Review
Behavioral genetic analyses of parent twin relationship quality

https://doi.org/10.1016/j.paid.2012.02.022Get rights and content

Abstract

This study estimates genetic and environmental effects on parent–offspring relationship quality. In addition, we tested to what degree offsprings’ and parents’ personality characteristics affect relationship quality. In a sample of 226 monozygotic and 168 dizygotic twin pairs we collected self-report relationship quality data for relationships with fathers and mothers. Fathers and mothers likewise reported relationship quality with each twin sibling. Independent of informants (parents versus twins) parent–offspring relationship quality was moderately to strongly heritable. Shared environmental influences had moderate and statistically significant effects on father–twin relationships. Effects of the shared environment were smaller and not statistically significant for mother–twin relationships. Offsprings’ personality traits were significantly related to relationship quality. Neuroticism, agreeableness, and conscientiousness showed consistent genetic correlations with all relationship quality measures, but there were no consistent environmentally mediated correlations. Parents’ agreeableness and neuroticism were consistently related to relationship quality, while conscientiousness showed correlation only for fathers. Our results suggest the importance of common genetic influences for the dynamic interplay of actor and partner that formed their relationship over time.

Highlights

► Parent–child relationship quality shows heritability. ► Relationship quality is reported by twins reared together and their parents. ► Heritability relationship quality is only partially mediated by personality.

Introduction

The relationship between parents and their children is regarded as among the most important relationships an individual experiences. Classical theoretical positions presume that this relationship has long term consequences for shaping an individual’s psychological functioning or personality formation and a wealth of research has shown associations of several aspects of parent–child relationships with important life outcomes in children, such as conduct problems (e.g., Ge et al., 1996) or antisocial personality disorder in adulthood (Johnson, Cohen, Chen, Kasen, & Brook, 2006). The current study focuses on the quality of parent–offspring relationships and explores its genetic and environmental influences in a study of twins.

There are several behavioral genetic studies of relationship quality, focusing on marital relations, attachment, and parent–offspring relationship quality. In these studies relationship with parents was mainly examined as a characteristic of the children’s environment. Several studies found a moderate heritability (Burt, 2003, Elkins et al., 1997, McGue et al., 2005, Plomin et al., 1994), increasing genetic influence during adolescence (Elkins et al., 1997, McGue et al., 2005), and a moderator effect of adolescents’ personality on the genetic and environmental influence on relationship quality (South, Krueger, Johnson, & Iacono, 2008). While previous studies have demonstrated, that adolescents according to their personality may evoke responses from their parents, South et al. (2008) found Gene × Environment interactions between personality and relationship quality or in other words, differential genetic expression in different environments. For example, positive relationship quality (regard) was more influenced by genetic and less by environmental variance in adolescents high on positive emotionality than in adolescents low on this broad personality trait.

Frequently, quality in parent–child relationships has been regarded as a latent construct characterizing the mutual relationship (e.g., Denissen et al., 2009, South et al., 2008). In contrast, in this study, we considered judgments of relationship quality as unique to each participant in the relationship (i.e. relationship quality experienced by a child need not necessarily be the same as the relationship quality mother or father experience) and so refrained from averaging across different perspectives.

For a thorough analysis, it is useful to adopt Kenny’s social relations model (Kenny et al., 2006, Snijders and Kenny, 1999; see also Rasbash, Jenkins, O’Connor, Tackett, & Reiss, 2011). This model distinguishes actor effects on perceptions of relationship quality from partner and relationship effects. The actor effect represents a person’s average level of experienced relationship quality in the presence of a variety of partners, while the partner effect represents the average level of relationship quality which a person elicits from a variety of “actors”. The relationship effect represents the deviation from actor and partner effects of an actor’s experienced quality of relationship with a particular partner.

Estimation of actor, partner, and relationship effects requires methods of data collection and data analysis that have been elaborated by Kenny (1994) in his social relations model. Examining how actor, partner, and relationship effects may contribute to twin similarity allows specifying hypotheses about the relative impacts of genetic and environmental influences on twin- and parent-reported relationship quality.

In our analyses we considered three sources of variance: additive genetic variance (denoted as a2), environmental variance shared by members of a family (c2) and variance due to environmental factors affecting members of a family individually (e2). In behavioral genetics the distinction between the shared and non-shared environment is based exclusively on the effects of the environment: c2 contributes to the similarity of family members, whereas e2 increases differences among family members. The effects of these sources on phenotypic variation are estimated by analyzing the covariation of phenotypes among family members (e.g., monozygotic and dizygotic twins). In the case of twin-reported relationship measures, actor effects affect the sizes of twin correlations positively to the degree they are themselves influenced by genetic or shared environmental sources of variance. In addition to actor effects, genetically or shared environmentally influenced rater effects (e.g., individual differences in the perception of relationship quality) contribute to twin similarity.

Partner effects are environmental effects, if there is no genetic relatedness between actor and partner (e.g., the relationship between parents and their adopted children) and no genetically influenced selection of partners (e.g., no assortative mating). But the characteristics influencing partner selection tend to show genetic influences, so genetic influences on partner effects are probably not uncommon, and shared environmental influences may operate similarly. Social homogamy, or the tendency for people with similar characteristics to congregate socially, in schools, workplaces, churches, leisure activity centers, etc., is an example of shared environmental effects on partner effects. To the degree that neither genetic nor shared environmental factors influence the selection of partners, partner effects reflect environmental influences not shared by siblings. In the case of parent–twin relationships, partner effects are by definition shared environmental effects if we analyze the twins’ ratings of relationship quality, because partner effects reflect the uniform effect of one person (mother or father) on two others. If identical parental behaviors had differential effects on the relationship, this would be a relationship effect.

Relationship effects reflect both an actor’s and a partner’s responses to specific characteristics of the interacting person. To the degree that relationship effects reflect partner’s responses to actor’s characteristics they should show the same etiology as the characteristics (e.g., personality traits) the partner responds to. To the degree they reflect an actor’s response to partner’s characteristics, the etiology depends on genetic and environmental influences on the tendency to react to partner’s characteristics. In biological relatives, actors’ and partners’ responses as well as their characteristics are correlated due to shared genetic influences. For example, if the mother is a friendly person, this may result in a partner effect contributing to positive relationships on average. In addition to this environmental effect, genetic relatedness results in a correlation of actor and partner effects, because, for example, genetic influences contributing to friendly mothers are transmitted and tend to bring about friendly offspring. In the study of twins’ ratings of quality of relationship with their parents, genetic and shared environment influences on the actor’s tendency to react to a certain interaction partner (mother or father) contribute to twin similarity, because both twins have the same parents (unlike in the study of relationships to peers).

If we consider measures provided by (individual) parents on their twins, the sources of variance of actor, partner, and relationship effects differ from the respective sources of variance based on offsprings’ reports. Note first, that now parents are the actors and the twins are in the role of partners. With respect to the variance components of the actor effect there is no genetic influence but only between-family variance. Any rater effect (or rater bias) contributes to twin similarity here (irrespective of the twins’ genetic similarity), inflating estimates of the shared environment, because one parent judges the relationships with both twins, presumably to the same degree in MZ and DZ twins. In addition, contrast effects that are well documented for parental judgments of children’s temperament (Buss and Plomin, 1984, Hwang and Rothbart, 2003, Saudino, 2003, Spinath and Angleitner, 1998) are likely to play greater roles here. Contrast effects may result from comparing relationship quality between twin siblings instead of judging each individual relationship with reference to the population of all relationships. They result in underestimating the true correlation between dizygotic twins more than the correlation between monozygotic twins, because in DZ twins there are more characteristics to contrast. As a consequence, if there are contrast effects, they inflate genetic effects, leading to an underestimation of shared environmental effects.

Partner effects on parental reports have a genetic basis, since individual attributes that affect relationship quality and similarity in those attributes may depend on twins’ genetic relatedness. Influences on relationship effects should in principle be identical for parents’ reports and twins’ reports. Because of the systematic effects of rater bias on estimates of these sources, however, higher estimates of genetic or shared environmental influences may result from estimates of parents’ reports than from twins’ reports. This is not because parental reports are more likely to be subject to rater bias, but because of the systematic effects of the location of genetic relatedness in actors or partners.

If we find genetic influences on parent–offspring relationship quality, it is reasonable to hypothesize that these influences are at least partially mediated by genetic influences on personality traits, which in turn affect relationship quality. Asendorpf (2002) argued that the quality of all dyadic relationships is strongly influenced by the interacting partners’ personality traits mainly because small effects of personality summate over time and over different relationships and result in a characteristic relationship pattern. Denissen et al. (2009) found substantial correlations between five-factor model personality traits and relationship quality. In particular, all traits correlated significantly with the warmth and restrictive control dimensions of relationship quality. The warmth dimension tended to correlate more highly with personality characteristics than the control dimension. Agreeableness and openness were the best parental predictors of relationship quality. Offsprings’ agreeableness was a good predictor of relationship quality.

Given the significance of parent–offspring relationships as well as their roles in explanations of offsprings’ behavior it is important to understand how genetic and environmental influences may be involved. We estimated genetic and environmental influences separately for twin- and parent-reported relationship quality. In addition, we estimated the influences of offsprings’ and parents’ personalities on relationship quality to investigate how genes might be involved in relationship quality.

Section snippets

Participants

We analyzed data of twins reared together and their parents from the Jena Twin Study of Social Attitudes (JeTSSA; Stößel, Kämpfe, & Riemann, 2006). The sample consisted of 875 individuals (646 of all twins were women, 74%) including 226 monozygotic, 168 dizygotic twin pairs, and 87 unmatched twins (data available only for one twin sibling) mainly from the Jena Twin Registry. Part of the sample consists of twins that were recruited from the Berlin Twin Registry (Busjahn, 2006) and the Bielefeld

Results

Means and standard deviations indicated that on average parents as well as offspring judged the quality of their relationships positively. We ran a separate two factorial ANOVA for each twin sibling encompassing the factors “Parent versus Offspring” and “Mother versus Father”. They revealed a main effect for parents versus offspring (Twin 1: df = 1, 217, F = 113.82, p < .001, partial eta2 = .34; Twin 2: df = 1, 221, F = 105.62, p < .001, partial eta2 = .32), no main effect of relationship with mother versus

Discussion

Across informants (parents and twins), parent–offspring relationship quality was moderately to strongly heritable. There were moderate and statistically significant shared environmental influences on father–twin relationships. Contrary to expectations, shared environmental influences were smaller and not statistically significant for mother–twin relationships. Again, these results were observed for parents’ and twins’ reports of relationship quality. Offsprings’ as well as parents’ personality

References (34)

  • I.J. Elkins et al.

    Genetic and environmental influences on parent-son relationships: Evidence for increasing genetic influence during adolescence

    Developmental Psychology

    (1997)
  • X. Ge et al.

    The developmental interface between nature and nurture: A mutual influence model of child antisocial behavior and parent behaviors

    Developmental Psychology

    (1996)
  • L.R. Goldberg

    An alternative “description of personality“: The Big-Five factor structure

    Journal of Personality and Social Psychology

    (1990)
  • W.G. Hill et al.

    Data and theory point to mainly additive genetic variance for complex traits

    PLOS Genetics

    (2008)
  • J.G. Johnson et al.

    Parenting behaviors associated with risk for offspring personality disorder during adulthood

    Archives of General Psychiatry

    (2006)
  • Kandler, C., Bleidorn, W., & Riemann, R. (2011). Left or right? Sources of political orientation: The roles of genetic...
  • C. Kandler et al.

    Genetic and environmental mediation between measures of personality and family environment in twins reared together

    Behavior Genetics

    (2009)
  • Cited by (8)

    View all citing articles on Scopus
    1

    Tel.: +49 521 1064540.

    View full text