European positive psychology conference in Amsterdam: what proportion of well-being is genetically determined? (3rd post)
Last updated on 20th September 2014
I have already written a couple of blog posts about this 7th European Conference on Positive Psychology - the first on pre-conference workshops about supervision & about relationships and the second on love, national happiness league tables, and life satisfaction assessment. After the coffee break I went on to, what for me turned out to be, one of the most interesting sets of presentations at this conference - an invited symposium on "Biological aspects of wellbeing and resilience". The key take-home message I left this symposium with is to be much more cautious in how I explain that, when comparing populations, about 36% (a little over a third) of well-being is due to genetic effects, while about 64% (nearly two thirds) is due to individual environmental effects. For a specific individual however, we're much less clear. Even if we knew the state of all relevant genes that were involved in affecting the individual's well-being, these effects could still be mostly cancelled out by favourable environmental circumstances (such as loving, responsive parenting). So many of us should back track on the way we, as professionals interested in well-being, have too easily made comments to people about what percentage of their well-being is genetically and what percentage is environmentally determined. As these heavy-hitting geneticists highlighted, the research studies coming up with these kinds of percentages are NOT studies on wellbeing in individuals, they are studies comparing different populations. So when comparing large groups of people we have some justification for saying that about 40% of wellbeing is heritable (36 to 41% in Meike Bartels recent meta-analysis described in this symposium), but this doesn't apply with any great accuracy to a specific individual. Secondly, even when making this kind of statement about large populations, the percentage of wellbeing attributable to genetic factors tends to decrease in more difficult environmental conditions. There is increasing evidence for this kind of variation in gene-environment ratio depending on environmental characteristics - see Ragnhild Nes below and also recent research by Bartels & colleagues e.g. this year's paper "Child care, socio-economic status and problem behavior: A study of gene-environment interaction in young Dutch twins" with its conclusion that "heritability is lower in circumstances associated with more problem behaviors" and "the decrease in heritability was explained by a larger influence of the environment, rather than by a decrease in genetic variance."
The "Biological aspects of well-being and resilience" symposium had presentations from five impressive scientists. Bart Rutten spoke on "Resilience: linking psychological and neuro-biological perspectives". He mentioned the role of epigenetics, for example effects on methylation produced by early maternal care. Interestingly, he also referred to using experience sampling via mobile phones to get data that is closer to "real time" than our typical daily, weekly or even less frequent questionnaire methods. Googling "experience sampling" and "android" or "iphone" highlights the kinds of activity in this area ... one to watch for the future. Ragnhild Nes spoke about "Causes of individual differences in well-being: an overview of twin and family studies". My scribbled notes include the following comments: (in large populations, for example of twins) about 30 to 50% of well-being is genetically determined, very little is due to 'shared environment', whereas about 30 to 70% is due to 'unique environment'. A 2014 meta-analysis of the current research literature gives a figure of about 40% of well-being as heritable, but this percentage decreases in harsher environments. About 50% of the variance in well-being is stable & this is mostly because of genetic contributions. About 50% of the variance is variable and this is mostly environmental. Genetic influences on well-being act primarily via personality, for example neuroticism and extroversion. All personality disorders are associated with decreases in well-being, especially avoidant, borderline & paranoid disorders.
The third speaker was Meike Bartels, one of those enviable people who look wonderful and whose brains seem razor sharp. She talked about "Molecular genetics and well-being". She said that 36 to 41% of well-being is heritable. In fact the most up-to-date figure from meta-analysis seems to be 36% which pushes the genetic component of well-being down towards only about a third ... quite a decrease from the often quoted 50%. Like pretty much all the speakers in this symposium though, she underlined that this does NOT explain heritability in a specific individual ... only in comparative studies looking at differences between individuals. She said that we'll only be able to assess the genetic contribution to well-being for an individual when we know all the relevant genes and, even then, high genetic risk in an specific person might make little difference to their level of well-being because good environment could cancel out the genetic effect. See the last speaker, Claire Haworth's example of phenylketonuria (below) for more on this point.
Barbara Fredrickson now spoke on "Gene expression and well-being". In my experience, Barbara is usually the star of the show when she gives a talk. She spoke well here, but she isn't a geneticist so it was interesting to see her in the different, heavily statistical environment of this symposium. Again looking back at scribbled notes, I have written that genomic correlates of loneliness and adversity constitute a "forward-looking immune system" ... isolation & adversity "expect" tissue damage & associated bacterial infection. She contrasted the "feel-good" quality of hedonic well-being and the "do-good" quality of eudaimonic well-being. She commented how the two are often reciprocally related with "feel-good" encouraging "do-good" and vice-versa. She talked about the potential sequence from positive thoughts to positive feelings to positive meaning to healthy genomes. She mentioned too that eudaimonic well-being seems more strongly related to healthy gene expression than hedonic well-being is. This reminds me of a series of posts I wrote a while ago beginning with "Purpose in life: reduces dementia risk, increases life expectancy, treats depression and builds wellbeing".
The fifth & final speaker at the symposium was Claire Haworth from the University of Warwick, talking about "Gene by well-being intervention effects". Again from my notes, I see that she discussed a study done with Sonja Lyubomirsky teaching 47 sets of twins (some monozygotic, some dizygotic, all the same gender) to perform three acts of kindness per day and write a ten minute letter of gratitude (not posted). These experimental subjects showed increases in well-being. She commented that baseline levels of kindness had an important impact on response to this intervention with subjects who were already strong in this area benefitting most (another example of the increasingly supported advice that interventions tend to do best if building on strengths rather than trying to correct weaknesses). She also emphasised that genetic studies are not particularly relevant to well-being in individuals, highlighting that genetic risk is probabilistic not deterministic. I found it particularly helpful when she illustrated this point with the example of phenylketonuria (PKU). As Wikipedia comments "This genetic disorder is characterised by the inability to metabolise the amino acid phenylalanine." It was only identified in 1934. Untreated it leads to "intellectual disability, seizures, and other serious medical problems. The mainstream treatment for classic PKU patients is a strict PHE-restricted diet supplemented by a medical formula containing amino acids and other nutrients ... Patients who are diagnosed early and maintain a strict diet can have a normal life span and normal mental development." What a nice example ... a genetic life-destroying sentence beautifully reversed by an appropriate, skilful environmental response. Absolutely, for a given individual, genetic risk is probabilistic not deterministic. Great stuff ... and a great symposium!
For the next post in this sequence about the European Positive Psychology Conference in Amsterdam, click on "Self-determination theory, positive aging, and the economic crisis."