Summary The aim of this study is to investigate the role of gene-environment correlation (rGE) in the association between emotional symptoms and home chaos, parental feelings, discipline, monitoring and control and stressful life events using data from the Twins Early Development Study (TEDS). We will explore the role of rGE in emotional symptoms within age, at ages 4, 7, 9, 12, 16, and 21 using parent and twin reports. Firstly, phenotypic analyses will be conducted to assess the association between each environmental factor and emotional symptoms at each age. Secondly, to decompose the genetic and environmental influences, bivariate twin analyses will be conducted between emotional symptoms and each environmental factor. Background Emotional symptoms (e.g. low mood, tearfulness, fearfulness, and excessive worrying) in children and adolescents, have been identified as key vulnerability factors for developing emotional problems such as depression and anxiety (Birmaher et al., 1996; Goodman & Goodman, 2009), which in turn persist over time, conferring increased risk for later mental health problems (Gregory et al., 2007; Hannigan et al., 2017; Roza et al., 2003; Waszczuk et al., 2016). Depression and anxiety disorders are the two most common psychiatric conditions, with a global lifetime prevalence of 16-20% (Kessler et al., 2005, 2009). They onset in late childhood and adolescence (anxiety: 11 years old; Kessler 2005, depression: 18 years old; (Hankin et al., 1998; Kovacs & Devlin, 1998), making them two of the most prevalent disorders affecting youth (Merikangas et al., 2010). They are highly comorbid with other mental health illnesses and particularly with each other (Kessler et al., 2005). Evidence suggests that more than 75% of children diagnosed with depression, also express anxiety symptoms (Ghandour et al., 2019). Depression and anxiety disorders have complex and multifactorial aetiologies, with evidence from twin and genome-wide association studies suggesting that they are moderately heritable (Craske et al., 2017; Fernandez-Pujals et al., 2015; Howard et al., 2019; Polderman et al., 2015; Purves et al., 2020). Evidence suggests that depression and anxiety are genetically correlated in childhood and adolescence, potentially explaining their high comorbidity (Eley & Stevenson, 1999; Franić et al., 2010; Waszczuk et al., 2016; Waszczuk et al., 2014). Genetic factors are found to also explain the increased stability of the disorders (Hannigan et al., 2017; Nivard et al., 2015). Additionally, environmental factors such as the family environment and stressful life events are implicated in their development (Goodyer, 1990; Rapee et al., 2009) Family environment The family environment plays a key role in the development of children’s depression and anxiety, due to its relatively constant presence across development (Bronfenbrenner, 1986; Hannigan et al., 2017). Components of the family environment have been found to impact childrens’ depression and anxiety both concurrently and prospectively. For example, household chaos, defined as living in crowded, noisy and/or chaotic homes and lacking a routine, has been associated with symptoms of depression and anxiety in both children (Gregory et al., 2005; McAnally et al., 2019) and adolescents (Erickson & Newman, 2017; Evans et al., 2003). However, most of the evidence on the link between household chaos and depression and anxiety comes from studies in early childhood samples, and thus further research is required to further support this association in adolescence and early adulthood. Parenting style and the parent-child relationship have also been associated with the development and maintenance of depression and anxiety disorders in both childhood (Bögels & Brechman-Toussaint, 2006; McLeod et al., 2007; National Research Council (US) and Institute of Medicine (US) Committee on Depression et al., 2009; Rapee, 1997) and adolescence (Branje et al., 2010; Delay et al., 2013; National Research Council (US) and Institute of Medicine (US) Committee on Depression et al., 2009; Rapee et al., 2009). Parenting styles have been broadly grouped into two dimensions, namely, positive and negative parenting. Positive parenting includes behaviours such as increased responsiveness, acceptance and approval (Baumrind, 1991; McLeod et al., 2007) and is associated with increased autonomy and problem coping in children (Wolfradt et al., 2003). Negative parenting includes behaviours such as over controlling and critical parenting (Rapee, 1997) or demandingness (Oliver et al., 2014) and has been associated with poor emotion regulation and mental health outcomes and increased child dependence on the parent. Stressful life events Stressful life events have also been associated with the development of emotional problems and symptoms (Brown & Harris, 1978; Garber, 2010; Harland et al., 2002; National Research Council (US) and Institute of Medicine (US) Committee on Depression et al., 2009). Interestingly, most studies on young samples focus on depression (Goodyer, 2001) or on the association between traumatic life events (such as sexual abuse or neglect) and anxiety or depression (Cohen et al., 2001; Collishaw et al., 2007; Molnar et al., 2001). However, evidence from adult population (McLaughlin et al., 2010; Miloyan et al., 2018) and clinical (Francis et al., 2012; Taher et al., 2015) sample studies suggests that exposure to non-traumatic stressful life events is associated with the subsequent development of anxiety disorders and broader emotional problems (Kessler et al., 2010). At present there is limited evidence regarding the association between non-traumatic life events and the development of anxiety and broader emotional problems in young people. The evidence on the role of environmental factors as risk factors for emotional problems is strong. However, most cross sectional studies tend to assume a simple directionality of effects from the environment to children and adolescent behaviour and psychopathology, and do not consider the possibility of a bidirectional association which some studies have demonstrated (e.g. Hannigan et al., 2017; Hudson et al., 2009; Lewis et al., 2014). Additionally, the findings from cross sectional studies are limited by the presence of confounding variables such as genetic factors, which are known to influence environmental factors (Kendler & Baker, 2007) and may thus be underlying the bidirectional association. Gene-Environment correlation (rGE) One of the most interesting lessons learned from decades of twin studies is that these environmental factors do not occur independently of individuals’ behaviours, but instead, are shaped by individuals’ heritable characteristics (Bolinskey et al., 2004; Hanscombe et al., 2010; Jaffee & Price, 2007, 2007; Plomin, 1994; Plomin & Bergeman, 1991; Saudino et al., 1997; Silberg et al., 1999; Spotts et al., 2005). This is called gene-environment correlation (Plomin et al., 1977). Three rGE mechanisms have been identified, namely, passive, evocative and active (Plomin et al., 1977). Passive rGE refers to the correlation between the genotype inherited by a child and the environment provided by the parents, as influenced by their own heritable characteristics, and occurs independently of the child. Evocative rGE refers to the association between an individual’s behaviour that is genetically influenced and the response to this behaviour from others in the individual’s environment. Finally, active rGE refers to the association between the individual’s behaviour that is genetically influenced and the environmental situations into which the individual selects themself. Environmental factors are moderately heritable (Kendler & Baker, 2007). A meta analysis reported that genetic factors account for 18%-30% for family environment dimensions such as cohesion, conflict and organisation, for 35% of parenting behaviours (e.g. intrusiveness), and 28% of stressful life events (Kendler & Baker, 2007). Given the genetic influence on these aspects of the environment that are known to be associated with the development of depression, anxiety and emotional problems, which are also heritable, it is possible that underlying genetic factors influence their association. Several studies have conducted rGE analyses with depression and have explored the genetic covariation with home environment factors (e.g. Hannigan et al., 2017; Jaffee & Price, 2007; Lau & Eley, 2008; Neiderhiser et al., 1999; Pike et al., 1996; Wilkinson et al., 2013) and stressful life events (e.g. Lau & Eley, 2008; Rice et al., 2003; Silberg et al., 1999; Thapar et al., 1998). More analytically, a study (Wilkinson et al., 2013) exploring the genetic correlation between home chaos and parenting at age 9 and depression at age 12 reported that home chaos and parenting were genetically correlated with depression. These results suggest that the same genetic factors account for depression and home environment factors, which are also considered risk factors for developing depression. Moreover, another study (Lau & Eley, 2008) that investigated the rGE with depression and maternal discipline and dependent negative life events (i.e. events associated with an individual's behaviour e.g. break-up) in adolescents (mean age 15), found a significant genetic overlap between depression and both environmental factors. Furthermore, other studies exploring the genetic covariation between depression and family connectedness (e.g. parent-child relationship and parental warmth) (Jacobson & Rowe, 1999), parental hostility (Lewis et al., 2014), and parental negativity (Pike et al., 1996) in children in late childhood and adolescence have also supported these findings. Additionally, although the association is found to be weaker, there is evidence supporting the genetic covariation between total scores of negative life events (i.e. total number of the dependent and the independent stressful life events an individual has experienced) and depression in late childhood and adolescence (Rice et al., 2003; Silberg et al., 1999; Thapar et al., 1998). Surprisingly, very few studies have conducted rGE analyses with anxiety or with the broader category of emotional symptoms, which includes both anxiety and depression, suggesting this is a gap in the literature. One gene-environment correlation analysis with anxiety found that high levels of child anxiety were genetically correlated with over-controlling parenting (Eley et al., 2010), consistent with the presence of an evocative rGE. Based on these preliminary findings and evidence from research in depression, which has a similar genetic basis with anxiety (Eley & Stevenson, 1999), it is possible that the genetic factors contributing to the development of anxiety also influence the environmental factors associated with the development of anxiety. Furthermore, some studies that have explored the role of rGE in depression from a developmental perspective, suggest that the role of genetic and environmental factors changes during development and that genetic factors play a greater role as the child enters adolescence. For example, a study investigating developmental change in the association between the home environment (i.e. parental feelings and discipline and home chaos) and depression (Hannigan et al., 2017) found that the influence of genetic factors and non-shared environmental factors increased in the transition from childhood to adolescence, while the influence of shared environmental factors decreased. A study on the developmental change in the association between negative life events and depression suggested a similar pattern of change (Rice et al., 2003). These findings are consistent with the developmental perspective of rGE that was proposed by Scarr & McCartney (Scarr & McCartney, 1983), based on which the influence of the rGE mechanisms on behaviour and psychopathology changes over time. More analytically, they suggested that the influence of passive rGE is prominent in infancy and early childhood, when the child primarily attends to environmental stimuli that are shaped and provided by the parents, and declines as the child enters late childhood and adolescence. Adolescence is a period in which the child begins to explore other environments outside the family one, and constructs its own environments, and thus, active rGE has a greater influence on the child’s development. Evocative rGE exerts an influence throughout development, as children and adolescents evoke responses from people around them based on various characteristics that are influenced by their genotypes (e.g. personality). Taken together, these studies suggested that the increase in the role of genetic factors may explain the sharp increase in the onset of depression that comes with adolescence. It is again possible that similar developmental patterns of change influence the association between anxiety or broader emotional problems and environmental factors, however, at present, there are no studies that have investigated this. Additionally, there are no studies that have systematically explored the age-related change in the rGE, from early childhood to young adulthood. Most studies have been conducted on late-childhood and early adolescent samples, due to the critical role that the transition from childhood to adolescence plays in the development of psychopathology. However, it is equally important to understand the underlying mechanisms involved in the early years of development and the transition from adolescence to young adulthood and to identify the environmental factors that are most influential at each age. Evidence suggests that some anxiety disorders (e.g. separation anxiety and specific phobias) may start as early as 6-7 years old (Kessler et al., 2005), thus emphasising the need to understand the role of rGE in their development. Also, the transition to young adulthood is associated with important life changes, increased independence (Keller et al., 2007; Rimfeld et al., 2019) and the mean onset of depression (Hankin et al., 1998), stressing the importance of better understanding the role that genetic and environmental factors play in psychopathology. Gaps in the literature & the present study To date, studies on the role of rGE in depression and anxiety have emphasised the importance of disentangling the role of genetic and environmental effects in the development of psychopathology. However, as highlighted above, gaps exist in the literature. In summary, there is a need for more studies exploring the role of rGE in anxiety and broad emotional problems and a need for studies to adopt a developmental perspective and to systematically explore the age-related change in rGE in the association between depression, anxiety and broader emotional problems, and environmental factors. The present study aims to explore the role of rGE in emotional symptoms across development, in the Twins Early Development Study cohort, and specifically at ages 4, 7, 9, 12, 16, and 21. The TEDS cohort gives us the unique opportunity to cover these gaps in the literature and to improve our understanding of the rGE mechanisms involved in emotional symptoms across development. Exploring the role of rGE in the association between emotional symptoms, home chaos, parenting, and stressful life events, will allow us to disentangle the genetic and environmental effects influencing the emergence of depression and anxiety and broad emotional symptoms, and consequently, to better understand the aetiologies and maintenance mechanisms underlying them. Additionally, as the effects of specific environments may be more influential at certain time-points, the longitudinal design will allow us to identify the most influential factors at each age and consequently to inform prevention and treatment interventions.