Psychiatry Research 215 (2014) 453–459

Contents lists available at ScienceDirect

Psychiatry Research journal homepage: www.elsevier.com/locate/psychres

Genetic and environmental bases of the interplay between magical ideation and personality Paolo Brambilla a,b,n,1, Corrado Fagnani c,1, Filippo Cecchetto a, Emanuela Medda c, Marcella Bellani d, Miriam Salemi c, Angelo Picardi e, Maria Antonietta Stazi c a

DISM, InterUniversity Center for Behavioural Neurosciences (ICBN), University of Udine, Udine, Italy "E. Medea", UDGEE, Udine, Italy c Genetic Epidemiology Unit, National Centre of Epidemiology, Surveillance and Health Promotion, Italian National Institute of Health, Rome, Italy d Department of Public Health and Community Medicine, Section of Psychiatry and Clinical Psychology, Inter-University Centre for Behavioural Neurosciences (ICBN), University of Verona, Verona, Italy e Mental Health Unit, National Centre of Epidemiology, Surveillance and Health Promotion, Italian National Institute of Health, Rome, Italy b

art ic l e i nf o

a b s t r a c t

Article history: Received 26 June 2013 Received in revised form 20 November 2013 Accepted 21 November 2013 Available online 2 December 2013

Sub-threshold psychotic symptoms are quite commonly present in general population. Among these, Magical Ideation (MI) has been proved to be a valid predictor of psychosis. However, the genetic and environmental influences on the interplay between MI and personality have not fully been explored. A total of 534 adult twins from the population-based Italian Twin Register were assessed for MI using the MI Scale (MIS) and for personality with the temperament and character inventory (TCI). A Multivariate Cholesky model was applied with Mx statistical program. The best-fitting model showed that additive genetic and unshared environmental factors explain approximately the same proportion of variance in MI, whereas a less strong genetic influence on personality traits emerged. Relevant correlations between MI and specific personality traits (novelty seeking, cooperativeness, self-directedness, self-transcendence) were found, suggesting shared influences for MI and these traits. Both genetic and environmental factors explained these correlations, with genetic factors playing a predominant role. Moderate-to-substantial genetic effects on MI and personality were found. Shared genetic and environmental effects underlie the phenotypic correlation between MI (psychosis-proneness) and personality traits, i.e. self-directedness (negative association) and self-transcendence (positive association), potentially representing predictive markers of psychosis liability related to schizotypy and personality. & 2013 Elsevier Ireland Ltd. All rights reserved.

Keywords: Magical ideation Twins Temperament and character inventory Schizotypy Psychosis Self-directedness Self-transcendence

1. Introduction Psychotic-like experiences or sub-threshold psychotic symptoms, such as perceptual aberrations, paranoid ideation or magical thinking are quite commonly experienced by the general population (Fagnani et al., 2011; Fonseca-Pedrero et al., 2011; Nelson et al., 2013), with an average prevalence of 5% in adults (van Os et al., 2009). Epidemiological surveys across specific subgroups of general population such as adolescents, have nonetheless reported different prevalence for some type of attenuated psychotic symptoms, ranging from 8.4% to 44.1% among different studies, as reported by Fonseca-Pedrero et al. (2011). Such psychotic-like experiences have traditionally been associated to schizophrenia-related personality disorders, like schizotypal, paranoid, and schizoid personality n Correspondence to: Department of Experimental and Clinical Medicine, University of Udine, P.le Kolbe 3, 33100 Udine, Italy. E-mail address: [email protected] (P. Brambilla). 1 The two authors contributed equally to this study.

0165-1781/$ - see front matter & 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.psychres.2013.11.021

disorder (Camisa et al., 2005). In particular, schizotypy, which involves up to 10% of the general population (Nelson et al., 2013), is generally referred to psychosis-liability and attenuated schizophrenia-like traits (Vollema and van den Bosch, 1995). In a phenotypic perspective, schizotypal features include sub-clinical psychotic symptoms like bizarre behavior, magical ideation, social withdrawal/anxiety, lack of feelings, and perceptual abnormalities (Raine, 2006). To a certain extent, it represents a major psychological substrate predisposing to schizophrenia (Fonseca-Pedrero et al., 2011; Nelson et al., 2013) even though only a small group of schizotypal individuals will further develop the disease (Vollema and van den Bosch, 1995). It may therefore be considered as a part of a continuum between psychological health and psychosis vulnerability among the general population (Fonseca-Pedrero et al., 2009). Over the past few decades, schizotypy has been a matter of major interest primarily for translational implications, such as the implementation of preventive strategies for psychosis (McDonald et al., 2005; Bellani et al., 2010; Correll et al., 2010). Likewise, several conceptualizations on its causative factors have been

454

P. Brambilla et al. / Psychiatry Research 215 (2014) 453–459

developed over the past decades. In Meehl's “schizotaxia–schizotypy” model (Meehl, 1962; Kwapil et al., 2008) an ubiquitarious synaptic impairment called hypokrisia, caused by a single major locus (“schizogene”), would underlie specific cerebral abnormalities (“schizotaxia”). This would represent the neural underpinnings of an inherited disposition towards developing psychosis and schizotypal personality traits (Jang et al., 2005). Successively, Tsuang et al. (1999) have revised Meehl's model, defining schizotaxia as the result of the interaction between polygenic and environmental factors (especially pregnancy and neonatal complications) in a complex multifactorial model. These models, differently implying a genetic contribution to schizotypal dimensions and psychosis proneness, have been corroborated by family studies, which revealed that neurological soft signs and cognitive deficits including attention, working memory, and executive functioning are present in first-degree relatives of patients with schizophrenia (Johnson et al., 2003; Solanki et al., 2012), and are associated with schizotypal features (Vollema and Postma, 2002; Johnson et al., 2003). Further evidence of a genetic contribution comes from a number of twin studies (Sullivan et al., 2003; Lataster et al., 2009). Some other twin investigations showed that neuroticism correlates with perceptual and ideational components of schizotypy (Macare et al., 2012) and that the observed relationship between psychotic and personality features is mainly caused by common genetic factors (Jang et al., 2005). In a previous twin study conducted by our group, we showed that genetic factors play a prevalent role in the co-occurrence of psychotic and obsessive traits (Fagnani et al., 2011). Yet, the genetic and environmental influences on the interplay between schizotypal traits and personality in the general population have not been fully elucidated. Specifically, magical ideation is a prominent feature of schizotypal personality and schizophrenia-proneness (Chapman and Chapman, 1980, 1987; Kwapil et al., 1997). It can be identified as “beliefs in forms of causation that by conventional standards are invalid” and can be assessed by the Magical Ideation Scale (MIS) (Eckblad and Chapman, 1983). For example, it puts in causal relation events that are not actually linked, such as thoughts and external events, or supports strong beliefs in paranormal phenomena. Magical ideation is distributed as a continuum from normal population to schizophrenia and does not pertain to psychosis only (Haslam, 2003). This is in line with dimensional psychopathology, which consider the existence of quantitative and not qualitative differences between normal and psychotic experiences (van Praag et al., 1990; Johns and van Os, 2001; van Os, 2003). The main objective of this study was to clarify whether genetic effects on magical ideation are independent of higher-order personality traits. Therefore, we used a multivariate twin design to: (i) explore the genetic and environmental architecture of magical ideation and of personality traits in the Italian general population; (ii) investigate the relationship between magical ideation and personality; (iii) unravel the genetic and environmental bases of this relationship.

From February to November 2010, twins aged 18–65 years, previously enrolled in the ITR, were invited to participate in the survey and were asked to donate a saliva sample to be stored for future genetic studies. In the same mail contact, the twins received the study instruments to be completed and returned. A total of 534 twins aged 18–65 years (mean age ¼40 years, SD ¼ 12 years) corresponding to 267 complete pairs were included in this analysis, leaving out unmatched twins whose data were subjected to parallel psychometric validation analyses. Of the 267 pairs, 161 were monozygotic (MZ; 45 male–male, 116 female–female pairs) and 106 were dizygotic (DZ; 11 male–male, 49 female–female, 46 unlike-gender pairs). Zygosity was assigned by means of a standard questionnaire aimed at assessing the degree of physical similarity of the twins during infancy; this is a wellestablished procedure in twin studies, which is known to have an accuracy level well over 90%. The reliability of this method in the ITR population was recently estimated in an independent sample of 158 same-gender adult twin pairs by using nine microsatellite markers; 149 pairs (94.3%) were correctly classified by the questionnaire. After complete description of the study to the twins, written informed consent was obtained. 2.2. Assessments 2.2.1. Magical ideation Magical ideation was assessed by the self-report Magical Ideation Scale (MIS) (Eckblad and Chapman, 1983), which contains 30 true–false items regarding beliefs in magical influences. It highlights a wide range of subjective magical beliefs such as attenuated forms of thought-transmission experiences, astrology, thought withdrawal, aberrant beliefs like spirit influences, conspiracy theories, UFOs, reincarnation, good luck charms, or the spread of mental energies between people (Chapman and Chapman, 1980). The MIS has been shown to have sound construct validity and internal consistency (Chapman and Chapman, 1985; Norman et al., 1996); furthermore independent longitudinal studies reported that people with high MIS scores have a greater likelihood of developing psychosis (Kwapil et al., 1997; Gooding et al., 2005). In the present study, the Alpha coefficient for this 30-item scale was 0.82. 2.2.2. Personality Personality traits were assessed with the 125-item version of the Temperament and Character Inventory (TCI-125) (Cloninger et al., 1993; Urgesi et al., 2012). It provides information on four temperament dimensions and three character dimensions. The temperament dimensions are: novelty seeking [NS: 20 items measuring exploratory excitability, impulsiveness, extravagance, disorderliness]; harm avoidance [HA: 20 items measuring anticipatory worry and pessimism, fear of uncertainty, shyness with strangers, fatigability and asthenia]; reward dependence [RD: 15 items measuring sentimentality, openness to warm communication, attachment, dependence]; persistence [P: 5 items measuring eagerness of effort, work hardening, ambitious overachieving, perfectionistic perseveration]. The character dimensions are: self-directedness [SD: 25 items measuring ability to control, regulate and adapt one's behavior to concur with existing situations to achieve one's goals and values]; cooperativeness [CO: 25 items measuring identification with and acceptance of others]; self-transcendence [ST: 15 items measuring imaginativeness and spirituality]. In the current study, the Alpha values for NS, HA, RD and P were 0.66, 0.83, 0.62 and 0.61, respectively; they were 0.84, 0.72 and 0.82 for SD, C and ST, respectively. Temperament dimensions (NS, HA, RD and P) refer to the individual-specific behavioral responses to external stimuli and are expected to be highly heritable, relatively stable through lifetime, and genetically determined (Cloninger, 1994; Miettunen et al., 2008). On the other hand, character dimensions (SD, CO, ST) correspond to the representation of the self in relation to other people and to external world. Unlike temperament factors, character factors are mostly prone to change during lifetime, as they are postulated to dynamically result from the complex interaction of several external and internal factors, such as maturation, social learning, cultural and life experiences (Eklund et al., 2004). MIS total score and TCI-125 dimensions scores were calculated for each twin and were used as continuous variables in subsequent analyses.

2. Methods 2.3. Statistical analysis 2.1. Subjects The study sample was derived from the population-based Italian Twin Register (ITR). The procedures that led to the establishment of the ITR are described in detail elsewhere (Stazi et al., 2002). Currently, the ITR contains information on approximately 25,000 twins, and is involved in both general population- and clinical-based studies on various complex phenotypes, with behavioral and psychiatric genetics as major areas of investigation (Brescianini et al., 2013). The study subjects were recruited within a broad mail survey on health and psychological well-being in adulthood in three metropolitan areas of Northern (Milan), Central (Rome) and Southern (Palermo) Italy. The survey was approved by the Ethics Committee of the Italian National Institute of Health (Istituto Superiore di Sanità, Rome, Italy).

2.3.1. Sample statistics For each dimension, descriptive statistics of the scores were computed for twins as individuals and were compared across genders and zygosity groups using robust t-tests (as implemented in Stata, version 9.2) to take the dependence of twin data into account. 2.3.2. Correlations The correlations (i) between the different dimensions within a twin individual (named ‘phenotypic’), (ii) between twin and co-twin for the same dimension (named ‘cross-twin/within-trait’) in MZ and DZ twin pairs separately, and (iii) between one dimension in a twin and another dimension in the co-twin (named ‘cross-twin/cross-trait’) in MZ and DZ twin pairs separately were estimated and

P. Brambilla et al. / Psychiatry Research 215 (2014) 453–459

455

interpreted under the assumptions of the twin design (Neale and Cardon, 1992). A higher cross-twin/within-trait correlation in MZ than in DZ pairs suggests genetic effects on the trait. A significant phenotypic correlation between a pair of traits points to etiological influences common to the traits. A higher cross-twin/crosstrait correlation in MZ compared to DZ pairs reveals genetic factors shared by the traits. To estimate the correlations, the maximum-likelihood method as implemented in the software Mx (Neale et al., 2006) was used, and a saturated model was fitted to raw data from MZ and DZ twins. Gender and age were incorporated as covariates in the model. 2.3.3. Genetic modeling A Cholesky decomposition encompassing additive genetic (A), shared (familial) environmental (C) and unshared (individual-specific) environmental (E) components, and incorporating gender and age as covariates was fitted to MIS total and TCI-125 dimensions scores (Fig. 1) with the software Mx (Neale et al., 2006), using raw data from MZ and DZ twins as input. Additive genetic influences are associated with the average effect of alleles without allelic or gene–gene interaction. Shared environmental influences relate to exposures that are common to all members of a family, thus contributing to within-family aggregation. Unshared environmental factors are specific to an individual, and therefore are responsible for differences between family members; measurement error is also included in this component. Relevant statistics that can be derived from such a decomposition include: (i) ‘heritability’ of each dimension, defined as the proportion of total variance in the dimension explained by genetic effects; (ii) ‘genetic correlation’ between dimensions, that measures the extent to which the dimensions share the same genetic factors; (iii) ‘bivariate heritability’ of dimensions, namely the proportion of phenotypic correlation between the dimensions that is due to shared genetic effects (Neale and Cardon, 1992). Nested models were compared using likelihoodratio χ2 tests. Model fitting started with a full ACE decomposition, and then proceeded with sub-models to test the significance of specific parameters. Parameter estimates were reported under the most parsimonious solution (the bestfitting model).

3. Results 3.1. Sample statistics No significant differences emerged between zygosity groups in the mean MIS total and TCI-125 dimensions scores; this was in agreement with the assumption of twin method that MZ and DZ twins, as individuals, can be regarded as coming from the same reference population. With respect to gender comparison, females scored significantly higher than males on HA, RD, CO, and ST (Table 1). 3.2. Correlations Table 2 shows the correlation structure of MIS total and TCI-125 dimensions. For MIS total and each of the TCI-125 dimensions, cross-twin/within-trait correlation was substantially higher in MZ (range: 0.27–0.51) than in DZ pairs (range:  0.01–0.26), suggesting the contribution of additive genetic factors to individual differences in each dimension. Furthermore, the pattern of MZ vs. DZ correlations provided no evidence for shared environmental effects, except for HA and ST; indeed, according to quantitative genetic theory, these effects are evoked when twice DZ correlation exceeds MZ correlation. MIS total score was negatively correlated with the score of SD (  0.37) and CO (  0.22), and positively correlated with the score of NS (0.22) and ST (0.59). Cross-twin/cross-trait correlations of MIS total with the aforementioned TCI-125 dimensions were higher in MZ compared to DZ pairs, revealing that overlapping genetic effects may underlie the phenotypic correlations. 3.3. Genetic modeling Likelihood-ratio χ2 tests between nested models were performed. As a first test, dropping the shared environmental component from the full ACE Cholesky model showed a non-significant fit deterioration

Fig. 1. Path diagram of the Cholesky decomposition for MIS and TCI dimensions in one twin. Squares denote observed scores. Circles indicate latent sources of variance and covariance. Ai and Ei represent additive genetic and unshared environmental influences, respectively. Latent genetic factors correlate 1 between monozygotic twins and 0.5 between dizygotic twins. Although model fitting also included shared environmental influences (Ci), the corresponding latent sources were not shown in the diagram for reasons of clarity. MIS, magical ideation scale; TCI, temperament and character inventory; NS, novelty seeking; HA, harm avoidance; RD, reward dependence; P, persistence; SD, self-directedness; CO, cooperativeness; and ST, self-transcendence.

(χ 236 ¼8.93, p¼1.00). Then, genetic effects common to the dimensions were tested in the AE model by setting the genetic correlations between the dimensions to zero; this produced a highly significant change in the fit of the AE model (χ 228 ¼127.47, p¼1.2  10  14). Similarly, unshared environmental correlations could not be dropped from the AE model (χ 228 ¼ 159.60, p¼2.2  10  20). Therefore, the bestfitting model included additive genetic and individual-specific environmental effects on the dimensions, possibly shared between them. Genetic and unshared environmental effects for MIS total and TCI-125 dimensions, as estimated from the best-fitting AE Cholesky model, are reported in Table 3. About half of the variation in MIS total score was due to genetic effects, while heritability estimates for TCI-125 traits ranged from 0.22 (P) to 0.50 (SD). The genetic correlation between MIS total and each of NS, SD and CO was moderate, whereas that between MIS total and ST was high; shared genetic effects explained from 54% (for MIS and NS) to 93% (for MIS and CO) of phenotypic correlation between MIS total and the aforementioned TCI-125 dimensions. The overlap between individual-specific environmental factors – as indicated by unshared environmental correlations – was more modest; however, this overlap still explained a substantial proportion of phenotypic correlation between MIS total and TCI125 dimensions.

4. Discussion The main purpose of this population-based twin study was to estimate the effects of genetic and environmental factors on magical ideation (a core feature of psychosis-proneness), personality traits, and the relationship between them. Although subthreshold psychotic symptoms are quite common in the general population and well documented (Kendler et al., 1996; Scott et al., 2008; van Os et al., 2009), this is the first twin study, to our best knowledge, aimed to investigate the genetic and environmental bases of the interplay between these symptoms and personality traits as conceptualized in Cloninger's model.

456

P. Brambilla et al. / Psychiatry Research 215 (2014) 453–459

Table 1 Descriptive statistics of MIS and TCI scores. Gender

Zygosity

Males

MIS TCI NS TCI HA TCI RD TCI P TCI SD TCI CO TCI ST

Females

Monozygotic

Dizygotic

N

Mean 7S.D.

N

Mean 7 S.D.

N

Mean 7 S.D.

N

Mean 7S.D.

130 154 154 155 155 149 150 151

5.277 4.61 8.477 3.76 8.60 7 3.73 7.99 7 2.57 3.017 1.54 18.137 4.75 18.487 4.04 4.727 3.26

326 357 354 363 364 356 358 354

6.067 4.63 8.447 3.44 10.46 7 4.09* 9.45 7 2.60* 3.117 1.52 17.55 7 5.12 19.517 3.58† 5.58 7 3.4†5

276 306 303 311 309 303 303 302

5.917 4.70 8.247 3.50 10.067 4.07 9.20 7 2.62 3.127 1.50 17.87 7 5.01 19.38 7 3.61 5.487 3.47

180 205 205 207 210 202 205 203

5.717 4.54 8.767 3.58 9.667 4.08 8.747 2.74 3.017 1.57 17.497 5.03 18.95 7 3.94 5.08 7 3.32

MIS, magical ideation scale; TCI, temperament and character inventory; NS, novelty seeking; HA, harm avoidance; RD, reward dependence; P, persistence; SD, self-directedness; CO, cooperativeness; and ST, self-transcendence. n

†

p o 0.01 for gender difference in mean. 0.01 op o 0.05 for gender difference in mean.

Table 2 Correlations for MIS and TCI. TCI NS

HA

P

SD

CO

ST

Cross-twin/within-trait correlations MIS MZ 0.51 0.37 0.44 0.30 DZ 0.26 0.06 0.26  0.01

0.27 0.01

0.51 0.21

0.49 0.19

0.37 0.25

Phenotypic correlations MIS 0.22

 0.09

0.09

 0.37

 0.22

0.59

0.00  0.02

0.08  0.11

 0.23  0.16

 0.19  0.12

0.38 0.16

0.06

Cross-twin/cross-trait correlations MIS MZ 0.12 0.04 DZ 0.08  0.01

RD

63 DZ). This may lead to differences in heritability estimates if true genetic influence changes with age or if there are gender-specific genetic effects, which may not be properly modeled. Conversely, heritability estimates for TCI-125 traits ranged from 0.22 (P) to 0.50 (SD), suggesting an overall less strong genetic influence. Nevertheless, previous twin studies reported substantial genetic effects on temperament as well as on character dimensions with no strong differences in levels of heritability (Gillespie et al., 2003; Ando et al., 2004). As recently reported, a biological mechanism underlying genetic effects on personality traits may encompass both additive genetic variants and non-additive genetic effects such as a combination of dominance and epistasis (Verweij et al., 2012). 4.2. Environmental effects on magical ideation and personality traits

MIS, magical ideation scale; TCI, temperament and character inventory; NS, novelty seeking; HA, harm avoidance; RD, reward dependence; P, persistence; SD, selfdirectedness; CO, cooperativeness; ST, self-transcendence; MZ, monozygotic; and DZ, dizygotic.

4.1. Genetic contribution to magical ideation and personality traits The best-fitting multivariate Cholesky model shows that MIS and TCI-125 scores are mainly influenced by additive genetic (A) and unshared environmental (E) factors, with no shared environmental effects. For MIS, genetic proportion of total variance was estimated at 51%, suggesting that genetic and individual-specific environmental factors explain approximately the same proportion of interindividual phenotypic differences. This finding is consistent with the literature showing that schizotypal symptoms are common among relatives of patients with schizophrenia, although the association seems to be stronger for negative symptoms than for positive symptoms (Faraone et al., 2001). In accordance to our study, Kendler and Hewitt (1992) and Miller (1993) reported a substantial genetic influence on magical ideation (even over 50%), with little influence of shared environment. However, using a univariate model, MacDonald et al. (2001) showed that magical ideation was significantly influenced by shared and unshared environmental components which explained, respectively, 41% and 59% of total variance, with no genetic contribution. Discrepancies among these studies may arise from age and gender differences in study sample. Indeed, in MacDonald et al. (2001) same sex young adults with small age range were included (mean age¼ 21.472.9 years; 98 MZ twin pairs of which 53 females and 45 males; 59 same-sex DZ of which 33 females and 26 males). Older twins with larger age range also including opposite-sex DZ pairs were recruited in the study of Kendler and Hewitt (1992) (mean age¼37.7713 years; 70 MZ and

The best-fitting multivariate Cholesky model shows that the environmental contribution is exclusively accounted for by unshared environmental effects (E), in accordance to prior twin studies (Kendler and Hewitt, 1992; Miller, 1993). These effects reflect the influence of risk factors that are related to individual experiences (for instance, smoking, substance abuse, childhood adverse experiences) as well as to measurement errors (Fagnani et al., 2011). Interestingly, MacDonald et al. (2001) found that magical ideation was mainly influenced by shared (familial) environmental factors; however, this study was based on a younger twin population, and thus shared environmental influences, which generally tend to decline with ageing, may have been easier to detect. The best-fitting Cholesky model also shows that a predominant portion of total variance for personality traits is explained by environmental effects, which are entirely unshared by the twins. Accordingly to our results, previous twin studies reported that unshared environment plays a pivotal role in shaping personality traits, whereas shared environment plays a minor, almost negligible, influence (Ando et al., 2002; Keller et al., 2005). Nevertheless, the effects of both unshared and shared environment, encompassing individual and familiar stressful experiences and adverse life events or substances exposure, have greatly been debated and still remains a matter of major interest. In addition, a possible biological underpinning of environmental effects on personality traits modulation and development emerged from the field of epigenetics. In this regard, it has been recently proposed that several external stressors (social, situational and chemical) are involved in influencing the development of behavioral and personality traits through genome-expression

P. Brambilla et al. / Psychiatry Research 215 (2014) 453–459

457

Table 3 Genetic and unshared environmental effects for MIS and TCI, as estimated from the best-fitting AE Cholesky model. TCI NS

HA

RD

P

SD

CO

ST

Genetic (A) and unshared environmental (E) proportions of variance MIS A 0.51 0.35 0.45 0.27 E 0.49 0.65 0.55 0.73

0.22 0.78

0.50 0.50

0.48 0.52

0.39 0.61

Genetic (rA) and unshared environmental (rE) correlations MIS rA 0.28 0.06 rE 0.18 0.05

0.14 0.08

 0.46  0.28

 0.41  0.03

0.82 0.41

0.62 0.38

0.93 0.07

0.62 0.38

 0.05  0.12

Genetic (A) and unshared environmental (E) proportions of phenotypic correlations MIS A 0.54 0.52 0.18 0.49 E 0.46 0.48 0.82 0.51

MIS, magical ideation scale; TCI, temperament and character inventory; NS, novelty seeking; HA, harm avoidance; RD, reward dependence; P, persistence; SD, self-directedness; CO, cooperativeness; and ST, self-transcendence. Genetic proportions of phenotypic correlations are ‘bivariate heritabilities’.

regulation, in a multifaceted scenario of GxE interaction (see Svrakic and Cloninger, 2010). 4.3. Correlations among MIS and TCI-125 dimensions MIS total score showed relevant phenotypic correlations with one temperament dimension (i.e. novelty seeking, NS, 0.22) and three character dimensions (i.e. cooperativeness, CO,  0.22; self-directedness, SD,  0.37; self-transcendence, ST, 0.59). The negative correlation between MIS and CO ( 0.22) seems to be almost exclusively driven by genetic effects common to those two psychological dimensions. Indeed, there is a substantial genetic correlation (0.41) and a very large bivariate heritability (i.e. genetic proportion of phenotypic correlation, 0.93). As expected, a negative phenotypic correlation has been detected between MIS and SD ( 0.37), the latter being a major proxy of psychological well-being. Such correlation is mainly genetic in origin (genetic correlation:  0.46; bivariate heritability: 0.62), but overlapping environmental effects also play a role (environmental correlation:  0.28; environmental proportion of correlation: 0.38). Interestingly, a similar finding of substantial genetic overlap was recently reported (Macare et al., 2012) for perceptual and ideational components of schizotypy and the high-order personality trait neuroticism, which shares many similarities with low SD. The positive phenotypic correlation between MIS and ST (0.59) arises from a large genetic overlap between the two dimensions (genetic correlation: 0.82) and from a more modest, though still substantial, environmental overlap (environmental correlation: 0.41). These overlapping genetic and environmental factors explain exactly the same proportions of phenotypic correlation as in the negative correlation between MIS and SD. With respect to MIS and NS, genetic and environmental factors influencing the two dimensions are suggested to be distinct (weak genetic and environmental correlation). The few overlapping genetic and environmental factors explain similar proportions of phenotypic correlation. In summary, these phenotypic correlations suggest the existence of underlying genetic and environmental etiological factors in common for magical ideation and specific personality traits (i.e. SD, CO, ST and NS). However, the genetic overlap (and its impact in terms of observed correlation) between magical ideation and personality traits seems to prevail on the environmental overlap, especially for CO, ST and SD, though with differences in magnitude. Hypothetically, the positive correlation between MI and ST might identify the phenotypic relationship with creativity, spirituality, and mysticsm, ultimately leading to unusual feelings, thoughts and behaviors. Likewise, it has similarly been shown that schizotypal

personality traits, assessed with the Schizotypal Personality Questionnaire (SPQ), and especially “positive-schizotypy”, including odd beliefs/magical thinking, are positively associated to ST character dimension (Hori et al., 2014). In contrast, the negative correlation with SD and CO could reflect characteristics of the schizotypal personality organization, such as low socialization, tolerance, and cooperativeness. In this regard, previous studies largely supported the evidence of a negative correlation between SPQ score and SD and CO in schizotypal personality disorder (Hori et al., 2014).

4.4. Limitations and strengths Some methodological issues, which could potentially affect the reliability of our findings and their interpretation, need to be considered. First, this study relied exclusively on self-report measures, which may introduce some degree of bias in the assessment of targeted symptoms. Such biases are typically captured as measurement errors in twin models, thus contributing to the magnitude of unshared environmental effects. Second, a relatively small sample may not have provided enough power to assess shared environmental effects on the dimensions, and did not allow us to test further hypotheses, such as gender differences in genetic and environmental effects. Third, the use of unlike-gender twin pairs may have reduced DZ correlation, and this may have ultimately led to an overestimation of heritability, as gender-specific genetic effects could not be modeled. Fourth, a low response rate (around 30%) was achieved, which is however in line with all previous studies conducted by the Italian Twin Register in the area of behavior and psychiatric genetics. This rate may partially reflect cultural reluctance towards the use of a nationwide public health instrument to openly address the issue of genetic and environmental determinants of mental health. It is unlikely that results were affected by selection biases, because no differences were detected between respondents and non-respondents in terms of age, gender, education and geographical area of residence. Points of strength of the current study include the enrollment of a sample from the community rather than from a specific subgroup of population, such as undergraduate students that generally have higher schizotypy scores than general population samples (Chmielewski et al., 1995). Furthermore, the wider age range of our sample increases the representativeness of the target population, thus allowing the generalization of our results.

458

P. Brambilla et al. / Psychiatry Research 215 (2014) 453–459

5. Conclusions This is the first twin study exploring the relationship between magical ideation (the core feature of psychosis-proneness) and temperament and character as conceptualized in Cloninger's model. Moderate to substantial genetic effects on magical ideation and on specific temperament and character dimensions were found. Shared genetic and environmental effects underlie the phenotypic correlation between tendency to magical ideation and some personality traits, especially self-directedness (negative association) and selftranscendence (positive association). In particular, self-directedness relates to individual's competence toward autonomy, reliability, and maturity, while self-transcendence characterizes the attitude toward mysticism, religion, and idealism. They could therefore represent predictive markers of psychosis liability related to schizotypy and personality. However, our results need to be replicated by larger and prospective twin investigations, including child and adolescent subjects. Such studies will allow to further understand the relative role of genetic and environmental factors in the interplay between psychosis-proneness and personality during development.

Acknowledgments Dr. Paolo Brambilla was partly supported by grants from the Italian Ministry of Health (GR-2010-2316745) and by the BIAL Foundation (Fellowship #262/12). References Ando, J., Ono, Y., Yoshimura, K., Onoda, N., Shinohara, M., Kanba, S., Asai, M., 2002. Rhe genetic structure of Cloninger's seven-factor model of temperament and character in a Japanese sample. Journal of Personality 70, 583–609. Ando, J., Suzuki, A., Yamagata, S., Kijima, N., Maekawa, H., Ono, Y., Jang, K.L., 2004. Genetic and environmental structure of Cloninger's temperament and character dimensions. Journal of Personality Disorders 18, 379–393. Bellani, M., Cerruti, S., Brambilla, P., 2010. Orbitofrontal cortex abnormalities in schizophrenia. Epidemiologia e Psichiatria Sociale 19, 23–25. Brescianini, S., Fagnani, C., Toccaceli, V., Medda, E., Nisticò, L., D'Ippolito, C., Alviti, S., Arnolfi, A., Caffari, B., Delfino, D., Ferri, M., Penna, L., Salemi, M., Sereni, S., Serino, L., Cotichini, R., Stazi, M.A., 2013. An update on the Italian Twin Register: advances in cohort recruitment, project building and network development. Twin Research and Human Genetics 16, 190–196. Camisa, K.M., Bockbrader, M.A., Lysaker, P., Rae, L.L., Brenner, C.A., O'Donnell, B.F., 2005. Personality traits in schizophrenia and related personality disorders. Psychiatry Research 133, 23–33. Chapman, L.J., Chapman, J.P., 1980. Scales for rating psychotic and psychotic-like experiences as continua. Schizophrenia Bulletin 6, 477–489. Chapman, L.J., Chapman, J.P., 1985. Psychosis proneness. In: Alpert, M. (Ed.), Controversies in Schzophrenia: Changes and Constancies, 1st edition. Guilford Press, New York, NY, pp. 157–174. Chapman, L.J., Chapman, J.P., 1987. The search for symptoms predictive of schizophrenia. Schizophrenia Bulletin 13, 497–503. Chmielewski, P.M., Fernandas, L.O.L., Yee, C.M., Miller, G.A., 1995. Ethnicity and gender in scales of psychosis proneness and mood disorders. Journal of Abnormal Psychology 104, 464–470. Cloninger, C.R., 1994. Temperament and personality. Current Opinion in Neurobiology 4, 266–273. Cloninger, C.R., Svrakic, D.M., Przybeck, T.R., 1993. A psychobiological model of temperament and character. Archives of General Psychiatry 50, 975–990. Correll, C.U., Hauser, M., Auther, A.M., Cornblatt, B.A., 2010. Research in people with psychosis risk syndrome: a review of the current evidence and future directions. Journal of Child Psychology and Psychiatry and Allied Disciplines 51, 390–431. Eckblad, M., Chapman, L.J., 1983. Magical ideation as an indicator of schizotypy. Journal of Consulting and Clinical Psychology 51, 215–225. Eklund, M., Hansson, L., Bengtsson-Tops, A., 2004. The influence of temperament and character on functioning and aspects of psychological health among people with schizophrenia. European Psychiatry: the Journal of the Association of European Psychiatrists 19, 34–41. Fagnani, C., Bellani, M., Tansella, M., Balestrieri, M., Toccaceli, V., Patriarca, V., Stazi, M.A., Brambilla, P., 2011. Investigation of shared genetic effects for psychotic and obsessive symptoms in young adult twins. Psychiatry Research 188, 276–282. Faraone, S.V., Green, A.I., Seidman, L.J., Tsuang, M.T., 2001. “Schizotaxia”: Clinical implications and new directions for research. Schizophrenia Bulletin 27, 1–18.

Fonseca-Pedrero, E., Paíno-Piñeiro, M., Lemos-Giráldez, S., Villazón-García, U., Muñiz, J., 2009. Validation of the schizotypal personality questionnaire – brief form in adolescents. Schizophrenia Research 111, 53–60. Fonseca-Pedrero, E., Santarén-Rosell, M., Lemos-Giráldez, S., Paino, M., SierraBaigrie, S., Muñiz, J., 2011. Psychotic-like experiences in the adolescent general population. Actas Españolas de Psiquiatría 39, 155–162. Gooding, D.C., Tallent, K.A., Matts, C.W., 2005. Clinical status of at-risk individuals 5 years later: further validation of the psychometric high-risk strategy. Journal of Abnormal Psychology 114, 170–175. Gillespie, N.A., Cloninger, C.R., Heath, A.C., Martin, N.G., 2003. The genetic and environmental relationship between Cloninger's dimensions of temperament and character. Personality and Individual Differences 35, 1931–1946. Haslam, N., 2003. Categorical versus dimensional models of mental disorder: the taxometric evidence. Australian and New Zealand Journal of Psychiatry 37, 696–704. Hori, H., Teraishi, T., Sasayama, D., Matsuo, J., Kinoshita, Y., Ota, M., Hattori, K., Kunugi, H., 2014. A latent profile analysis of schizotypy, temperament and character in a nonclinical population: association with neurocognition. Journal of Psychiatric Research 48, 56–64. Jang, K.L., Woodwards, T.S, Lang, D., Honer, W.G., Livesley, W.J., 2005. The genetic and environmental basis of the relationship between schizotypy and personality: a twin study. Journal of Nervous and Mental Disease 193, 153–159. Johns, L.C., van Os, J., 2001. The continuity of psychotic experiences in the general population. Clinical Psychology Review 21, 1125–1141. Johnson, J.K., Tuulio-Henriksson, A., Pirkola, T., Huttunen, M.O., Lönnqvist, J., Kaprio, J., Cannon, T.D., 2003. Do schizotypal symptoms mediate the relationship between genetic risk for schizophrenia and impaired neuropsychological performance in co-twins of schizophrenic patients? Biological Psychiatry 54, 1200–1204. Keller, M.C., Coventry, W.L., Heath, A.C., Martin, N.G., 2005. Widespread evidence for non-additive genetic variation in Cloninger’s and Eysenck’s personality dimensions using a twin plus sibling design. Behavior Genetics 35, 707–721. Kendler, K.S., Gallagher, T.J., Abelson, J.M., Kessler, R.C., 1996. Lifetime prevalence, demographic risk factors, and diagnostic validity of nonaffective psychosis as assessed in a US community sample. The National Comorbidity Survey. Archives of General Psychiatry 53, 1022–1031. Kendler, K.S., Hewitt, J., 1992. The structure of self-report schizotypy in twins. Journal of Personality Disorders 6, 1–17. Kwapil, T.R., Barrantes-Vidal, N., Silvia, P.J., 2008. The dimensional structure of the Wisconsin Schizotypy Scales: factor identification and construct validity. Schizophrenia Bulletin 34, 444–457. Kwapil, T.R., Miller, M.B., Zinser, M.C., Chapman, J., Chapman, L.J., 1997. Magical ideation and social anhedonia as predictors of psychosis proneness: a partial replication. Journal of Abnormal Psychology 106, 491–495. Lataster, T., Wichers, M., Jacobs, N., Mengelers, R., Derom, C., Thiery, E., van Os, J., MyinGermeys, I., 2009. Does reactivity to stress cosegregate with subclinical psychosis? A general population twin study. Acta Psychiatrica Scandinavica 119, 45–53. Macare, C., Bates, T.C., Heath, A.C., Martin, N.G., Ettinger, U., 2012. Substantial genetic overlap between schizotypy and neuroticism: a twin study. Behavior Genetics 42, 732–742. MacDonald, A.W., Pogue-Geile, M.F., Debski, T.T., Manuck, S., 2001. Genetic and environmental influences on schizotypy: a community-based twin study. Schizophrenia Bulletin 27, 47–58. McDonald, C., Bullmore, E., Sham, P., Chitnis, X., Suckling, J., MacCabe, J., Walshe, M., Murray, R.M., 2005. Regional volume deviations of brain structure in schizophrenia and psychotic bipolar disorder: computational morphometry study. British Journal of Psychiatry: the Journal of Mental Science 186, 369–377. Meehl, P.E., 1962. Schizotaxia, schizotypy, schizophrenia. American Psychologist 17, 827–838. Miettunen, J., Lauronen, E., Kantojärvi, L., Veijola, J., Joukamaa, M., 2008. Intercorrelations between Cloninger's temperament dimensions – a meta-analysis. Psychiatry Research 160, 106–114. Miller, M.B., 1993. A study of schizotypal traits in young male twins. University of Wisconsin-Madison, Madison, WI. Neale, M.C., Boker, S.M., Xie, G., Maes, H., 2006. Mx: Statistical Modeling, 7th ed Richmond, Virginia. Neale, M.C., Cardon, L.R., 1992. Methodology for Genetic Studies of Twins and Families. Dordrecht, The Netherlands. Nelson, M.T., Seal, M.L., Pantelis, C., Phillips, L.J., 2013. Evidence of a dimensional relationship between schizotypy and schizophrenia: a systematic review. Neuroscience and Biobehavioral Reviews 37, 317–327. Norman, R.M., Davies, F., Malla, A.K., Cortese, L., Nicholson, I.R., 1996. Relationship of obsessive-compulsive symptomatology to anxiety, depression and schizotypy in a clinical population. British Journal of Clinical Psychology – the British Psychological Society 35 (Pt 4), 553–566. Raine, A., 2006. Schizotypal personality: neurodevelopmental and psychosocial trajectories. Annual Review of Clinical Psychology 2, 291–326. Scott, J., Welham, J., Martin, G., Bor, W., Najman, J., O'Callaghan, M., Williams, G., Aird, R., McGrath, J., 2008. Demographic correlates of psychotic-like experiences in young Australian adults. Acta Psychiatrica Scandinavica 118, 230–237. Solanki, R.K., Swami, M.K., Singh, P., Gupta, S., 2012. Identification of vulnerability among first-degree relatives of patients with schizophrenia. East Asian Archives of Psychiatry: Official Journal of the Hong Kong College of Psychiatrists ¼Dong Ya jing shen ke xue zhi: Xianggang jing shen ke yi xue yuan qi kan 22, 118–125. Stazi, M.A., Cotichini, R., Patriarca, V., Brescianini, S., Fagnani, C., D'Ippolito, C., Cannoni, S., Ristori, G., Salvetti, M., 2002. The Italian Twin Project: from the

P. Brambilla et al. / Psychiatry Research 215 (2014) 453–459

personal identification number to a national twin registry. Twin Research 5, 382–386. Sullivan, P.F., Kendler, K.S., Neale, M.C., 2003. Schizophrenia as a complex trait: evidence from a meta-analysis of twin studies. Archives of General Psychiatry 60, 1187–1192. Svrakic, D.M., Cloninger, R.C., 2010. Epigenetic perspective on behavior development, personality, and personality disorders. Psychiatria Danubina 22, 153–166. Tsuang, M.T., Stone, W.S., Seidman, L.J., Faraone, S.V., Zimmet, S., Wojcik, J., Kelleher, J.P., Green, A.I., 1999. Treatment of nonpsychotic relatives of patients with schizophrenia: four case studies. Biological Psychiatry 45, 1412–1418. Urgesi, C., Romanò, M., Fornasari, L., Brambilla, P., Fabbro, F., 2012. Investigating the development of temperament and character in school-aged children using a self-report measure. Comprehensive Psychiatry 53, 875–883. van Os, J., 2003. Is there a continuum of psychotic experiences in the general population? Epidemiologia e Psichiatria Sociale 12, 242–252. van Os, J., Linscott, R.J., Myin-Germeys, I., Delespaul, P., Krabbendam, L., 2009. A systematic review and meta-analysis of the psychosis continuum: evidence for

459

a psychosis proneness-persistence-impairment model of psychotic disorder. Psychological Medicine 39, 179–195. van Praag, H.M., Asnis, G.M., Kahn, R.S., 1990. Nosological tunnel vision in biological psychiatry. Annals of the New York Academy of Sciences 600, 501–510. Verweij, K.J., Yang, J., Lahti, J., Veijola, J., Hintsanen, M., Pulkki-Råback, L., Heinonen, K., Pouta, A., Pesonen, A.K., Widen, E., Taanila, A., Isohanni, M., Miettunen, J., Palotie, A., Penke, L., Service, S.K., Heath, A.C., Montgomery, G.W., Raitakari, O., Kähönen, M., Viikari, J., Räikkönen, K., Eriksson, J.G., Keltikangas-Järvinen, L., Lehtimäki, T., Martin, N.G., Järvelin, M.R., Visscher, P.M., Keller, M.C., Zietsch, B. P., 2012. Maintenance of genetic variation in human personality: testing evolutionary models by estimating heritability due to common causal variants and investigating the effect of distant inbreeding. Evolution 66, 3238-3251. Erratum in Evolution 67, 1537. Vollema, M.G., van den Bosch, R.J., 1995. The multidimensionality of schizotypy. Schizophrenia Bulletin 21, 19–31. Vollema, M.G., Postma, B., 2002. Neurocognitive correlates of schizotypy in first degree relatives of schizophrenia patients. Schizophrenia Bulletin 28, 367–377.