Family, twin and adoption studies demonstrate that most behavioural phenotypes are influenced by genetic variation and some substantially. This includes phenotypes where environmental influences are thought to be important, including socially influenced behaviours such as the frequency and quantity of alcohol consumption. Laucht et al. (2012) present a study investigating the influences of both environmental measures (parental supervision, involvement and autonomy granting) and genetic effects (COMT Val158Met genotype) on drinking behaviour (frequency of drinking and average daily consumption) at the age of 19. After controlling for potential confounding factors including gender, alcohol use at age 15, parental alcohol use disorder, adolescent internalising and externalising symptoms and temperament, they found no main effects of either parental behaviour or COMT genotype on adolescent drinking behaviour. However, they did find evidence of significant Gene × Environment interaction.

The pattern of the interaction reported in this study is consistent with the hypothesis of differential susceptibility, which proposes that some individuals by virtue of their genotype are more responsive to environmental influences, whereas others are relatively unresponsive (Ellis, Boyce, Belsky, Bakermans-Kranenburg, & Van Ijzendoorn, 2011). In the example reported here, Laucht and colleagues found that Met/Met homozygotes had either good or bad outcomes, depending on the quality of parental supervision and involvement. In contrast, Val carriers showed no significant effects of the environmental measures on risk for drinking behaviour. These are potentially important findings if it turns out that some individuals are more sensitive to their parenting environment, for better or for worse, when it comes to drinking behaviour. The study conclusions emphasise the need for parents to be aware of the need for appropriate parental supervision and guidance in determining both positive and negative outcomes for their children; but also suggests that this is more important for some individuals than others, depending on underlying genetic and neurobiological factors that lead to differential sensitivity to the parenting environment.

The question then arises as to how we would use this information to improve health outcomes? Can we envisage a time when we would use genetic testing as a way to target behavioural strategies towards individual children, or to identify those children for whom we need to pay more attention to environmental risks than others? Given the low predictive power of nearly all genetic findings on behavioural outcomes, we need to be very cautious in presenting such an argument, although if the case was proven for such a mechanism, we might devise a direct (nongenetic) test of environmental sensitivity. However, the strength of the finding may not lie in such direct testing of individual sensitivities to the environment, but rather in the use of genetic data to identify an important environmental effect in the absence of a main effect of the environmental risk, as reported here for the effect of parental supervision on offspring drinking behaviour. This at least establishes the importance of parenting behaviour on offspring outcomes and supports the further development of programs that seek to enhance positive aspects of parenting.

However, before we start to think further about the application of this new knowledge, we need to consider the veracity of the findings. Laucht and colleagues interpretation of their results as evidence of differential susceptibility rests on four criteria: (a) evidence of a crossover interaction between genotype and parenting, (b) independence of the genotype and parenting, (c) no association between genotype and drinking, and (d) significant regression slope of drinking on parenting for the susceptible genotype, but not for the nonsusceptible genotype. Evidence of a crossover interaction – namely, one in which the regression slopes of phenotype on environment, stratified by genotype, cross over within the range of the observed environment – is crucial for establishing differential susceptibility rather than the alternative mechanism of genetic vulnerability (Dick, 2011). This criterion, however, is not a very exact one: the same set of regression coefficients may result in lines that cross, or not, depending on the breadth of the environmental measure surveyed. We note, for example, that Laucht and colleagues oversampled individuals born at the risky end of the environmental spectrum, perhaps increasing their chances of detecting a crossover interaction; future studies may find it harder to detect such an interaction unless the environmental measures cover both positive and negative effects. Statistical power is also a consideration, given that the study used a relatively small sample. The fact that greater power is required to detect crossover interactions than interactions of other types means that we would expect a larger proportion of crossover effects to be detected as a statistical artefact (Dick, 2011).

More generally there has been considerable controversy about the interpretation of gene by environment interactions on behavioural phenotypes. A case in point is the reported interaction between genetic variation in the promoter region of the serotonin transporter gene (5-HTT) and exposure to environmental adversity, leading to increased risk of developing depression (Caspi, Hariri, Holmes, Uher, & Moffitt, 2010; Risch et al., 2009). The difficulties in performing a literature based meta-analysis of the statistical interaction have highlighted the need for high quality measurement of environmental exposures. Perhaps of greater significance, in terms of the underlying mechanism, is the converging evidence that has accrued for this particular finding from human observational studies, experimental neuroscience and nonhuman primate and other animal models.

The 5-HTT finding provides considerable encouragement for gene–environment research (McGuffin, Alsabban, & Uher, 2011). Nevertheless extreme caution should be taken towards all preliminary reports of candidate gene associations, particular where a priori expectations are not clearly specified and where replication studies have yet to be completed. Several steps are required to identify reproducible and therefore believable findings, starting with the replication of initial reports in independent samples. There are many potential pitfalls that have led to the candidate gene literature being littered with unsubstantiated findings that in most cases fall by the wayside as further data are accumulated; and there is no reason to believe that gene–environmental research is immune from this problem. One major reason for this relates to the frequent occurrence of nominally significant associations and inadequate adjustment being made for multiple testing when a number of putative risk alleles (candidate genes) are used to search for associations across multiple phenotypes and risk models. The extent to which any candidate has better than even odds of being associated with any behavioural phenotype is unclear, so setting higher a priori odds relies mainly on intuition. Gene–environment strategies present researchers with further multiple testing problems that are exacerbated when multiple gene by environment models are evaluated.

So in the light of this discussion what can we say about the findings reported by Laucht and colleagues. As discussed, these are potentially interesting findings but they are not yet firmly established. The COMT gene does however appear to be a good candidate because we know that the polymorphism investigated has effects on dopamine levels and is associated with altered function of the prefrontal cortex and other brain regions including those related to reward pathways and aspects of self control (Camara et al., 2010; Kaenmaki et al., 2010; Meyer-Lindenberg et al., 2005; Savitz, Solms, & Ramesar, 2006).

A recent finding reported in four independent samples (plus further unpublished datasets) is the association of the high activity low dopamine Val/Val variant with conduct problems and antisocial behaviour in children with ADHD, but not in controls (Caspi et al., 2008; Langley, Heron, O’Donovan, Owen, & Thapar, 2010). These data appear to firmly establish the influence of the COMT Val158Met polymorphism on behavioural outcomes. In the search for processes that mediated this association, social understanding was found to be an important factor (Langley et al., 2010), which could link back to the effects of adequate parental engagement and supervision reported here. Unfortunately, we would have to explain why the findings reported by Laucht et al. on drinking behaviour are with the opposite high dopamine Met/Met genotype to the low dopamine Val/Val genotype associated with antisocial behaviour in children with ADHD. Could it be that alterations of dopamine regulation in either direction confer risks on different behavioural phenotypes? These apparently contradictory findings with regard to the direction of effect might for example be explained by previous reports that the required level of dopamine for optimal function follows an inverted U shaped distribution (Mattay et al., 2003). Overall research on COMT Val158Met continues to provide evidence of functional significance, with effects on important human behaviours. Further work to understand the interactions with genetic and environmental risk factors is clearly indicated by reports on this gene.


  1. Top of page
  2. Acknowledgement
  3. Correspondence
  4. References

This Commentary article was contributed by invitation of the JCPP Editors.


  1. Top of page
  2. Acknowledgement
  3. Correspondence
  4. References

Philip Asherson, MRC Social Genetic and Developmental Psychiatry, Institute of Psychiatry, Kings College London, London SE5 8AF, UK; Email:


  1. Top of page
  2. Acknowledgement
  3. Correspondence
  4. References
  • Camara, E., Kramer, U.M., Cunillera, T., Marco-Pallares, J., Cucurell, D., Nager, W., … & Munte, T.F. (2010). The effects of COMT (Val108/158Met) and DRD4 (SNP -521) dopamine genotypes on brain activations related to valence and magnitude of rewards. Cerebral Cortex, 20, 19851996.
  • Caspi, A., Hariri, A.R., Holmes, A., Uher, R., & Moffitt, T.E. (2010). Genetic sensitivity to the environment: The case of the serotonin transporter gene and its implications for studying complex diseases and traits. American Journal of Psychiatry, 167, 509527.
  • Caspi, A., Langley, K., Milne, B., Moffitt, T.E., O’Donovan, M., Owen, M.J., … & Thapar, A. (2008). A replicated molecular genetic basis for subtyping antisocial behavior in children with attention-deficit/hyperactivity disorder. Archives of General Psychiatry, 65, 203210.
  • Dick, D.M. (2011). Gene-environment interaction in psychological traits and disorders. Annual Review of Clinical Psychology, 7, 383409.
  • Ellis, B.J., Boyce, W.T., Belsky, J., Bakermans-Kranenburg, M.J., & Van Ijzendoorn, M.H. (2011). Differential susceptibility to the environment: An evolutionary–neurodevelopmental theory. Development and Psychopathology, 23, 728.
  • Kaenmaki, M., Tammimaki, A., Myohanen, T., Pakarinen, K., Amberg, C., Karayiorgou, M., Gogos, J.A., & Mannisto, P.T. (2010). Quantitative role of COMT in dopamine clearance in the prefrontal cortex of freely moving mice. Journal of Neurochemistry, 114, 17451755.
  • Langley, K., Heron, J., O’Donovan, M.C., Owen, M.J., & Thapar, A. (2010). Genotype link with extreme antisocial behavior: The contribution of cognitive pathways. Archives of General Psychiatry, 67, 13171323.
  • Laucht, M., Blomeyer, D., Buchmann, A.F., Treutlein, J., Schmidt, M.H., Esser, G., … & Banaschewski, T. (2012). Catechol-O-methyltransferase Val158Met genotype, parenting practices and adolescent alcohol use: Testing the differential susceptibility hypothesis. Journal of Child Psychology and Psychiatry. Advance online publication. doi:10.1111/j.1469-7610.2011.02408.x
  • Mattay, V.S., Goldberg, T.E., Fera, F., Hariri, A.R., Tessitore, A., Egan, M.F., … & Weinberger, D.R. (2003). Catechol O-methyltransferase val158-met genotype and individual variation in the brain response to amphetamine. Proceedings of the National Academy of Sciences of the United States of America, 100, 61866191.
  • McGuffin, P., Alsabban, S., & Uher, R. (2011). The truth about genetic variation in the serotonin transporter gene and response to stress and medication. British Journal of Psychiatry, 198, 424427.
  • Meyer-Lindenberg, A., Kohn, P.D., Kolachana, B., Kippenhan, S., McInerney-Leo, A., … & Berman, K.F. (2005). Midbrain dopamine and prefrontal function in humans: Interaction and modulation by COMT genotype. Nature Neuroscience, 8, 594596.
  • Risch, N., Herrell, R., Lehner, T., Liang, K.Y., Eaves, L., Hoh, J., … & Merikangas, K.R. (2009). Interaction between the serotonin transporter gene (5-HTTLPR), stressful life events, and risk of depression: A meta-analysis. Journal of American Medical Association, 301, 24622471.
  • Savitz, J., Solms, M., & Ramesar, R. (2006). The molecular genetics of cognition: Dopamine, COMT and BDNF. Genes, Brain, and Behavior, 5, 311328.