Classical Twin Design Research Articles

The aetiology of educational attainment: A nuclear twin family study into the genetic and environmental influences on school leaving certificates.

Educational attainment is connected to many important life outcomes, and the previous research has already focused on identifying its genetic and environmental components. However, most of these studies used twin data only and did not incorporate information from other family members. Twin studies typically decompose the phenotypic variance into genetic, shared, and unique environment components. In this study design, the shared environment component encompasses the influence of parents and the shared environments of twins and siblings independent of parents (e.g., teachers, schools, and peers). The classical twin design (CTD) conflates these influences as part of the shared environment. This shortcoming can be overcome using the nuclear twin family design (NTFD), which enables separation of the parental and shared twin/sibling environmental components. The aim of this study was to broaden the understanding of the aetiology of educational attainment using the nuclear twin family design to provide a detailed account of the genetic and environmental effects on the type of school leaving certificate. The data of 1,048 monozygotic and 916 dizygotic same-sex twins, their biological parents, and non-twin full biological siblings of the German project TwinLife were used in the nuclear twin family design. Structural equationmodelling (SEM) techniques were used to analyse the variance-covariance patterns of the ordinal outcome variable. Genetic influences were found to make up around 60% of variance, whilst environmental influences shared by all siblings, educational influences shared by the twins only, and non-shared environmental influences explained the remaining variance in equal amounts. Environmental transmission from parent to offspring was found to be negligible. Through its advanced design, our study extends the previous work enabling more detailed and robust estimations of sources of variance and contributes to a better understanding of the complex aetiology of educational attainment.

Actigraphic sleep and cortisol in middle childhood: A multivariate behavioral genetics model

To date, behavioral genetic studies investigated either sleep or cortisol levels in middle childhood, but not both simultaneously. Therefore, a pertinent question is the degree to which genetic factors and environmental factor contribute to the correlation between sleep and cortisol levels. To address this question, we employed the classical twin design. We measured sleep in 6-9-year-old twins (N = 436 twin pairs, “Together Unique” study) over four consecutive nights using actigraphy, and we measured morning cortisol on two consecutive days. Sleep duration, sleep efficiency, and wake episodes were used as indicators of sleep. Morning cortisol level was used as cortisol indicator. A structural equation model was fitted to estimate the contribution of additive genetic effects (A), shared (common) environmental effects, (C) and unique environmental effects (E) to phenotypic variances and covariances. Age, cohort, and sex were included as covariates. The heritability of sleep duration, sleep efficiency, and wake episodes were 52%, 45%, and 55%, respectively. Common environmental factors played no significant role. High genetic correlations between sleep duration and sleep efficiency and high genetic correlations between sleep efficiency and wake episodes were found. Shared environmental (29%) and unique environmental factors (53%) explained the variance in morning cortisol levels. Because the sleep and cortisol measures were found to be uncorrelated, we did not consider genetic and environmental contributions to the association between the sleep and cortisol measures. Our findings indicate that sleep duration, sleep efficiency, and wake episodes in children are mostly impacted by genetic factors and by unique environmental factors (including measurement error).

The Causes of Individual Differences in Autism Spectrum Disorder

Autism spectrum disorder (ASD) is characterized by various symptoms including impaired social interactions, unusually repetitive behaviors, and highly restricted interests etc. People with ASD differ significantly on their clinical profiles and the causes of such individual differences are not yet fully understood. The present paper provides an overview of the causes of individual differences in ASD from three different perspectives: genetic, environmental, and neurobiological perspectives. The present paper also describes one study design in detail within each perspective (i.e., classical twin design, epidemiological case-control design, and magnetic resonance imaging), and explains how each study design is informative about the causes of ASD.

Trait-specific testing of the equal environment assumption: The case of school grades and upper secondary school attendance

Objective: This paper tests the equal environment assumption for school grades and upper secondary school attendance and describes the conditions under which violations are problematic.
 Background: A growing number of sociologists use twin-based research designs, particularly the Classical Twin Design (CTD), to differentiate between genetic and social causes of social inequalities. One key assumption of CTD is that environmental influences are shared by monozygotic and dizygotic twins to the same extent; called the equal environment assumption (EEA). This assumption is frequently contested and the target of concern, because violation can result in an overestimation of heritability and an underestimation of the role of the social environment.
 Method: Using data from the first wave of the German TwinLife study, the paper illustrates two approaches to test EEA for school grades and enrolment in upper secondary school (Gymnasium). The analysis is based on a sample of twins (N = 1,576) aged ten to twelve years.
 Results: The results show that the approaches are able to detect violations of EEA (though in different ways), depending on the environmental variables that might causally be involved in trait variance. Only in one case was a violation was observed; it had no effect on heritability estimates.
 Conclusion: While EEA holds for school grades, violations do not automatically invalidate CTD in case of upper secondary school attendance.

Examining the association between childhood autistic traits and adolescent hypomania: a longitudinal twin study.

There is evidence that autism spectrum disorders (ASDs) co-occur with bipolar disorder (BD) relatively frequently. Individuals with BD often report symptoms of mania and hypomania during adolescence, prior to the age of onset for BD. It is unknown whether these symptoms are associated with ASDs. We examined whether diagnoses of ASDs and autistic traits were associated with hypomania in a large, population-based Swedish twin sample. Parental structured interviews assessed autistic traits, and were used to assign screening diagnoses of ASDs, when twins were aged 9 or 12 (N = 13 533 pairs). Parents then completed questionnaires assessing hypomania when the twins were aged 15 and 18 (N = 3852 pairs at age 15, and 3013 pairs at age 18). After investigating the phenotypic associations between these measures, we used the classical twin design to test whether genetic and environmental influences on autistic traits influence variation in adolescent hypomania. Autistic traits and ASD diagnoses in childhood were associated with elevated scores on the measures of adolescent hypomania. Twin analyses indicated that 6-9% of the variance in hypomania was explained by genetic influences that were shared with autistic traits in childhood. When repeating these analyses for specific autistic trait domains, we found a stronger association between social interaction difficulties and hypomania than for other autistic trait domains. These results indicate a genetic link between autistic traits and hypomania in adolescence. This adds to the growing evidence base of genetic factors associated with ASDs showing links with psychiatric outcomes across childhood and into adulthood.

The Augmented Classical Twin Design: Incorporating Genome-Wide Identity by Descent Sharing Into Twin Studies in Order to Model Violations of the Equal Environments Assumption.

The Classical Twin Method (CTM) compares the similarity of monozygotic (MZ) twins with that of dizygotic (DZ) twins to make inferences about the relative importance of genes and environment in the etiology of individual differences. The design has been applied to thousands of traits across the biomedical, behavioral and social sciences and is arguably the most widely used natural experiment known to science. The fundamental assumption of the CTM is that trait relevant environmental covariation within MZ pairs is the same as that found within DZ pairs, so that zygosity differences in within-pair variance must be due to genetic factors uncontaminated by the environment. This equal environments assumption (EEA) has been, and still is hotly contested, and has been mentioned as a possible contributing factor to the missing heritability conundrum. In this manuscript, we introduce a new model for testing the EEA, which we call the Augmented Classical Twin Design which uses identity by descent (IBD) sharing between DZ twin pairs to estimate separate environmental variance components for MZ and DZ twin pairs, and provides a test of whether these are equal. We show through simulation that given large samples of DZ twin pairs, the model provides unbiased estimates of variance components and valid tests of the EEA under strong assumptions (e.g. no epistatic variance, IBD sharing in DZ twins estimated accurately etc.) which may not hold in reality. Sample sizes in excess of 50,000 DZ twin pairs with genome-wide genetic data are likely to be required in order to detect substantial violations of the EEA with moderate power. Consequently, we recommend that the Augmented Classical Twin Design only be applied to datasets with very large numbers of DZ twin pairs (> 50,000 DZ twin pairs), and given the strong assumptions relating to the absence of epistatic variance, appropriate caution be exercised regarding interpretation of the results.

Heritability in Friendship Networks

There is considerable evidence nowadays that friendship networks account for a large part of an individual's success or failure in life. Little, however, is known about the extent to which friendship networks are associated with an individual's genotype. Using data from the German TwinLife study, we explore, within a classical twin design, whether friendship networks are related to genes. We find a substantial heritability component in twins' network sizes and network homophily, but not in twins' network closeness. The genetic influence on network characteristics may be attributable to traits which are themselves influenced by genetic factors. Addressing indirect ways in which genes could influence network characteristics, we do not find evidence that shared hobbies or Big Five personality traits affect networks.

Using Multimodel Inference/Model Averaging to Model Causes of Covariation Between Variables in Twins.

To explore and apply multimodel inference to test the relative contributions of latent genetic, environmental and direct causal factors to the covariation between two variables with data from the classical twin design by estimating model-averaged parameters. Behavior genetics is concerned with understanding the causes of variation in phenotypes and the causes of covariation between two or more phenotypes. Two variables may correlate as a result of genetic, shared environmental or unique environmental factors contributing to variation in both variables. Two variables may also correlate because one or both directly cause variation in the other. Furthermore, covariation may result from any combination of these sources, leading to 25 different identified structural equation models. OpenMx was used to fit all these models to account for covariation between two variables collected in twins. Multimodel inference and model averaging were used to summarize the key sources of covariation, and estimate the magnitude of these causes of covariance. Extensions of these models to test heterogeneity by sex are discussed. We illustrate the application of multimodel inference by fitting a comprehensive set of bivariate models to twin data from the Virginia Twin Study of Psychiatric and Substance Use Disorders. Analyses of body mass index and tobacco consumption data show sufficient power to reject distinct models, and to estimate the contribution of each of the five potential sources of covariation, irrespective of selecting the best fitting model. Discrimination between models on sample size, type of variable (continuous versus binary or ordinal measures) and the effect size of sources of variance and covariance. We introduce multimodel inference and model averaging approaches to the behavior genetics community, in the context of testing models for the causes of covariation between traits in term of genetic, environmental and causal explanations.

Heritability of overlapping impulsivity and compulsivity dimensional phenotypes

Impulsivity and compulsivity are traits relevant to a range of mental health problems and have traditionally been conceptualised as distinct constructs. Here, we reconceptualised impulsivity and compulsivity as partially overlapping phenotypes using a bifactor modelling approach and estimated heritability for their shared and unique phenotypic variance within a classical twin design. Adult twin pairs (N = 173) completed self-report questionnaires measuring psychological processes related to impulsivity and compulsivity. We fitted variance components models to three uncorrelated phenotypic dimensions: a general impulsive–compulsive dimension; and two narrower phenotypes related to impulsivity and obsessiveness.There was evidence of moderate heritability for impulsivity (A2 = 0.33), modest additive genetic or common environmental effects for obsessiveness (A2 = 0.25; C2 = 0.23), and moderate effects of common environment (C2 = 0.36) for the general dimension, This general impulsive–compulsive phenotype may reflect a quantitative liability to related mental health disorders that indexes exposure to potentially modifiable environmental risk factors.

Genetic and environmental variation in educational attainment: an individual-based analysis of 28 twin cohorts

We investigated the heritability of educational attainment and how it differed between birth cohorts and cultural–geographic regions. A classical twin design was applied to pooled data from 28 cohorts representing 16 countries and including 193,518 twins with information on educational attainment at 25 years of age or older. Genetic factors explained the major part of individual differences in educational attainment (heritability: a2 = 0.43; 0.41–0.44), but also environmental variation shared by co-twins was substantial (c2 = 0.31; 0.30–0.33). The proportions of educational variation explained by genetic and shared environmental factors did not differ between Europe, North America and Australia, and East Asia. When restricted to twins 30 years or older to confirm finalized education, the heritability was higher in the older cohorts born in 1900–1949 (a2 = 0.44; 0.41–0.46) than in the later cohorts born in 1950–1989 (a2 = 0.38; 0.36–0.40), with a corresponding lower influence of common environmental factors (c2 = 0.31; 0.29–0.33 and c2 = 0.34; 0.32–0.36, respectively). In conclusion, both genetic and environmental factors shared by co-twins have an important influence on individual differences in educational attainment. The effect of genetic factors on educational attainment has decreased from the cohorts born before to those born after the 1950s.

The (Broad-Sense) Genetic Correlations Among Four Measures of Inattention and Hyperactivity in 12 Year Olds

We estimated the genetic covariance matrix among four inattention (INATT) and four hyperactivity (HYP) measures in the classical twin design. Data on INATT and HYP symptom counts were obtained in mono- and dizygotic twin pairs (N = 1593) with an average age of 12.2 years (sd = .51). We analyzed maternal ratings of INATT and HYP based on the Conners’ Parent Rating Scale (CPRS), the Strengths and Weaknesses of ADHD-symptoms and Normal-behavior (SWAN), and teacher ratings based on the Conners' Teacher rating scale (CTRS) and the ASEBA Teacher Rating Form (TRF). Broad-sense heritabilities, corrected for the main effects of sex and for random teacher rater effects, were large (ranging from .658 to .912). The results reveal pervasive and strong broad-sense genetic effects on INATT and HYP phenotypes with the phenotypic covariance among the phenotypes largely due to correlated genetic effects. Specifically between 79.9 and 99.9% of the phenotypic covariance among the HYP measures, and between 81.0 and 93.5% of the INATT measures are attributable to broad-sense genetic effects. Overall, the present results, pertaining to the broad-sense heritabilities and shared genetic effects, support the current genome-wide association meta-analytic approach to identifying pleiotropic genetic variants.

A Comparison of the ASEBA Adult Self Report (ASR) and the Brief Problem Monitor (BPM/18-59)

The adult self report (ASR) is a well-validated instrument with multiple scales relating to adult psychopathology. Recently, an 18-item version has been introduced, the brief problem monitor (BPM) to measure Internalizing behavior (INT), Externalizing behavior (EXT), and attention problems (ATT). The present study compared the BPM and ASR and investigated how well the BPM can serve as a supplement or an alternative for the ASR for specific clinical and scientific purposes. In a large sample of adult twins (N = 9.835) from the Netherlands Twin Register (NTR), we compared the internal consistency, clinical classification concordance, means, and variances of the ASR and BPM. Using the classical twin design, we investigated the genetic covariance structure. For external validation, the associations between subjective well-being and different subscales of the ASR and BPM were compared. The internal consistency of the BPM scales (around α = 0.75) was somewhat lower than the ASR (α ~ 0.85). The BPM Externalizing scale showed the lowest internal consistency (α = 0.63). ASR and BPM scores showed good clinical classification concordance (0.61–0.80) and high correlations (r > 0.88). A small reversed sex difference in the BPM Externalizing scale appeared (women > men). Genetic (0.34–0.54) and environmental components (0.46–0.66) explained the variance to a similar extent for the ASR and BPM. The phenotypic and genetic associations with well-being were comparable. In situations where sum scores are sufficient, the BPM performs as well as the longer ASR. Depending on the situation and goal, it is worth considering the BPM as an alternative for the ASR to reduce the participant burden.

Genetic and environmental influences on human height from infancy through adulthood at different levels of parental education

Genetic factors explain a major proportion of human height variation, but differences in mean stature have also been found between socio-economic categories suggesting a possible effect of environment. By utilizing a classical twin design which allows decomposing the variation of height into genetic and environmental components, we tested the hypothesis that environmental variation in height is greater in offspring of lower educated parents. Twin data from 29 cohorts including 65,978 complete twin pairs with information on height at ages 1 to 69 years and on parental education were pooled allowing the analyses at different ages and in three geographic-cultural regions (Europe, North America and Australia, and East Asia). Parental education mostly showed a positive association with offspring height, with significant associations in mid-childhood and from adolescence onwards. In variance decomposition modeling, the genetic and environmental variance components of height did not show a consistent relation to parental education. A random-effects meta-regression analysis of the aggregate-level data showed a trend towards greater shared environmental variation of height in low parental education families. In conclusion, in our very large dataset from twin cohorts around the globe, these results provide only weak evidence for the study hypothesis.

Statistical Power and the Classical Twin Design.

Dr Nick Martin has made enormous contributions to the field of behavior genetics over the past 50 years. Of his many seminal papers that have had a profound impact, we focus on his early work on the power of twin studies. He was among the first to recognize the importance of sample size calculation before conducting a study to ensure sufficient power to detect the effects of interest. The elegant approach he developed, based on the noncentral chi-squared distribution, has been adopted by subsequent researchers for other genetic study designs, and today remains a standard tool for power calculations in structural equation modeling and other areas of statistical analysis. The present brief article discusses the main aspects of his seminal paper, and how it led to subsequent developments, by him and others, as the field of behavior genetics evolved into the present era.

Genetic and environmental variation in political orientation in adolescence and early adulthood: A Nuclear Twin Family analysis.

Political orientation is often assumed to be shaped by socialization processes; however, previous studies have shown substantial genetic variance components in party affiliation, political attitudes and behaviors, or closely related personality traits. The majority of these studies have relied on the Classical Twin Design, which comes with restrictive assumptions, some of which are easily violated. Moreover, most analyses lack a perspective of age-group differences. In this study, we investigated political orientation in adolescents (age: 16-18) and young adults (age: 21-25) in a cross-sectional Nuclear Twin Family Design. We used data of the German TwinLife project, including data from same-sex twins reared together, their biological parents, and nontwin full siblings. We found genetic variation in political orientation, which was significant in the older cohort, possibly indicating an increasing importance of active gene-environment correlation from adolescence to adulthood. Individual differences in political orientation because of passive gene-environment correlation and shared environmental effects were larger in the younger cohort, substantiating the same theoretical consideration and the importance of shared socialization contexts for adolescents' political views. By running Nuclear Twin Family model analyses, and considering age-group differences, as well as the relationship of political orientation with the Big Five personality traits, our study extended previous work, and resulted in more robust and fine-grained estimates of genetic and environmental sources of variance in political orientation. Therefore, it contributed to a better understanding of the complex nature-nurture interplay that forms political orientation in emerging adulthood. (PsycINFO Database Record (c) 2020 APA, all rights reserved).

Musings on Visscher et al. (2006).

The classical twin design relies on a number of strong number of assumptions in order to yield unbiased estimates of heritability. This includes the equal environments assumption - that monozygotic and dizygotic twins experience similar degrees of environmental similarity - an assumption that is likely to be violated in practice for many traits of interest. An alternative method of estimating heritability that does not suffer from many of these limitations is to model trait similarity between sibling pairs as a function of their empirical genome-wide identity by descent sharing, estimated from genetic markers. In this review, I recount the story behind Nick Martin's and my development of this method, our first attempts at applying it in a human population and more recent studies using the original and related methods to estimate trait heritability.

Birmingham and Beyond.

Nick Martin was a doctoral student of mine at the University of Birmingham in the mid 1970s. In this review, I discuss two of Nick's earliest and most seminal contributions to the field of behavior genetics. First, Martin and Eaves' (1977) extension of the model-fitting approach to multivariate data, which laid the theoretical groundwork for a generation of multivariate behavior genetic studies. Second, the Martin etal.'s (1978) manuscript on the power of the classical twin design, which showed that thousands of twin pairs would be required in order to reliably estimate components of variance, and has served as impetus for the formation of large-scale twin registries across the world. I discuss these contributions against the historical backdrop of a time when we and others were struggling with the challenge of figuring out how to incorporate gene-by-environment interaction, gene-environment correlation, mate selection and cultural transmission into more complex genetic models of human behavior.

The moderating role of SES on genetic differences in educational achievement in the Netherlands

Parental socioeconomic status (SES) is a strong predictor of children’s educational achievement (EA), with an increasing effect throughout development. Inequality in educational outcomes between children from different SES backgrounds exists in all Western countries. It has been proposed that a cause of this inequality lies in the interplay between genetic effects and SES on EA, which might depend on society and the equality of the education system. This study adopted two approaches, a classical twin design and polygenic score (PGS) approach, to address the effect of parental SES on EA in a large sample of 12-year-old Dutch twin pairs (2479 MZ and 4450 DZ twin pairs with PGSs for educational attainment available in 2335 children) from the Netherlands Twin Register (NTR). The findings of this study indicated that average EA increased with increasing parental SES. The difference in EA between boys and girls became smaller in the higher SES groups. The classical twin design analyses based on genetic covariance structure modeling pointed to lower genetic, environmental, and thus phenotypic variation in EA at higher SES. Independent from a child’s PGS, parental SES predicted EA. However, the strength of the association between PGS and EA did not depend on parental SES. In a within-family design, the twin with a higher PGS scored higher on EA than the co-twin, demonstrating that the effect of the PGS on EA was at least partly independent from parental SES. To conclude, EA depended on SES both directly and indirectly, and SES moderated the additive genetic and environmental components of EA. Adding information from PGS, in addition to parental SES, improved the prediction of children’s EA.

Pre- and Perinatal Characteristics Associated with Apgar Scores in a Review and in a New Study of Dutch Twins.

A literature review was carried out to identify pre and perinatal characteristics associated with variation in Apgar scores in population-based studies. The parameters identified in the literature search were included in the classical twin design study to estimate effects of pre and perinatal factors shared and nonshared by twins and to test for a contribution of genetic factors in 1- and 5-min Apgar scores in a large sample of Dutch monozygotic (MZ) and dizygotic (DZ) twins. The sample included MZ and DZ twins (N = 5181 pairs) recruited by the Netherlands Twin Register shortly after birth, with data on prenatal characteristics and Apgar scores at first and/or fifth minutes. The ordinal regression and structural equation modeling were used to analyze the effects of characteristics identified in the literature review and to estimate genetic and nongenetic variance components. The literature review identified 63 papers. Consistent with the review, we observed statistically significant effects of birth order, zygosity and gestational age (GA) for 1- and 5-min Apgar scores of both twins. Apgar scores are higher in first-born versus second-born twins and DZ first-born versus MZ first-born twins. Birth weight had an effect on the 5-min Apgar of the first born. Fetal presentation and mode of delivery had different effects on Apgar scores of first- and second-born twins. Parental characteristics and chorionicity did not have significant main effects on Apgar scores. The MZ twins' Apgar correlations equaled the DZ Apgar correlations. Our analyses suggest that individual differences in 1- and 5-min Apgar scores are attributable to shared and nonshared pre and perinatal factors, but not to genotypic factors of the newborns. The main predictors of Apgar scores are birth order, zygosity, GA, birth weight, mode of delivery and fetal presentation.

Genetic and Environmental Influences on Urinary Conditions in Men: A Classical Twin Study

ObjectiveTo evaluate the genetic and environmental relationship among prostatitis and other urological conditions, including benign prostatic hyperplasia (BPH) and prostate cancer (CaP), a classical twin design and biometric modeling was used. While prostatitis—characterized by pain and voiding symptoms, no clear etiology, and functional and quality of life impairments—co-occurs with other urinary conditions, the degree of shared overlapping etiologic processes among them remains unclear. We examined the contribution of genetic and environmental factors to these conditions and the etiology of their associations at the level of genetic and environmental influences. Methods4380 monozygotic and dizygotic male twin pairs from the Vietnam Era Twin Registry reported lifetime physician-diagnosed prostatitis (combined acute and chronic), bladder problems, enlarged prostate/BPH, and CaP. Multivariate biometrical modeling estimated the magnitude of genetic and environmental influences for each condition, as well as their genetic and environmental covariance. The common pathway model tested the assumption that covariation among these urinary conditions is determined by a single latent factor. ResultsOverall prevalence of prostatitis was 2.7%. Heritability estimates ranged from 19% for bladder problems to 42% for CaP. Significant shared environmental influences were present for CaP (12%), enlarged prostate/BPH (10%) but were smaller than genetic influences. A reduced one factor common pathway model provided the best fit, suggesting that covariation among the conditions is determined by a shared latent factor. ConclusionWe identified a common, genetically-influenced factor that accounts for much of the comorbidity among these 4 disease conditions. Nonshared environmental factors also make a significant contribution.