Gene-Environment Interplay and Antisocial Behavior PDF

Gene-environment interplay and antisocial behaviour - Reading 04 December 2023 19:41 Source Notes Accounting for Genetic and Environmental Confounds in Associations Between Parent and Child Characteristics: A Systematic Review of Children-of-Twins Studies Introduction Parental Characteristics and Ch...

Gene-environment interplay and antisocial behaviour - Reading 04 December 2023 19:41 Source Notes Accounting for Genetic and Environmental Confounds in Associations Between Parent and Child Characteristics: A Systematic Review of Children-of-Twins Studies Introduction Parental Characteristics and Child Characteristics ○ Parents influence their children's development in a variety of ways. ○ Many traits tend to run in families, suggesting that parent behavior impacts child development. ○ Children of anxious parents are more likely to be anxious themselves. ○ Children learn behaviors from their parents through imitation and modelling. ○ Parents can also influence their children's behavior through verbal transmission of information. (McAdams et al., 2014) Parenting Behaviors and Child Outcomes ○ Parents often seek to influence their children's behavior directly. ○ Punishment and praise can be used to condition and reinforce desired behaviors. ○ Parental monitoring is associated with reduced adolescent externalizing behaviors. ○ Harsh parental discipline is associated with elevated levels of psychopathology. Parenting Styles and Child Outcomes ○ Parenting style has been linked to a host of child outcomes, including personality, educational achievement, and psychopathology. ○ Authoritarian parenting is associated with offspring conduct problems. ○ Parenting styles characterized by warmth and positive expressivity are associated with reduced externalizing problems in children. Bidirectional Effects of Parenting and Child Behavior ○ The direction of effect between parenting and child outcomes is often conceptualized as running from parent to child. ○ However, children's behavior can also affect the parenting they receive. ○ The association between parenting and child outcome is often reciprocal, with each affecting the other over time. Other Elements of the Family Environment ○ Beyond parental characteristics and behaviors, other elements of the family environment also impact child development. ○ These elements include the organization of the home environment, the provision of play materials, and the marital relations of parents. ○ Examples of the impact of these elements include the link between household chaos and children's problem behavior, the association between violent video games and increased aggression, and the increased emotional and behavioral problems in children from divorced or separated families. The Confounding Effects of Genetic Relatedness Overview ○ Researchers often study the role of parents in child development by examining associations between parent behavior, parenting style, and the family environment and child outcomes. Genetic Transmission as a Complicating Factor ○ Due to genetic similarities between parents and children, associations between their characteristics can reflect genetic transmission rather than environmental influence. Generalist Genes Hypothesis ○ The generalist genes hypothesis suggests that genes can affect multiple traits, leading to correlations between conceptually distinct phenotypes. Psychopathology ○ All forms of psychopathology share common variance via a single general psychopathology dimension, and this dimension is highly heritable. Genetic Pleiotropy ○ Genetic pleiotropy means that any association between a parental measure and child outcome is potentially confounded by shared genes. Bidirectional Effects ○ Child behavior can also affect parenting style, further complicating the interpretation of associations between parent and child characteristics. Gene-Environment Correlation (rGE) ○ rGE refers to a correlation between an individual's genome and the environment they inhabit. ○ rGE can be passive, active, or evocative. ○ rGE is a common source of confound in studies of parent-child relationships. Examples of rGE ○ A child may inherit genetic factors involved in conduct problems from their parent, who may also be more likely to use harsh discipline due to the same genetic factors (passive rGE). ○ A child's genetically influenced conduct problems may lead them to actively seek confrontation with their parent (active rGE) or evoke harsh discipline from their parent (evocative rGE). Implications for Research and Intervention ○ Many ostensibly environmental aspects of the rearing environment are subject to genetic influence. ○ Associations between these variables and measures of child outcome may be subject to the confounding effects of rGE. ○ Getting an accurate picture of which associations are confounded and which are not is crucial for designing effective interventions. Behavioral Scientists' Methods to Account for Confounds ○ Experimental tradition: Random allocation of participants to conditions, such as in randomized control trials of parenting interventions. ○ Quasi-experiments: Using naturally occurring groups of individuals that differ in their genetic and/or environmental relatedness, such as twin studies. Children-of-Twins (COT) Design ○ COT studies involve using samples of twins who have children. ○ COT studies have risen in popularity due to the increasing availability of twin samples with children. The Children-of-Twins Method Overview ○ The Children-of-Twins (COT) method is a research design that utilizes samples of twins who have children. ○ COT studies aim to disentangle genetic and environmental factors in parent-child relationships. Logic and Strengths ○ COT studies address limitations of traditional parent-child studies by controlling for genetic confounds. ○ COT studies allow for the examination of bidirectional associations between parent and child characteristics. Weaknesses and Methodological Considerations ○ COT studies require large sample sizes to achieve adequate statistical power. ○ COT studies may be susceptible to selection bias and potential confounds from non-genetic factors. Careful design and analysis are crucial to maximize the validity and interpretability of COT studies. PSYC0036 Genes and Behaviour Page 1 ○ Careful design and analysis are crucial to maximize the validity and interpretability of COT studies. The Logic of the Children-of-Twins Method Genetic Relatedness in Families ○ The genetic relatedness between two people can be described as the proportion of genetic variance they share. ○ Siblings with the same parents share an average of 50% of their genetic variance. ○ Dizygotic (DZ) twins share an average of 50% of their genetic variance. ○ Monozygotic (MZ) twins share 100% of their genetic variance. ○ As a result, the offspring of MZ twins are as genetically related to their parents' co-twin as they are to their own parent (50%). Children-of-Twins (COT) Method ○ The COT method is a research design that utilizes samples of twins who have children. ○ COT studies aim to disentangle genetic and environmental factors in parent-child relationships. ○ COT studies use the unique genetic relationships within MZ twin families to distinguish between genetic and environmental transmission from one generation to the next. Shared Environment ○ The "shared environment" is a title given collectively to nongenetic factors that make members of a nuclear family similar to one another. ○ These shared environmental effects tend to be estimated as far smaller in magnitude than genetic effects. Confounding Effects in COT Studies ○ COT analyses are capable of accounting for the confounding effects of the extended family environment as well as genetic confounds. Analysing COT Data Early COT Studies ○ Early COT studies analysed families of MZ twins to distinguish between genetic and environmental factors in parent-child relationships. ○ Nance and Corey (1976) developed a method for analysing such families using parent-offspring and avuncular correlations. Limitations of MZ-Only COT Studies ○ MZ-only COT studies cannot fully disentangle genetic and environmental effects in parent phenotypes. ○ MZ-only COT studies lack power to detect certain parameters, such as intergenerational pathways and offspring etiology. COT Studies with MZ and DZ Twins ○ COT studies with both MZ and DZ twins can estimate the etiological structure of both parent and child phenotypes. ○ Including more twin pairs increases power and generalizability of results. ○ COT studies with MZ and DZ families rely on comparisons of intrafamilial correlations. Interpreting COT Correlations ○ Differences between MZ and DZ correlations for parenting phenotypes indicate genetic influence on parenting. ○ Differences between MZ and DZ correlations for child phenotypes indicate genetic influence on child characteristics. ○ Larger parent-child correlations than avuncular correlations suggest an effect of parental phenotype on child phenotype. Analytical Techniques for COT Studies ○ Between-families comparisons ○ Hierarchical linear modelling ○ Structural equation modelling Between-Families Comparisons Overview ▪ Between-families comparisons involve grouping offspring into risk categories based on their genetic and environmental exposure. ▪ Statistical tests compare these groups to infer whether associations between parent and child phenotypes are genetic or environmental. Applications ▪ This method is suitable for analysing COT data where the parental phenotype is dichotomous, such as psychiatric diagnosis. Example Study ▪ Haber et al. (2005) used the presence or absence of alcohol dependence in twins (parents) to create four offspring groups: 1) Exposed to parental alcoholism and at high familial risk 2) Not exposed to parental alcoholism but at high familial risk 3) Not exposed to parental alcoholism but at moderate familial risk 4) Not exposed to parental alcoholism and at low familial risk ▪ Significant differences between groups indicate whether parental alcoholism predicts child outcome above and beyond familial risk. ▪ Higher prevalence rates in Group 2 (high genetic risk, no exposure) than Group 3 (moderate genetic risk, no exposure) indicat e genetic effects. Limitations ▪ Meaningful inferences about causal effects depend on comparisons between offspring of discordant twin pairs. ▪ For highly heritable phenotypes, selective sampling may be necessary to obtain a sufficient number of discordant twin pairs. Hierarchical Linear Models Overview ▪ Hierarchical linear modelling (HLM) is a statistical technique that considers the nested structure of COT data, with offsprin g nested within nuclear families and families nested within twin families. ▪ HLM can incorporate various covariates and control variables and is suitable for analyzing both continuous and categorical data. ▪ HLM involves fitting multiple regression models to examine associations between parent and child phenotypes at different levels of the analysis and with different covariates. Assessing Genetic and Environmental Confounds ▪ Cousin comparisons, or within-twin-family effects, are particularly useful for distinguishing between genetic and environmental confounds. ▪ Cousins share genetic and environmental familial confounds, so if differences in parental phenotype predict differences in offspring phenotype within twin families, it indicates an environmental effect of parent on child. ▪ Comparing the strength of within-twin-family effects in MZ and DZ twin families reveals whether genetic confounding is present. ▪ A stronger within-twin-family association in DZ families than MZ families implies genetic confounding, as the effect is attenuated more in MZ families due to their higher genetic relatedness. ▪ Negligible differences between MZ and DZ families suggest that genetic confounding is not a factor, and any familial confounding is likely environmental. Comparison with Between-Families Comparisons ▪ A study by Slutske et al. (2008) found that between-families comparisons and within-family HLM approaches yielded similar conclusions. ▪ However, HLM is considered more sophisticated and allows for more rigorous hypothesis testing by directly comparing cousins PSYC0036 Genes and Behaviour Page 2 ▪ However, HLM is considered more sophisticated and allows for more rigorous hypothesis testing by directly comparing cousins rather than groups of individuals at different risk levels. Structural Equation Models Overview ▪ Structural equation models (SEMs) explicitly quantify latent genetic and environmental influences on phenotypes. ▪ SEMs decompose variance into additive genetic effects (A), common environment effects (C), and nonshared environment effects (E). ▪ SEMs of COT data can determine the proportion of covariance between parent and child phenotypes attributable to genetic and environmental effects. Limitations ▪ COT models do not account for bidirectional associations between parent and child phenotypes (child's phenotype affecting parent's phenotype). ▪ The extended children-of-twins (ECOT) model is designed to account for bidirectional associations. The ECOT Model ▪ The ECOT model estimates both parent-to-child and child-to-parent effects. ▪ The ECOT model can distinguish between passive rGE (genes and correlated rearing environment) and active/evocative rGE (child's genetically influenced behavior affects a correlated environment). ▪ The ECOT model requires two data sets: ▪ Twins and their cousins (estimates genetic and environmental effects on parenting) ▪ Twin children and their parents (estimates genetic and environmental effects on child phenotype) Detecting Active/Evocative rGE ▪ Accurate detection of active/evocative rGE requires estimation of genetic effects on offspring phenotype. ▪ ECOT studies introduce a second data set to increase power to detect genetic effects on offspring phenotype. ▪ The second data set consists of twin children and their parents. Bidirectional Pathways ▪ The ECOT model estimates bidirectional pathways between offspring phenotype and parenting. Methodological Considerations in COT Studies Assumption of Random Mating ○ Random mating: The assumption that mates select each other randomly. ○ Assortative mating: The situation where mates select each other based on similarity. ○ COT models should include information on the spouse to control for assortative mating. Equal Environments Assumption (EEA) ○ EEA in classical twin studies: The assumption that environments of MZ twins are not substantially more similar than those of DZ twins. ○ EEA in COT studies: The assumption that offspring of MZ twin pairs are not influenced by their parent's co-twin more than offspring of DZ twins. ○ Violation of EEA can lead to false conclusions about genetic influence. ○ Avuncular contact (time spent together or in contact with one another) is one possible route for EEA violation. ○ EEA violation can be assessed by measuring avuncular contact and including it as a covariate. Age Differences ○ Cousins are typically not the same age. ○ Most phenotypes and their intergenerational associations change with age. ○ Age differences between cousins and parental age at child's birth should be controlled for or accounted for in sampling strategies. Additional Considerations ○ Power: COT studies often face statistical power problems due to small sample sizes, categorical measures, high twin correlations, and highly heritable dichotomous disorders. ○ Interpreting results: Tentatively interpret point estimates when power is low, but acknowledge any lack of statistical significance. ○ Power analyses: Use power analyses and simulations to determine ideal sample sizes. Limitations of the COT Design General Limitations ▪ The COT design does not allow researchers to make causal conclusions due to the possibility of unmeasured confounding variables. ▪ Including spousal information in COT models can help control for confounding variables, but it only provides information at t he phenotypic level. ▪ Dyadic phenotypes (e.g., divorce) can affect the COT model's ability to distinguish genetic from environmental intergenerational transmission. ECOT Model Limitations ▪ Estimating genetic and shared environmental effects on both child and parent phenotypes limits power to reject false causalit y hypotheses. ▪ ECOT models should only be applied where the influence of the shared environment on the parenting phenotype is minimal. ▪ Error terms in bidirectional effects models must be constrained to be the same in parents and offspring. Statistical Power Limitations ▪ COT studies often face statistical power problems due to: □ Small sample sizes □ Use of categorical measures □ High twin correlations □ Highly heritable dichotomous disorders Impact of Low Power ▪ Low power can lead to overlapping confidence intervals, making it impossible to distinguish genetic from environmental effects. A Systematic Review of COT Studies Literature Search ○ Initial search using phrase "children of twins" or "offspring of twins" yielded 81 articles. ○ Excluded non-relevant articles, reviews, and editorial piece. ○ Excluded methodological articles and non-psychological/behavioral articles. ○ Refined broader search using "children" and "twins" keywords. ○ Checked titles, abstracts, and contents of articles for relevance. ○ Contacted researchers for additional articles. Total COT Articles ○ Identified 43 COT articles examining intergenerational transmission of phenotypes. ○ 36 articles included in Table 3. ○ 7 articles not included in Table 3 due to data overlap or non-psychological/behavioral focus. Thematic Discussion ○ Focused on intergenerational transmission of emotional and behavioral disorders, parenting style-child outcome associations, and family environment-child outcome associations. Overarching Themes PSYC0036 Genes and Behaviour Page 3 Overarching Themes ○ COT method provides evidence for both genetic and environmental transmission of various phenotypes. ○ Findings support the importance of both nature and nurture in shaping individual outcomes. Alternative Uses of COT Data ○ COT data can be used to examine intergenerational transmission of risk or resilience to disease. ○ COT data can be used to study gene-environment interactions. Intergenerational Transmission of Emotional and Behavioral Disorders Parental Psychoses Investigating Parental Psychosis and Its Influence on Offspring ▪ Early COT studies aimed to understand the intergenerational transmission of psychotic disorders, particularly schizophrenia. ▪ The goal was to determine whether rearing by a schizophrenic parent or shared familial factors contribute to psychosis risk i n offspring. Fischer's COT Study (1971) ▪ Examined psychosis prevalence in offspring of MZ twins discordant for schizophrenia diagnosis. ▪ Found no significant difference in psychosis rates between offspring of schizophrenic and non -schizophrenic twins. ▪ Concluded that intergenerational transmission of psychosis is familial, not solely due to parental influence. Follow-up COT Study (Gottesman & Bertelsen, 1989) ▪ Included DZ twins and reassessed participants from Fischer's study, including their offspring. ▪ Followed up with participants well past the risk period for schizophrenia development. ▪ Reaffirmed Fischer's findings, further supporting the familial nature of psychosis transmission. Replication Study (Kringlen, 1987) ▪ Replicated findings in a separate MZ twin sample using structured interviews for schizophrenia assessment. ▪ Similar results: Intergenerational transmission of schizophrenia is primarily familial, not solely due to parental influence. Parental Depression Association between Parental Depression and Offspring Depression ▪ COT studies have investigated the intergenerational transmission of depression. ▪ Two studies, Silberg et al. (2010) and Singh et al. (2011), examined the association between parental depression and offsprin g depression. ▪ Both studies found that the association was not significantly inflated by genetic or environmental confounds. ▪ This suggests that exposure to a depressed parent is a risk factor for child or adolescent depression. Mediating Mechanism: Perceived Self-Competence ▪ Perceived self-competence has been suggested as a mediatory mechanism in the link between parent depression and offspring depression. ▪ Class et al. (2012) conducted COT analyses of this association in the Twin and Offspring Study of Sweden (TOSS). ▪ Analyses revealed sex differences in the nature of transmission. ▪ The association between maternal depressive symptoms and offspring perceived self -competence was not significant once shared genetic and environmental liability was controlled for. ▪ However, the association between paternal depression and offspring self -competence was independent of such confounds. Parental Antisocial Behaviour Impact of Parental Antisocial Behavior on Child Outcomes ▪ COT studies have examined the intergenerational transmission of antisocial behavior. ▪ Two studies, D'Onofrio, Slutske, et al. (2007) and Silberg, Maes, & Eaves (2012), investigated the effects of parental antiso cial behavior on child outcomes. Sex Differences in Transmission ▪ D'Onofrio, Slutske, et al. (2007) found that the nature of intergenerational transmission differed for boys and girls. ▪ In girls, the association between parental conduct problems and offspring conduct problems was purely genetic. ▪ In boys, there was evidence for an environmental effect even after genetic transmission was accounted for. Genetic and Environmental Transmission ▪ Silberg, Maes, & Eaves (2012) found that the link between parental antisocial behavior and child conduct disturbance involved both genetic and environmental transmission. ▪ Genetic overlap was reported between antisocial behavior in parents and their children. Antisocial Behavior and Hyperactivity ▪ Silberg, Maes, & Eaves (2012) found that the association between parent antisocial behavior and child hyperactivity was entirely genetic. ▪ This supports the generalist genes hypothesis that many of the same genes underlie distinct psychiatric disorders. Antisocial Behavior and Depression ▪ In contrast to hyperactivity, there was no significant genetic association between parental antisocial behavior and child depression. ▪ Instead, the association appeared to be environmental. ▪ Child depression may largely be a response to a negative rearing environment, with genetic factors playing a lesser role. Parental Alcoholism Impact of Parental Alcoholism on Offspring Psychopathology ▪ COT studies have examined the intergenerational transmission of alcoholism and other forms of psychopathology. ▪ Eight COT studies have investigated the effects of parental alcohol problems on child outcomes. ▪ Parental alcoholism has been linked to offspring alcoholism, attention-deficit/hyperactivity disorder (ADHD), conduct problems, internalizing problems, and suicidal behaviors. Genetic and Environmental Transmission ▪ Most studies found that phenotypic associations between parental alcoholism and offspring psychopathology were likely genetic in nature. ▪ There was limited evidence to support an environmental effect of parental alcoholism on offspring psychopathology. Limitations of COT Studies ▪ There are some limitations to these studies, such as the use of dichotomous variables to measure psychopathology and the use of measures that may have done a better job of capturing genetic risk than environmental risk. Parental Drug Use and Abuse Impact of Parental Drug Use on Offspring Psychopathology ▪ Parental drug use has been linked to offspring cannabis abuse/dependence and conduct disorder. Genetic and Environmental Transmission ▪ COT studies have provided mixed evidence for the role of genetic and environmental factors in transmitting parental drug use to offspring. ▪ One study found that the association between parental drug dependence and offspring cannabis abuse/dependence was both genetic and environmental. ▪ Another study found that parental drug dependence was associated with offspring conduct disorder after genetic risk was accounted for. ▪ However, other studies were inconclusive due to limitations such as the use of lifetime psychiatric diagnoses and poorly defined environmental risk. PSYC0036 Genes and Behaviour Page 4 defined environmental risk. Additional Considerations ▪ Substance abuse is heritable, so the association between parental substance use and child outcomes could be due to genetic factors. ▪ Substance abuse could also cause poor parenting practices that lead to neglect and abuse, which could then lead to negative child outcomes. ▪ COT studies can help to elucidate the nature of associations between substance abuse and child psychopathology by controlling for genetic confounds. Parental Smoking Impact of Parental Smoking on Offspring Health ▪ Parental smoking has been linked to negative health outcomes in offspring, including low birth weight and nicotine dependence. Genetic and Environmental Transmission ▪ COT studies have shown that the association between parental smoking and offspring health outcomes is partially genetic and partially environmental. ▪ Even after accounting for genetic factors, exposure to maternal smoking during pregnancy has been shown to have a significant effect on offspring birth weight. ▪ Nicotine dependence in offspring has been shown to be partially explained by both genetic transmission from parents and environmental exposure to parental smoking. Specificity of Transmission ▪ Risk for nicotine dependence in offspring is associated with familial risk for nicotine dependence but not alcohol dependence. ▪ Similarly, familial risk for alcohol dependence predicts offspring alcohol dependence but not nicotine dependence. The Effects of Parenting on Child Outcome Impact of Parenting on Child Outcomes ○ COT studies have shown that parenting behavior has a significant impact on child outcomes. ○ Harsh physical punishment has been linked to child behavioral problems, internalizing problems, and drug and alcohol use. ○ Maternal emotional overinvolvement and criticism have been linked to offspring internalizing problems and externalizing problems, respectively. ○ Paternal criticism has been linked to offspring externalizing problems. ○ Negative parenting has been linked to offspring externalizing behavior. Mechanisms of Transmission ○ The effects of parenting on child outcomes can be explained by several mechanisms, including passive rGE and active/evocative rGE. ○ Passive rGE occurs when parents and children share genetic predispositions to certain behaviors. ○ Active/evocative rGE occurs when the behavior of one individual evokes a response from the other. Bidirectional Effects ○ ECOT studies have shown that the relationship between parenting and child outcomes is bidirectional. ○ That is, child behavior can also influence parenting behavior. The Effects of the Family Environment on Child Well-Being Adolescent Motherhood ○ Adolescent motherhood has been linked to a range of adverse outcomes in offspring, including substance use, externalizing problems, and internalizing problems. ○ These associations are not entirely explained by genetic confounding. Family Structure ○ Maternal age at first birth has been inversely associated with offspring criminal convictions. ○ The presence of a stepfather has not been shown to cause early menarche in girls. Family Conflict ○ Family conflict is associated with the development of internalizing problems in children. ○ Family conflict is also associated with externalizing problems in children, but this association is partially explained by genetic confounding. Marital Instability ○ Parental marital instability is associated with a range of negative outcomes in offspring, including substance use, internalizing problems, externalizing problems, reduced academic performance, fewer years in education, and cohabitation. ○ These associations are not entirely explained by genetic confounding. Overview of Findings to Date COT studies have elucidated the mechanisms underlying many associations between parent and child phenotypes. Transmission of psychosis: Likely genetic in nature. Parental antisocial behavior and depression: Phenotypic effect on offspring well-being after accounting for familial confounds. Parental substance use: Not an environmental risk factor for offspring substance use and related psychopathology. Associations attributable to common familial factors and/or related psychopathology. Parenting practices: Several parenting behaviors may provide a pathway through which psychopathology is transmitted from one generation to the next. Child-to-parent effects: ECOT studies show that children can affect their parent's behavior just as much as parents affect their children. Marital discord, family conflict: Important predictors of offspring psychopathology even after accounting for familial confounds. Genetic overlap: Common in COT analyses, with many examples of cross-disorder transmission. Limitations of COT Research ○ Underpowered studies ○ Inconsistent findings across studies ○ Methodological criticisms Recommendations for Future Research ○ Apply multiple genetically informative methods to issues of intergenerational transmission of psychopathology, the influence of parenting practices on child development, and the effects of the family environment on offspring well-being. ○ Longitudinal studies of adoptees and their adoptive parents ○ Samples involving offspring conceived using assisted reproductive technologies ○ Samples of nontwin parent–child dyads where the parent generation comprises full/half siblings and/or cousins ○ Linking longitudinal national population registers Example Study ○ Jundong et al. (2012): Linked longitudinal national population registers in Sweden to compare school performance of the offspring of schizophrenic and non-schizophrenic parents, resulting in a sample of over 1.4 million individuals. ○ Results indicated that genetic factors accounted for the association between parental schizophrenia and poor school performance in offspring. Alternative Uses of COT Data Exploring the relative role of genes and environment in the etiology of parenting ○ COT studies have been used to examine the factors that contribute to parenting behaviors. ○ These studies have found that both genes and environment play a role in parenting. PSYC0036 Genes and Behaviour Page 5 Assessing the importance of marital partners as a source of influence on maternal adjustment ○ COT studies have been used to investigate the impact of spousal characteristics on maternal well-being. ○ These studies have shown that marital partners can have a significant influence on maternal adjustment. Testing evolutionary theories of inclusive fitness ○ COT studies have been used to examine gift-giving behaviors within avuncular relationships. ○ These studies have provided support for evolutionary theories of inclusive fitness. Controlling for genetic susceptibility and environmental risk ○ COT data can be used to control for genetic susceptibility and environmental risk when examining associations between measured environmental risk factors and offspring outcome. ○ This approach can help to isolate the effects of specific environmental factors on child development. Example Study ○ Duncan et al. (2008) used COT data to examine the association between childhood sexual abuse and cannabis abuse/dependence in offspring. ○ They found that childhood sexual abuse predicted cannabis abuse/dependence above and beyond the genetic and familial risks associated with having a drug-dependent father. Possible Future Directions The use of COT data could be extended to investigate positive phenotypes such as well-being, resilience, and ability. Physiological measures could be included in COT studies to elucidate the biological pathways through which behavioral phenotypes are transmitted from one generation to the next. COT studies could be applied to phenotypes outside of psychology and psychiatry, such as birth weight, education, asthma, and income. Collecting and utilizing prospective COT data could be a powerful technique to study infant and child development as it happens. Sex differences in COT associations should be investigated more thoroughly. The potential for bidirectional effects between parent and child characteristics should be explored. Multivariate COT analyses could be used to better understand the complex nature of intergenerational transmission. The COT design could be combined with other family designs to further study intergenerational associations. Disentangling prenatal and inherited influences in humans with an experimental design (Rice et al., 2008) Background: The study investigates the impact of maternal smoking during pregnancy on two adverse offspring outcomes: reduced birth weight and increased childhood antisocial behavior. Previous research indicates associations between prenatal smoking and these outcomes. However, disentangling whether these associations are due to inherited maternal factors or true prenatal effects has been challenging. The study aims to utilize in vitro fertilization (IVF) and unrelated maternal gestation to delineate prenatal effects from inherited influences. Methods: The researchers examined records of 779 IVF-conceived children gestated by related or unrelated mothers. They assessed the associations between maternal smoking during pregnancy and (i) birth weight and (ii) antisocial behavior in both related and unrelated pregnancies. Statistical analyses and standardized means were used to evaluate associations between smoking and offspring outcomes while controlling for relevant covariates. Results: Birth Weight: ○ Maternal smoking during pregnancy was significantly associated with lower birth weight in both related and unrelated pregnancies. ○ This association persisted even when analysing only singleton births, indicating a potential causal role for prenatal smoking in reducing birth weight. Offspring Antisocial Behavior: ○ Overall, maternal smoking was associated with higher antisocial behavior in offspring. ○ However, this association was observed only in the related group, not in pregnancies where the gestating mother was unrelated to the child. ○ This suggests that inherited factors, not prenatal effects, explain the association between maternal smoking and antisocial behavior. Critical Analysis: Strengths: ○ Innovative Methodology: Utilizing IVF-conceived children with unrelated maternal gestation allowed for distinguishing between prenatal effects and inherited factors. ○ Clear Outcome Patterns: The study found consistent associations between prenatal smoking and birth weight, showcasing potential causal effects. ○ Implications for Public Health: Clarification of true prenatal effects can aid in designing targeted interventions for maternal smoking cessation. Limitations: ○ Sample Size: Limited sample size, especially in the group with unrelated pregnancies, might affect the generalizability of findings. ○ Self-Reporting Bias: Reliance on maternal reports of smoking during pregnancy may introduce recall or reporting biases. ○ Complexity of Behavioral Outcomes: Antisocial behavior is multifaceted and influenced by various environmental and genetic factors, which might not be fully accounted for in this study. Conclusion: The study offers valuable insights into the impact of maternal smoking during pregnancy on birth weight and antisocial behavior. Clear associations between smoking and birth weight support the existence of true prenatal effects. However, the lack of association with antisocial behavior in unrelated pregnancies indicates that inherited factors likely explain this outcome, challenging the notion of a direct causal relationship. The study's methodology provides a unique approach to disentangle prenatal and inherited influences, contributing significantly to understanding developmental pathways influenced by maternal behaviors during pregnancy. The implications of genotype-environment correlation for establishing causal processes in psychology (Jaffee & Price, 2012) Background: The study focuses on the implications of genotype-environment correlation for understanding causal processes in psychopathology. It discusses the challenges of establishing causality in observational research due to factors like reverse causation and confounding. The research explores different types of genotype-environment correlations - passive, evocative, and active - and their impact on behavioral genetics. Methods: The study reviews evidence from quantitative behavioral genetics and molecular genetic literature, emphasizing the heritability of environmental measures and the potential confounding effects of genotype. It delves into the use of genotype-environment correlations to demonstrate causal mechanisms, specifically through the concept of Mendelian randomization (MR). The research designs discussed to address genotype-environment correlations include twin studies, adoption studies, and the Children of Twins design. Results: The study highlights the challenges associated with using MR, especially in identifying plausible genetic proxies for specific exposures. It emphasizes the need for replicated findings and the potential for MR to expand as more genetic determinants are identified. The document underscores the need to address genotype-environment correlations in research to avoid misidentification of causal agents in the relationship between exposures and outcomes. Critical Analysis: Strengths: ○ The study provides a comprehensive review of the implications of genotype-environment correlation for understanding causal processes in psychopathology. ○ It offers insights into the challenges and potential of using Mendelian randomization and other research designs to address genotype-environment correlations. PSYC0036 Genes and Behaviour Page 6 genotype-environment correlations. Limitations: ○ The study acknowledges the limitations of using MR, especially in identifying plausible genetic proxies for specific exposures. ○ It also highlights the challenges associated with replicating findings and the need for more research to address genotypeenvironment correlations. Overall, the study contributes to the field by shedding light on the complexities of genotype-environment correlations and their impact on causal inference in psychopathology research. It underscores the need for rigorous research designs and the careful consideration of limitations in studying genotype-environment correlations. Strategy for Investigating Interactions Between Measured Genes and Measured environments Background: The study focuses on gene-environment interactions (GxE) in relation to psychopathology outcomes, specifically in mental disorders. The research team conducted three studies to investigate GxE and its feasibility as a research strategy. (Moffitt et al., 2005) Methods: The first study hypothesized that a functional polymorphism in the gene encoding the neurotransmitter-metabolizing enzyme monoamine oxidase A would moderate the effect of child maltreatment in the cycle of violence. The researchers used a hypothesis-driven approach, consulting quantitative behavioral-genetic studies, identifying candidate susceptibility genes, and optimizing environmental risk measurement. They collected retrospective reports of exposure to environmental pathogens and used the life history calendar method to improve the quality of retrospective data. Results: The findings of the study showed that maltreated children with a genotype conferring low levels of monoamine oxidase A expression were more likely to develop conduct disorder and antisocial behavior. The study provided proof of principle that GxE occurs in relation to psychopathology outcomes and illustrated the feasibility of the GxE research strategy. Critical Analysis: Strengths: ○ The study demonstrated the potential of GxE research in understanding the complex nature of genetic and environmental influences on mental disorders. ○ It provided evidence of the causal pathway connecting genes, environmental pathogens, and disorders, contributing to the field's understanding of disorder-specific causal mechanisms. Limitations: ○ The retrospective nature of data collection and the reliance on self-reported environmental pathogen exposure may introduce recall bias and confounding factors. ○ Additionally, the study's focus on specific genetic and environmental factors may limit the generalizability of the findings to other mental disorders and populations. Overall, the study's findings contribute to the field by highlighting the importance of considering gene-environment interactions in psychiatric genetics and providing a strategic approach for conducting hypothesis-driven GxE studies. However, the limitations in data collection and generalizability should be considered when interpreting the results. Concurrent and Longitudinal Contribution of Exposure to Bullying in Childhood to Mental Background: Health The Role of Vulnerability and Resilience Childhood exposure to bullying is prevalent, with estimates suggesting that one-third of children report being bullied by their peers, leading to adverse mental health outcomes. Research faces challenges in establishing causality due to correlational studies and the influence of genetic factors predisposing individuals (Singham et al., 2017) to both being bullied and experiencing mental health issues. Ethical constraints limit experimental designs that allocate children to different levels of bullying exposure. The study aimed to examine the causal relationship between childhood exposure to bullying and mental health difficulties using a genetically informative design. Methods: Participants were drawn from the Twins Early Development Study (TEDS) born in England and Wales between 1994 and 1996, totaling 11,108 twins. Childhood exposure to bullying and various mental health outcomes were measured using multidimensional scales and questionnaires at ages 11, 14, and 16. Statistical analyses included phenotypic estimates, differences in DZ and MZ twins, and longitudinal assessments to examine the relationship between exposure to bullying and mental health outcomes. Results: Findings indicated significant associations between childhood exposure to bullying and mental health outcomes across informants and scales, with phenotypic associations remaining largely significant. Strong concurrent contributions of exposure to bullying were seen in depression, anxiety, conduct problems, hyperactivity, and inattention, particularly in MZ twin analyses. Longitudinal analyses revealed a decrease in the direct contribution of bullying exposure to mental health difficulties over time, indicating potential resilience among children. Critical Analysis: Strengths: ○ Rigorous use of a genetically informative design in a large prospective study. ○ Multidimensional measures of bullying exposure and comprehensive assessments of mental health outcomes across different informants and scales. ○ Robust evidence of immediate adverse effects of bullying on mental health, highlighting the need for targeted interventions. Limitations: ○ Inability to account for all nonshared environmental confounding factors. ○ Lack of control for certain forms of bullying and some mental health outcomes. ○ Potential bias due to attrition and inability to assess long-term causal relationships comprehensively. Implications: ○ Suggests a need for interventions addressing both primary prevention of bullying and secondary strategies focusing on preexisting vulnerabilities in children exposed to bullying for long-term mental health improvements. ○ Calls for further investigation into mechanisms of resilience and protective factors to aid in targeted interventions. Widespread covariation of early environmental exposures and trait-associated polygenic Background: variation Environmental Exposures as Predictors of Developmental Outcomes: (Krapohl et al., 2017) ○ Maternal smoking during pregnancy, socioeconomic status, television/video game exposure, and harsh parental discipline correlate with lower academic achievement and increased emotional/behavioral problems in children. ○ Paternal age affects various disorders and academic achievement, particularly linked strongly to autism spectrum disorders and schizophrenia. ○ Breastfeeding and higher parental socioeconomic status are protective factors for educational achievement. Genotype-Environment Correlation: ○ Genetic influences affect individuals' exposure to environments (genotype-environment correlation). ○ Parenting characteristics and socioeconomic variables are partially heritable. ○ Recent research explores genetic contributions to correlations between environmental factors and children's outcomes, including new designs to separate genetic and environmental influences. SNP-Heritability Studies: ○ Studies show that genetic variants influence social deprivation, income, stressful life events, and socioeconomic status. ○ Some reports extend SNP heritability analysis to estimate genetic correlations between environmental measures and children’s developmental outcomes. Gap in Polygenic Prediction Models: PSYC0036 Genes and Behaviour Page 7 Gap in Polygenic Prediction Models: ○ Polygenic prediction models often ignore the possibility that genetic effects on traits might capture environmental risk or protective factors. ○ These models typically focus on gene-environment interaction rather than gene-environment correlation. Methods: Study Sample: ○ Utilized 6,710 unrelated individuals from the Twins Early Development Study (TEDS), representative of the UK population. ○ Collected genotype data, a wide range of environmental exposure measures, and developmental outcomes from birth to adolescence. Polygenic Scores: ○ Constructed genome-wide polygenic scores for educational attainment, body mass index (BMI), and schizophrenia using GWAS summary statistics. ○ Employed a Bayesian approach to estimate the effect size of each marker based on a point-normal mixture prior on effect sizes and linkage disequilibrium information. Statistical Analysis: ○ Applied single-score models to estimate the univariate effect of each polygenic score on environmental exposures. ○ Used multiscore models to isolate the effects of each polygenic score while adjusting for the effects of others. ○ Controlled for population stratification by estimating polygenic effects while considering overall genetic relatedness through a genomic relatedness matrix. Results: Association Findings: ○ Significant associations between polygenic scores for education, BMI, and schizophrenia and various environmental measures were observed. ○ Education-associated polygenic variation correlated with breastfeeding, maternal smoking during pregnancy, household income, parental education level, parental discipline, and television watching. ○ Paternal age was positively associated with offspring genetic risk for schizophrenia. Polygenic Variation and Developmental Outcomes: ○ Education-associated polygenic variation explained a notable proportion (15%) of the associations between environmental measures and developmental outcomes. ○ Specifically, the education polygenic score significantly explained the covariance between various environmental factors and children's developmental outcomes. Critical Analysis: Strengths: ○ Integration of Family and Molecular Data: ▪ The study combines family and molecular data to explore gene-environment correlations. ○ Exploration of Polygenic Score Effects: ▪ Identifies associations between polygenic scores for education, BMI, and schizophrenia and diverse environmental exposures. ○ Explanation of Variance in Developmental Outcomes: ▪ Demonstrates the contribution of education-associated polygenic variation to explaining associations between environmental exposures and developmental outcomes. Limitations: ○ Inability to Distinguish Mechanisms: ▪ Unable to differentiate between passive, active gene-environment correlation, or environmentally mediated genetic effects due to study design limitations. ○ Parental Genotype Data Not Included: ▪ Lack of parental genotype data restricts the ability to disentangle cross -generational effects of genetics and environment. ○ Incomplete Capture of Genetic Effects: ▪ Polygenic scores might underestimate the total genetic effects on exposure -outcome associations due to limitations in detecting all genetic variants' additive effects. Conclusion: ○ Implications for Polygenic Prediction Models: ▪ Genetic variation captured by GWASs partly encompasses environmental risk or protective factors, suggesting potential improvements in polygenic prediction models by incorporating such genetic variants. ○ Relevance of Gene-Environment Correlation: ▪ Emphasizes that genetic variation identified by GWASs not only influences traits but also environmental factors, blurring the conventional dichotomy between traits and environments. ○ Importance of Integrating Family and Molecular Data: ▪ Illustrates the significance of combining family and molecular data to unravel mechanisms through which genetic variation translates into phenotypic variation. Methodological Insights: ○ Polygenic Score Construction: ▪ Employed a Bayesian approach and controlled for population stratification, ensuring robustness in estimating polygenic effects. ○ Decomposition Analysis: ▪ Used structural equation models to decompose covariance between environmental exposures and developmental outcomes, providing insights into polygenic contributions. Future Directions: Incorporation of Parental Genotype Data: Suggested using samples with parental genotype data to disentangle cross-generational effects of genetics and environment. Mechanistic Investigations: Calls for exploring mechanisms behind gene-environment correlations by elucidating the roles of passive, active, and environmentally mediated genetic effects. Using genetic data to strengthen causal inference in observational research (Pingault et al., 2018) Challenges to causal inference Confounding: a variable causes both the risk factor and the outcome, generating a spurious association. Genetic confounding: genetic factors affect both the exposure and the outcome. Gene-environment correlation: the environment experienced by an individual is partly influenced by their genotypes. Passive gene-environment correlation: children inherit parental genetic variants that contribute to the environment that parents create. Reverse causation: the direction of the causal relationship between risk factors and outcomes is unclear. Measurement error: imprecise measures of the exposure or the outcome can hinder the detection of causal effects. Misidentification: the putative causal risk factor is not the true causal risk factor. Genetically informed methods Family-based methods: exploit the expected degree of genetic similarity for different types of relationships. Mendelian randomization (MR): uses genetic variants as instrumental variables to strengthen causal inference. Embedding genetic instruments within family-based designs: integrates these two methods. Phenome-wide approaches to causal inference: uses genetic variation data to dissect the genetic architecture of phenotypes. Key figures 2 major observational studies concluded that higher consumption of vitamin E reduces risk of CHD. RCTs are inefficient in the absence of reliable evidence to prioritize targets. Low drug development success rates result in US$2.6 billion costs per approved drug. The genome is randomized at conception, which is critical in MR. Massive genotyped and phenotyped data sets have expanded the applicability of these methods. Novel informatics tools allow data mining of these resources at phenome-wide scale. PSYC0036 Genes and Behaviour Page 8 Novel informatics tools allow data mining of these resources at phenome-wide scale. A causal inference framework The counterfactual (also known as potential outcomes) approach is a unifying framework for causal inference. To attain consistent causal inference, achieving or sufficiently approximating exchangeability is essential. Directed acyclic graphs (DAGs) provide a formal yet intuitive representation of causal inference. Family-based designs Family-based designs can exploit genetic relatedness to control for genetic confounding. Sibling and twin designs are commonly used to approximate the counterfactual situation. Siblings share 50% of their genetic material, while twins share 100%. MZ twins are ideal for controlling for genetic confounding as they share all of their genetic material. Sibling and twin designs are limited in that they cannot account for non-shared environmental confounding. Adoption-at-birth and IVF designs can control for genetic confounding but not for environmental confounding. The DoC model can be used to investigate causal relationships using A(D)CE components as instruments. Key statistics Smoking during pregnancy is associated with lower birthweight. Maternal genetic factors contribute to smoking during pregnancy and birthweight. Smoking during pregnancy is predictive of lower birthweight in both genetically related and unrelated mother–child dyads. Mendelian Randomization Mendelian randomization (MR) is a method used to infer causal relationships between exposures and outcomes using genetic variants as instruments. Genetic variants are associated with exposures and can be used to approximate the counterfactual situation in which individuals are exposed to different levels of the exposure. MR is based on the assumption that genetic variants are randomly assigned during conception and therefore are not associated with confounding factors. MR has been used to investigate a wide range of causal relationships, including the effects of LDL-C on CHD, CRP on multiple phenotypes, and cannabis use on schizophrenia. Extensions of Mendelian Randomization Dealing with imperfect instruments: MR can be used with multiple genetic variants as instruments to increase power and mitigate problems caused by weak instruments. Bidirectional MR: MR can be used to investigate reciprocal causal relationships between two phenotypes. Multivariable MR: MR can be used to consider multiple exposures simultaneously to model possible pleiotropic pathways that would violate MR assumptions. Intergenerational MR: MR can be used to investigate causal relationships between exposures and outcomes in parents and offspring. Emerging Approaches Polygenic scores: MR can be used with polygenic scores, which are scores that measure the genetic predisposition to a phenotype. Integrating MR with family-based designs: MR can be combined with family-based designs to control for dynastic effects and pleiotropy. Phenome-wide Approaches and Shared Aetiology Investigating Relationships: Summary association statistics enable the exploration of connections between thousands of traits. Genetic Correlations: Quantify the shared genetic basis between different traits. Importance of Shared Aetiology: Shared genetic roots between traits are sometimes more crucial than strict causality, especially in cases where intervening on an exposure isn't feasible. Identification of Shared Loci: Phenome-wide studies help identify genetic variants influencing multiple traits. Colocalization Methods: Assist in identifying regions where variants associate with multiple traits, indicating common causal pathways. Understanding Cross-Phenotype Relationships: Asymmetry analysis and colocalization tests help distinguish between causal relationships and shared pathways among phenotypes. Dissecting Exposures and Delineating Pathways Heterogeneous Exposures: Some traits, like BMI, can be broken down into subcomponents for more targeted causal effects. Pathway Mapping: Mapping out pathways between exposures and outcomes reveals additional targets for intervention. Causal Analyses as Maps: Rather than one-to-one relationships, causal analyses represent complex networks between various phenotypes. Conclusions and Future Perspectives Toolbox of Genetically Informed Methods: Varied methods aid in different research scenarios and data types, offering options like twin designs, MR, and colocalization. Integration for Research Development: Future improvements involve integrating multiple methods to identify shared pathways among different phenotypes. Methodological Advances: Continued progress is expected in refining MR estimators, utilizing genome-wide information, and integrating genetic instruments with family-based designs. Expansion Across Disciplines: Causal inference methods are expanding their applications beyond traditional fields into social sciences and economics, offering significant benefits. Pitfalls of Causal Inference Sensitivity Analysis Using Genome-wide Data: Proposes a method to assess genetic confounding when adequate genetic instruments are unavailable for complex phenotypes. Triangulation for Strengthening Evidence: The importance of converging conclusions from multiple study designs to strengthen evidence for causality. Conclusions Rapid Developments: Genetically informed causal inference methods have rapidly evolved, offering significant benefits across various research fields. Credible Conclusions: Emphasizes that conclusions drawn from these methods depend on sound modelling decisions and credible assumptions, urging the need for convergence in conclusions. Association Between Genetic Risk for Psychiatric Disorders and the Probability of Living in Urban Settings This study investigates the relationship between genetic predispositions for various psychiatric disorders and population den sity across an individual's lifetime. The research aims to elucidate whether genetic liability influences an individual's choice of living e nvironment, particularly in urban or rural settings, and to explore the potential impact of this genetic predisposition on the risk of psychiatric dis orders. (Maxwell et al., 2021) Background: Urban living has long been associated with an increased risk of schizophrenia, primarily observed in northern European countries. However, contradictory findings in other regions, such as low- and middle-income or southern European countries, challenge this association. Understanding the interplay between genetic predisposition, environmental factors, and psychiatric disorders like schizophrenia is essential. The study seeks to investigate the genetic correlations between psychiatric disorders and population density at birth and during an individual's lifetime. Methods: The research employs data from the UK Biobank, focusing on individuals of European ancestry to avoid confounding factors. Various genetic analyses are conducted, including Polygenic Risk Score (PRS) analyses, Genome-Wide Association Studies (GWAS), SNV heritability estimations, and Mendelian Randomization (MR) analyses. Population density is derived from historical census data linked to participants' address history. Results: The study found significant associations between genetic predispositions for schizophrenia, bipolar disorder, autism spectrum disorder, and anorexia nervosa with increased population density across adulthood. Conversely, a genetic predisposition for ADHD was associated with decreased population density. Analysis of migration patterns indicated that individuals with higher genetic risk for certain psychiatric disorders were more likely to move from rural to urban areas. GWAS identified specific genetic loci associated with birth and current population density, albeit with low heritability. Critical Analysis: PSYC0036 Genes and Behaviour Page 9 Strengths: ○ The study uses a large and diverse dataset from the UK Biobank, enabling robust genetic analyses. ○ It integrates various genetic methodologies to explore the complex relationship between genetics, migration, and urbanicity. ○ Findings provide insight into the nuanced interplay between genetic predisposition and environmental factors, particularly in urban settings. Limitations: ○ The study's reliance on data from the UK Biobank might introduce selection bias, potentially limiting the generalizability of findings. ○ Low heritability in some analyses suggests the need for larger sample sizes or more comprehensive genetic datasets to capture subtle effects. ○ The complexity of associations among population density, socioeconomic status, and educational attainment may confound the results. Conclusion: ○ The study suggests that genetic predispositions for certain psychiatric disorders influence an individual's choice of living environment, potentially contributing to the association between urban living and mental health disorders. ○ However, it underscores the complexity of this relationship, emphasizing the need for more comprehensive approaches to understand the causative factors behind mental health disorders in urban settings. ○ While environmental factors remain crucial, the study's findings highlight the importance of considering genetic predisposition in understanding mental health risks associated with urban living. G = E: What GWAS Can Tell Us about the Environment (Gage et al., 2016) Background: The study discusses the implications of Genome-Wide Association Studies (GWAS) in identifying not only genetic factors linked to behavioral traits but also causal effects of modifiable or environmental influences on these traits. It emphasizes that some GWAS-identified loci might capture modifiable risk factors rather than direct biological pathways. Methods: The study employs the principles of Mendelian randomization, using genetic variants as instrumental variables for modifiable exposures, to understand causal relationships between these exposures and disease outcomes. It utilizes GWAS data on behavioral phenotypes like tobacco and alcohol use to illustrate how these associations might reflect causal effects. Results: Tobacco Use: The CHRNA5-A3-B4 locus is associated with smoking quantity and diseases like lung cancer. This locus not only influences smoking behavior but is likely linked to diseases via tobacco exposure. Alcohol Use: ALDH2 variants influence alcohol consumption and, subsequently, conditions like high blood pressure and oesophageal cancer, especially in East Asian populations. Critical Analysis: Strengths: ○ Insights on Causal Relationships: The study highlights how GWAS can reveal causality between behavioral phenotypes (like smoking or drinking) and diseases, aiding public health interventions. ○ Utilization of Mendelian Randomization: Leveraging genetic variants as proxies for exposures strengthens causal inference. ○ Identification of Modifiable Risk Factors: It emphasizes the importance of distinguishing between direct biological pathways and modifiable risk factors in GWAS findings. Limitations: ○ Need for Larger Sample Sizes: Weak genetic instruments often require large sample sizes for adequate statistical power in Mendelian randomization studies. ○ Misclassification Bias: Issues like misreporting of smoking status can affect results, leading to potential biases. ○ Complexity in Causal Inference: Distinguishing between mediated and biological pleiotropy poses challenges in establishing direct causal relationships. ○ Contribution to the Field: ▪ Understanding Disease Etiology: Helps distinguish between genetic predisposition and environmental factors in disease development. ▪ Potential for Public Health Impact: Identifies behavioral phenotypes' role in disease outcomes, aiding preventive strategies. Conclusion: ○ The study emphasizes the need for a cautious approach in interpreting GWAS findings, considering both genetic and environmental influences. It advocates for triangulating evidence from various approaches to ascertain causal relationships, thereby enabli ng more precise interventions and treatments. ○ This comprehensive analysis of GWAS data and causal inference sheds light on the intricate relationships between genetics, behavior, and disease, urging a shift towards recognizing and addressing modifiable environmental factors in the context of disease etiology and prevention. Early Cannabis Use, Polygenic Risk Score for Schizophrenia, and Brain Maturation in Adolescence (French et al., 2015) Introduction The study investigates the association between cannabis use during early adolescence and variations in brain maturation among typically developing youth, focusing particularly on the influence of genetic risk for schizophrenia. The research is carried out using three separate population-based samples: the Saguenay Youth Study (SYS) in Canada, the Avon Longitudinal Study of Parents and Children (ALSPAC) in England, and the IMAGEN Study across eight European cities. The study incorporates data on cannabis use, structural brain imaging using MRI, and the polygenic risk score for schizophrenia. Background: Cannabis is a prevalent illicit substance globally, especially among adolescents. Adolescence is marked by significant physiological and social changes that impact brain development, as observed in various MRI studies. There are known sex differences in brain maturation during this phase, potentially influencing vulnerability to external factors like cannabis exposure. Given the link between cannabis use during adolescence and schizophrenia, the study aims to explore whether early cannabis use (by 16 years) correlates with brain maturation variations based on genetic schizophrenia risk. Methods: The research involves three distinct population-based samples across Canada, England, and Europe, utilizing data from MRI brain scans, information on cannabis use, and polygenic risk scores for schizophrenia. Participants are categorized based on cannabis use by age 16, and cortical thickness is measured as a primary dependent variable. Results: SYS Sample: Among male adolescents, there's a correlation between cannabis use and a decrease in cortical thickness in individuals with a high genetic risk for schizophrenia. This correlation isn't evident in non-users or those with low genetic risk. IMAGEN Sample: Similar findings in male participants between cannabis use frequency and cortical thickness change during adolescence. However, in female participants, the results are less pronounced. ALSPAC Sample: Among male youths with high genetic risk, frequent cannabis users show a decrease in cortical thickness compared to non-users or light users. No substantial differences are observed in low-risk male participants. Critical Analysis: Strengths: ○ Population-based Approach: Utilizing multiple samples across diverse geographic locations adds robustness to the findings. ○ Longitudinal Data: The inclusion of longitudinal MRI data enhances the study's credibility in identifying potential causal relationships between cannabis use and brain changes. ○ Genetic Component: Incorporating polygenic risk scores for schizophrenia provides insight into the interplay between genetic predisposition and cannabis use in influencing brain development. Limitations: ○ Observational Nature: The study doesn't establish causality between cannabis use and brain changes but rather identifies associations. It's challenging to rule out reverse causation or other confounding factors. ○ Sample Variations: Inconsistencies in results across samples (especially ALSPAC) could be due to sample size or other unaccounted variables, limiting the generalizability of the findings. ○ Gender Discrepancies: Findings are predominantly observed in males with high genetic risk, limiting the broader applicability of the results to females or low-risk individuals. PSYC0036 Genes and Behaviour Page 10 Implications: ○ Schizophrenia Vulnerability: The study suggests a potential link between cannabis use in early adolescence, genetic risk for schizophrenia, and altered cortical thickness in males. This might indicate a vulnerability pathway for schizophrenia development. ○ Gender-Specific Effects: The findings highlight the need to consider gender-specific vulnerabilities concerning cannabis use and brain maturation, particularly in males at high genetic risk. Conclusion: ○ The study implies that cannabis use during early adolescence might influence cortical maturation among males with a higher genetic risk for schizophrenia. ○ These findings offer critical insights into the complex interplay between cannabis exposure, genetic predisposition, and brain development, emphasizing the need for further experimental studies to confirm causality and inform preventive strategies or interventions in vulnerable populations. Commentary: Will genomics revolutionise research on gene–environment interplay (Plomin & Viding. 2022) Background: The study revolves around the convergence of two major domains in genetics: quantitative genetics and molecular genetics. For a long time, molecular geneticists focused on understanding genes, while quantitative geneticists studied phenotypes. The historical context outlines the shift from investigating mutations and dichotomous disorders to exploring naturally occurring variations and continuous traits. Methods: Quantitative Genomics: Utilizes technological advancements in genotyping, specifically the DNA microarray, to conduct genome-wide association studies (GWAS) and identify numerous DNA variants associated with various complex traits. Polygenic Scores: Aggregates thousands of small-effect DNA variants to create scores reflecting genetic predispositions, allowing the assessment of genetic influence directly without relying on twins or adoptees. Quantitative Genomic Estimates: Utilizes methods like GCTA or GREML to estimate SNP heritability and genetic correlations directly from SNP genotypes, offering insights into the genetic basis of traits. Results: Polygenic Scores: Used in several papers within the study to investigate correlations and interactions with environmental factors, shedding light on gene–environment correlation. Genetic Correlations: Estimation of genetic correlations between traits using GCTA, uncovering insights into the influence of genetic and environmental interactions on behavioral traits. Critical Analysis: Strengths: ○ Technological Advances: The study highlights the transformative impact of technological breakthroughs like the DNA microarray and SNP chips, enabling large-scale genomic studies. ○ Polygenic Scores: Offering a direct measure of genetic influence without relying on traditional methods like twin studies. Limitations: ○ Complexity of Causal Modeling: The study emphasizes the challenges in deriving causality from correlational data, especially in the context of gene–environment interplay. ○ Sample Sizes and Replication: The historical context highlights the underpowered nature of earlier studies using candidate genes and emphasizes the need for replication in genomic research. Contribution to the Field: ○ Understanding Developmental Dynamics: The study suggests a shift toward understanding gene–environment correlation, which has considerable potential to unravel developmental dynamics critical for mental health. Conclusion: ○ The fusion of quantitative and molecular genetics has brought about remarkable advances, particularly in the ability to measure inherited DNA differences directly, leading to the emergence of polygenic scores and new methods for estimating heritability and genetic correlations. ○ The study stresses the importance of examining gene–environment interaction findings to ensure they are not influenced by gene– environment correlation. ○ While acknowledging the complexities, the study advocates for utilizing polygenic scores for prediction and intervention, offering insights into potential avenues for personalized interventions in mental health. Identifying the environmental causes of disease: how should we decide what to believe and when to take action? Context: The report addresses the challenge of identifying environmental causes of diseases and the decision-making process regarding believing and taking action on such causes. (Rutter, 2017) Key Points: Importance of Evidence: ○ Emphasizes the significance of credible evidence in determining environmental causes of diseases. ○ Advocates for a cautious approach, acknowledging the complexity of identifying causal factors. Evaluating Research Findings: ○ Highlights the need to critically assess research findings and the methods used in studies exploring environmental causes. ○ Stresses the importance of replication and consistency across studies to establish reliable causal links. Weight of Evidence: ○ Discusses the concept of "weight of evidence" in determining the strength of environmental causation, considering factors like consistency, coherence, and biological plausibility. Decision-Making and Action: ○ Addresses the challenge of determining when to take action based on identified environmental causes, balancing potential risks and benefits of interventions. Public Policy Implications: ○ Considers the role of these findings in shaping public policy and interventions related to disease prevention and health promotion. ○ Advocates for a cautious approach in policy decisions, considering the uncertainty inherent in identifying environmental causes. Conclusion: Overall, stresses the need for a rigorous evaluation of evidence and a cautious approach to determine environmental causes of diseases before taking significant action or implementing policies. Summary: Rutter's report underscores the importance of credible evidence, critical evaluation of research findings, and a cautious approach in identifying and acting upon environmental causes of diseases. It emphasizes the complexity of this task and the necessity of robust evidence before initiating interventions or policy changes. PSYC0036 Genes and Behaviour Page 11