotypal personality disorder. Schizophr. Res. 42, pp. 125–134.
R. and Sahakian, B.J., (1995) The neuro
k, H.J., Eysenck, S.B.G. & Barrett, P. (1985). A revised
e, S. V., Seidman, L. J., Kremen, W. S., Kennedy, D., Makris
schizophrenia: Implications for linkage studies. Schizophrenia research, 60(2-
125-140.
anhedonia in subjects at risk for schizophrenia. Schizophr. Res. 10, pp. 77–84
oli, P., Perez, J., Broome, M., B
review and meta-analysis. Neuroscience & Biobehavioral Reviews, 31(4), 465
484.
Galdos and J. van Os, Gender, psychopathology, and development: from puberty to early
adulthood, Schizophr. Res. 14 (1995) (2), pp. 105–112.
indicators. J Abnorm Psychol 88(3): 217-33.
g, D.C., Tallent, K.A. and Hegyi, J.V., (2001) Cognitive
an, II and Gould, TD. (2003) The endophenotype concept in psychiatry:
etymology and strategic intention

50

Gottesman, I.I., and Bertelsen, A. (1989) Confirming unexpressed genotypes for
al
.

endophenotypes. Schizophrenia bulletin, 33(1), 49-68.
Hall, M
mismatch negativity." Am J Med Genet B Neuropsychiatr Genet 141(4): 336-43.
Hall, M ramon,
nia and Event-Related Potentials Am J Psychiatry. 164: 804-812

nen,
predictors of schizophrenia in the Northern Finland 1966 Birth Cohort—a

Isohanni, P.B. Jones, K. Moilanen, P. Rantakallio, J. Veijola, H. Oja, M. Koiranen, J.
Jokelainen, T. Croudace and M.-R. Jarvelin, (2001) Early developmental
, pp. 1–19.
ia. Am

chizophrenia Bulletin,
22(3):511-520,
Kendle der: A
historical perspective. Schizophrenia Bulletin, ll(4):538-555,.

Gottesman II, Shields J. (1982). Schizophrenia: The Epigenetic Puzzle. New York, NY:
Cambridge University Press.

Gottesman, I.I. (1994) Schizophrenia epigenesis: Past, present, and future. Acta
Psychiatrica Scandinavica, 90 (384):26-33,
schizophrenia: Risks in the offspring of Fischer's Danish identical and fratern
discordant twins. Archives of General Psychiatry, 46:867-872
Gruzelier and J. Kaiser (1996). Syndromes of schizotypy and timing of puberty,
Schizophr. Res. 21 (3), pp. 183–194.
Gur, R. E., Calkins, M. E., Gur, R. C., Horan, W. P., Nuechterlein, K. H., & Seidman, L.
J.(2007). The consortium on the genetics of schizophrenia: Neurocognitive

. H., K. Schulze, (2006). "Genetic overlap between P300, P50, and duration

., Rijsdijk, F., Picchioni, M., Schulze, K., Ettinger, U., Toulopoulou, T., B
E., Murray, R.M., Sham, P., (2007) Substantial Shared Genetic Influences on
Schizophre
Isohanni, P. Jones, L. Kemppainen, T. Croudace, I. Isohanni, J. Veijola, S. Raesae
K.E. Wahlberg, P. Tienari and P. Rantakallio, (2000) Childhood and adolescent
descriptive life-span model, Eur. Arch. Psychiatry Clin. Neurosci. 250 (6), pp.
311–319.
milestones in adult schizophrenia and other psychoses. A 31-year follow-up of the
Northern Finland 1966 Birth Cohort, Schizophr. Res. 52

Kendler KS. (1983) Overview: a current perspective on twin studies of schizophren
J Psychiatry. 140:1413-1425.
Kendler, K.S.; Thacker, L,Walsh, D. (1996) Self-report measures of schizotypy as
indexes of familial vulnerability to schizophrenia. S

r, K.S. (1985) Diagnostic approaches to schizotypal personality disor

51

7,.
rmal
itive
57(1), 3-8.
Kwapil
psychosis proneness: a
partial replication. J. Abnorm. Psychology 106, pp. 491–495.
Lauren
Community screening for psychotic-like experiences and other putative
awrie, S. M., Whalley, H. C., Abukmeil, S. S., Kestelman, J. N., Donnelly, L., & Miller,
at
Lenzen ve
of schizophrenia liability. In: Lenzenweger, M.F., and Dworkin, R.H.,
eds. Origins and Development of Schizophrenia: Advances in Experimental
.

Lenzen phrenia
using a psychometric measure of schizotypy. Archives of General Psychiatry,
enzenweger M.F. and Korfine, L., (1994). Perceptual aberrations, schizotypy and the
Liddle, P., (1987). The symptoms of chronic schizophrenia: a re-examination of the
positive–negative dichotomy. Br. J. Psychiatry 151, pp. 221–234.

Kendler, K.S., and Hewitt, J. (1992) Structure of self-report schizotypy in twins. Journal
of Personality Disorders,6:1-1

Kerns, J. G. (2005). Positive schizotypy and emotion processing. Journal of abno
psychology, 114(3), 392-401.
Keshavan, M. S., Berger, G., Zipursky, R. B., Wood, S. J., & Pantelis, C. (2005).
Neurobiology of early psychosis. British Journal of Psychiatry. Special Issue:
Early psychosis: A bridge to the future, 187(Suppl48), s8-s18.
Kurachi, M. (2003). Pathogenesis of schizophrenia: Part I. symptomatology, cogn
characteristics and brain morphology. Psychiatry and clinical neurosciences,
, T.R., Miller, M.B., Zinser, M.C., Chapman, J. and Chapman, L.J., (1997).
Magical Ideation and Social Anhedonia as predictors of
s , S . Hodgins , B . Maughan , R . Murray , M . Rutter , E . Taylor (2007)
antecedents of schizophrenia in children aged 9–12 years . Schizophrenia
Research ,Volume 90 , Issue 1 - 3 , pp. 130 - 146
L
P. (2001). Brain structure, genetic liability, and psychotic symptoms in subjects
high risk of developing schizophrenia. Biological psychiatry, 49(10), 811-823.
weger, M.F. Schizotypy and schizotypic psychopathology: Mapping an alternati
expression
Psychopathology. Washington, DC: American Psychological Association, 1998
pp. 93-121.
weger, M.F., and Loranger, A.W. (1989) Detection of familial schizo
46:902-907.
L
Wisconsin Card Sorting Test. Schizophr. Bull. 20, pp. 345–357.

52
hizotypal personality traits, Psychol. Med. 33 ,
pp. 803–816.
Lykken ity in twins. Behavior Genetics, 8, 437-473.

al trait
dimens

MacDonald, III. A., Michael F. Pogue-Geile , Thomas T. Debski , and Stephen Manuck
Marcus, J., Hans, S.L., Lewow, E., Wilkinson, L. and Burack, C.M., (1985).
McGuffin, A.E. Farmer and I. Harvey (1991). A polydiagnostic application of operational
stem,
Meehl, atry 46, pp. 935–944.

Meehl, P.E. (1990) Toward an integrated theory of schizotaxia, schizotypy, and

Meijer d

iller, T.J., McGlashan, T.H., Woods, S.W., Stein, K., Driesen, N., Corcoran, C.M.,

Miller, T.J., McGlashan, T.H., Rosen, J.L., Somjee, L., Markovich, P.J., Stein, K.,
romes:
preliminary evidence of interrater reliability and predictive validity. Am. J.

Linney et al., Y. Linney, R. Murray, E. Peters, A. MacDonald and S. Rijsdijk, A (2003)
Quantitative genetic analysis of sc
(1978) The diagnosis of zygos
Loughland, C.M. and Williams, L.M., (1997) A cluster analytic study of schizotyp
ions. Pers. Individ. Differ. 23, pp. 877–883.
(2001) Genetic and Environmental Influences on Schizotypy: A Community-
Based Twin Study Schizophr Bull 27: 47-58.
Neurological findings in high-risk children: childhood assessment and 5-year
follow-up. Schizophr. Bull. 11, pp. 85–100.
criteria in psychotic illness: development and reliability of the OPCRIT sy
Arch. Gen. Psychiatry 48, pp. 50–55.
P., (1989) Schizotaxia revisited. Arch. Gen. Psychi
Meehl, P.E. (1962) Schizotaxia, schizotypy, schizophrenia. American Psychologist,
17:828-838.
schizophrenia. Journal of Personality Disorders, 4:1-99.
and P.D. Treffers, (1991) Borderline and schizotypal disorders in children an
adolescents, Br. J. Psychiatry 158, pp. 205–212.
M
Hoffman, R., Davidson, L., (1999) Symptom assessment in schizophrenic
prodromal states. Psychiatr. Q. 70 (4), 273– 287.
Woods, S.W., (2002). Prospective diagnosis of the initial prodrome for
schizophrenia based on the Structured Interview for Prodromal Synd
Psychiatry 15. (5), 863–865.

53

Interview for Prodromal Syndromes and the Scale of Prodromal
Symptoms: predictive validity, interrater reliability, and training to reliability.
adison,.
7.
ersonality
68

Murray, R.M., O'Callaghan, E., Castle, D. and Lewis, S.W., (1992) A
phr.
Bull. 18, pp. 318–332.
ory
escents from a pacific island isolate with elevated risk for
schizophrenia, Biol. Psychiatry 55 (7), pp. 663–667
Neale, dies of twins and
families. Dordrecht, The Netherlands: Kluwer Academic Publications.
Neale, , VA: Department of
Psychiatry, Medical College of Virginia
Niu, L. 004).
h schizophrenia: A magnetic
resonance imaging study. Psychiatry Research: Neuroimaging, 132(1), 41-51.
Obiols, 9).
us
.
28

Miller, T.J., McGlashan, T.H., Rosen, J.L., Cadenhead, K., Ventura, J., McFarlane, W.,
Perkins, D.O., Pearlson, G.D., Woods, S.W., (2003). Prodromal assessment with
the Structured
Schizophr. Bull. 29 (4), 703– 715.

Miller, M.B. (1993) A Study of Schizotypal Traits in Young Male Twins. Unpublished
Ph.D. dissertation, University of Wisconsin, M
Mohanty, A., Herrington, J. D., Koven, N. S., Fisher, J. E., Wenzel, E. A., & Webb, A.
G. et al. (2005). Neural mechanisms of affective interference in schizotypy.
Journal of abnormal psychology, 114(1), 16-2
Muntaner, C., Garcia-Sevilla, L.L., Fernandez, A. and Torrubia, R., (1988). P
dimensions, schizotypal and borderline personality traits and psychosis-
proneness. Pers. Individ. Differ. 9, pp. 257–2
Murray, R.M. and Lewis, S.W., (1987). Is schizophrenia a neurodevelopmental disorder?
Br. Med. J. 295, pp. 681–682.
neurodevelopmental approach to the classification of schizophrenia. Schizo
Myles-Worsley, L. Ord, F. Blailes, H. Ngiralmau and R. Freedman, (2004). P50 sens
gating in adol

M. C. & Cardon, L. R. (1992). Methodology for genetic stu

M. C. (1997). Mx: Statistical modeling. Richmond

, Matsui, M., Zhou, S., Hagino, H., Takahashi, T., & Yoneyama, E. et al. (2
Volume reduction of the amygdala in patients wit
J.E., Serrano, F., Caparrós, B., Subirà, S. and Barrantes-Vidal, N., (199
Neurological soft signs in adolescents with poor performance on the continuo
performance test: markers of liability for schizophrenia spectrum disorders?
Psychiatry Res. 86, pp. 217–2

54
Pearlso al Brain Imaging in Schizophrenia: A
Selective Review, Biol. Psychiatry, vol. 46, p. 627.
Peralta e than two syndromes in
schizophrenia? A critique of the positive–negative dichotomy. Br. J. Psychiatry
Plomin P. (2001). Behavioral Genetics.
United States of America: Worth Publisher.
Pogue- mplex
eurobehavioral Genetics:
Methods and Applications. Boca Raton, FL: CRC Press, 1999. pp. 247-263.
Poreh, R.D., (1995). Reexamination of executive
functions in psychosis-prone college students. Pers. Individ. Differ. 18 4, pp.
Raine, A. (1991). The SPQ: A scale for the assessment of schizotypal personality based

Raine, ening instrument for
schizotypal personality disorder. Journal of Personality Disorders 9 346-355.
Raine,
(1994).Cognitive-perceptual, interpersonal, and disorganized features of
aine, A., Sheard, C., Reynolds, G.P. and Lencz, T., (1992). Pre-frontal structural and
Rawlings, D. and McFarlane, C., (1994). A multidimensional schizotypal traits
Reitan, in
. 271–276.

Park, S., Holzman, P.S. and Lenzenweger, M.F., (1995). Individual differences in
spatial working memory in relation to schizotypy. J. Abnorm. Psychology 104,
pp. 355– 364.
n, G. and Marsh, L. (1998) Structur
, V., Cuesta, M.J. and de Leon, J., (1992). Are there mor
161, pp. 335–343.
, R., DeFries, J. C., McClearn, G. E. & McGuffin,

Geile, M.F., and Gottesman, I.I. Schizophrenia: Study of a genetically co
phenotype. In: Jones, B.C., and Mormede, P., eds. N
A.M., Ross, T.P. and Whitman,
535–539.
on DSM-III-R criteria. Schizophrenia Bulletin 17 556-564.
A. and Benishay, D. (1995). The SPQ-B: A brief scre

A.; Reynolds, C; Lencz, T; Scerbo, A.; Triphon, N.; and Kim, D
schizotypal personality. Schizophrenia Bulletin, 20(1): 191-201
R
functional deficits associated with individual differences in schizotypal
personality. Schizophr. Res. 7, pp. 237–247.
questionnaire for young adolescents. Pers. Individ. Differ. 17, pp. 489–496.
R.M., (1958). Validity of the Trail Making Test as an indicator of organic bra
damage. Percept. Mot. Skills 8, pp

55
nd

ic
uditory evoked responses.
Arch Gen Psychiatry; 41:617–612.
Siever,
Am. 16, pp. 217–245.

schizophrenia: A meta-analysis.
Schizophrenia research, 71(2-3), 285-295
Stefani
dimensions of psychosis have a distribution in the general population. Psychol.
Strauss, M.E., (1993). Relations of symptoms to cognitive deficits in schizophrenia.
Schizophr. Bull. 19, pp. 215–231.
Suhr, J psychosis-prone
college students. Schizophr. Res. 27, pp. 29–35.
Suhr, J
traits: A comparison of unselected and highly schizotypal samples. Schizophr.
Trestm
Roitman, S., Davidson, M., Aronson, A., Silverman, J. and Siever, L.J., (1995).
Tsuang, M.W. Gilbertson and S.V. Faraone (1991). Genetic transmission of negative and
arneros,
Schizophrenia, Springer, Berlin, pp. 265–291.
Tsuang enia: a review of genetic studies.
Harv. Rev. Psychiatry 7, pp. 185–208.
ent and
schizophrenia. Br. J. Psychiatry 178, s18–s24. (doi:10.1192/bjp.178.40.s18)

Rosa, A., Van Os, J., Fañanás, L., Barrantes-Vidal, N., Caparrós, B., Gutiérrez, B. a
Obiols, J.E., (2000). Developmental instability and schizotypy. Schizophr. Res.
43, pp. 1–10.
Siegel, C, Waldo, M, Mizner, G, (1984). Deficits in sensory gating in schizophren
patients and their relatives. Evidence obtained with a

L.J., Kalus, O.F. and Keefe, R.S.E., (1993). The boundaries of schizophrenia.
Psychiatr. Clin. North
Sitskoorn, M. M., Aleman, A., Ebisch, S. J. H., Appels, M. C. M., & Kahn, R. S. (2004).
Cognitive deficits in relatives of patients with
s, N.C., Hanssen, M., Smirnis, N.K., Avramopoulos, D.A., Evdokimidis, I.K.,
Stefanis, C.N., Verdoux, H. and van Os, J., (2002). Evidence that three
Med. 32 2, pp. 347–358.
.A., (1997). Executive functioning deficits in hypothetically
.A. and Spitznagel, M.B., (2001). Factor versus cluster models of schizotypal
Res. 52, pp. 231–239.
an, R.L., Keefe, R.S.E., Mitropoulou, V., Harvey, P.D., deVegvar, M.L., Lees-
Cognitive function and biological correlates of cognitive performance in
schizotypal personality disorder. Psychiatry Res. 59, pp. 127–136.
positive symptoms in the biological relatives of schizophrenics. In: A. M
N.C. Andreasen and M.T. Tsuang, Editors, Negative versus Positive

, W.S. Stone and S.V. Faraone (1999). Schizophr

Tsuang, M. T., Stone, W. S. & Faraone, S. V. (2001) Genes, environm

56

ate measures. Schizophrenia bulletin, 33(1), 69-94.
approach to the symptomatology of
recent–onset schizophrenia. J. Nerv. Ment. Dis. 181, pp. 744–749.
van Os
osis-like symptoms and the continuum
of psychosis. Soc. Psychiatry Psychiatr. Epidemiol. 34, pp. 459–463.
van Os factors
tic
chotic symptoms. Arch. Gen. Psychiatry
58, pp. 663–668.
Venabl izotypy, its relation to
subdiagnoses of schizophrenia and to sex and age. Br. J. Clin. Psychol. 33, pp.

Venables, P. and Rector, N (1999). The Content and Structure of Schizotypy: A Study
Using Confirmatory Factor Analysis Schizophr Bull 26: 587-602.
Vollem
Vollema, M.G.; Sitskoorn, M.M.; Appels, M.C.M.; and Kahn, R.S. (2002) Does the
Walker elopment and the emergence of
psychotic symptoms. Schizophrenia Research. Special Issue: NATO advanced

Turetsky, B. I., Calkins, M. E., Light, G. A., Olincy, A., Radant, A. D., & Swerdlow, N.
R. (2007). Neurophysiological endophenotypes of schizophrenia: The viability of
selected candid
van Beijsterveldt, C. E. and G. C. van Baal (2002). "Twin and family studies of the
human electroencephalogram: a review and a meta-analysis." Biol Psychol
61(1-2): 111-38.
van der Does, A.J.W., Linszen, D.H., Dingemans, P.M., Nugter, M.A. and Scholte,
W.F., (1993). A dimensional and categorical
, J., Verdoux, H., Maurice-Tison, S., Gay, B., Liraud, F., Salamon, R. and
Bourgeois, M., (1999). Self-reported psych
, J., Jones, P., Sham, P., Bebbington, P. and Murray, R.M., (1998). Risk
for onset and persistence of psychosis. Soc. Psychiatry Psychiatr. Epidemiol.
33, pp. 596–606.
van Os, J., Hanssen, M., Bijl, R.V. and Vollebergh, W., (2001). Prevalence of psycho
disorders and community level of psy
es, P.H. and Bailes, K., (1994). The structure of sch
277–294.
a, M.G. and Van den Bosch, R.J., (1995). The multidimensionality of
schizotypy. Schizophr. Bull. 21, pp. 19–32.
Schizotypal Personality Questionnaire reflect the biological-genetic vulnerability
to schizophrenia? Schizophrenia Research, 54:39
, E., & Bollini, A. M. (2002). Pubertal neurodev
research workshop on schizophrenia and schizotypy, 54(1-2), 17-23.

57
.F.
., Editors, 1998. Origins and Development of Schizophrenia,
APA Press, Washington, DC, pp. 469–492.
Walker
. 12–19.
Williams, L.M., (1994). The multidimensional nature of schizotypal traits: a cluster
William urther evidence for a multidimensional personality disposition
to schizophrenia in terms of cognitive inhibition. Br. J. Clin. Psychol. 34, pp.
William gnitive inhibition and schizophrenic symptom subgroups.
Schizophr. Bull. 22, pp. 139-151
Wolff,
beling. Br J Psychiatry 159: 615-20, 634-5.
isorder. Br J
Psychiatry 159: 620-9, 634-5.

Walker, E. and Lewine, R., (1988). The positive/negative symptom distinction in
schizophrenia: validity and etiological relevance. Schizophr. Res. 1, pp. 315–
328.
Walker, E.F., Baum, K.M. and Diforio, D., (1998). Developmental changes in the
behavioral expression of vulnerability for schizophrenia. In: Lenzenweger, M
and Dworkin, R.H
, E.F., Diforio, D. and Baum, K., (1999). Developmental neuropathology and the
precursors of schizophrenia. Acta Psychiatr. Scand. 99 Suppl. 395, pp
Weinberger, D.R., (1987). Implications of normal brain development for the
pathogenesis of schizophrenia. Arch. Gen. Psychiatry 44, pp. 660–669.
analytic study. Pers. Individ. Differ. 16, pp. 103–112.
s, L.M., (1995). F
193–213.
s, L.M., (1996). Co
S. (1991). Schizoid personality in childhood and adult life. I: The vagaries of
diagnostic la

Wolff, S., R. Townshend (1991).Schizoid personality in childhood and adult life. II:
Adult adjustment and the continuity with schizotypal personality d

Abstract (if available)

Abstract The study described in the present paper attempted to clarify further the genetic and environmental etiology of schizotypal personality traits through biometric model fitting of data from a sample of MZ (n=91 twin pairs) and DZ (n=87 twin pairs) adolescent twins (age 11-13 years old) drawn from the general population. Univariate genetic analyses found that schizotypal traits are modestly heritable (additive genetic effects ranging from 35 to 49%). Multivariate genetic model fitting results indicated that additive genetic and unique environmental influences acted through a single common latent pathway for cognitive-perceptual, interpersonal-affective and disorganization symptom dimensions of schizotypal personality. The covariation among the three schizotypy sub-factors could be accounted for by a common 'schizotypy' latent factor which was significantly heritable, with additive genetic factors explaining 60% of the latent factor variance.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Heritability and longitudinal stability of impulsivity in adolescence

PDF

Psychometric and biometric evaluation of psychopathic personality traits in 9 and 10 year old twins

PDF

Biometric models of psychopathic traits in adolescence: a comparison of item-level and sum-score approaches

PDF

Multivariate biometric modeling among multiple traits across different raters in child externalizing behavior studies

PDF

Patterns of EEG spectral power in 9-10 year old twins and their relationships with aggressive and nonaggressive antisocial behavior in childhood and adolescence

PDF

Developmental etiology of reactive and proactive aggression, electrodermal reactivity and their relationships: a twin study

PDF

Construct validity of psychopathic personality traits in a cohort of young twins

PDF

Different genetic and environmental structures for the overlap of three antisocial behavior factors with alcohol initiation

PDF

Sources of stability and change in the trajectory of openness to experience across the lifespan

PDF

Personality and cognitive aging: are positive traits protective?

PDF

Psychophysiological reactivity to stress, psychopathic traits, and developmental trajectories of antisocial behavior: a prospective, longitudinal study

PDF

Examining the effect of fertility on lifespan using data from the Swedish Twin Registry

Asset Metadata

Creator Ericson, Marissa (author)

Core Title Heritability of schizotypal traits during adolescence

School College of Letters, Arts and Sciences

Degree Master of Arts

Degree Program Psychology

Publication Date 12/01/2008

Defense Date 09/10/2008

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag genetics,heritability,OAI-PMH Harvest,schizophrenia,schizotypal personality,schizotypy

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Baker, Laura A. (committee chair), Mednick, Sarnoff (committee member), Prescott, Carol A. (committee member)

Creator Email marissae@usc.edu,mle9@cornell.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m1850

Unique identifier UC1261978

Identifier etd-Ericson-2469 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-128802 (legacy record id),usctheses-m1850 (legacy record id)

Legacy Identifier etd-Ericson-2469.pdf

Dmrecord 128802

Document Type Thesis

Rights Ericson, Marissa

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu