Gene-ecology interactions and psychological disorders

At Marked For Life - the recent Progress Educational Trust debate - speaker Fenno Outen suggested the term 'culture' described more precisely than 'environment' some causes of mental illness. There is reason to have sympathy with Outen. Humans are exposed to cultural influences from birth to death and they clearly have a causal role to play. Emphasising culture would recognise some of the core demands facing humans today. But culture is a broad term capturing diverse activities and beliefs, and involving numerous psychological mechanisms. A better and more precise term would be 'ecology' - a notion we will defend in this article.

From the perspective of a behavioural biologist, 'environment' refers to everything outside the system undergoing study - it can be everything outside a whole organism or a gene. 'Ecology' refers to the portion of the environment used by the system or which the system is adapted to use. In this sense, ecological accounts rely on evolutionary theory. Evolutionary theory is our current best theory of gene-environment interaction - it enables us to focus on relevant aspects of the environment when developing explanations for phenotypic traits.

Behavioural geneticists studying mental health are interested in two things: patterns of inheritance in the population and genes that potentially contribute to a disorder or illness. The patterns of inheritance seen in many mental illnesses suggest polygenic traits and several models exist to account for such phenotypes persisting in nature. For example, it is possible that schizophrenia remains the population because - like sickle-cell disease - some genes that raise disease risk also convey advantages. What is interesting about sickle cell is that the gene responsible occurs at greater frequencies in the malarial regions of the world. Those regions represent the relevant ecology for the adaptation.

Sickle cell disease is clearly detrimental to fitness and is an unfortunate consequence of selection for malarial resistance. However, as Nesse (1) has pointed out, not all apparently harmful traits are harmful in all places and circumstances. Where we see alleles with widespread dispersion throughout the human population, they are likely to be benign as they have resisted being selected against over a long period. So when they are associated with disorders in particular places, it is likely  there is a mismatch with the current environment. Or, more precisely, they are no longer operating in the appropriate ecology (2). This could be due to migration or a change in the local ecology.

Persistent negative behaviours, e.g. conduct disorder, may not be detrimental to long-term fitness. Evolutionary models are only concerned with the average fitness of an organism over its lifetime. Behavioural biologists assume organisms act to maximize their fitness, but this can mean taking on some immediate-term costs. This notion is captured by Life History Theory.

Life History Theory assumes that organisms cannot simultaneously devote energy and time to growth, development and survival, and reproductive effort. Throughout their lives, they must decide which to expend energy on. Both are important to lifetime fitness, but at certain points individuals must forgo one for the other. This trade-off will be shaped by local ecological factors; such as sex, age, sex ratio, local mortality rates and availability of resources. Optimal inclusive fitness will depend upon the success of the organism's strategy to manage these trade-offs.

Adopting different life-history strategies is partly due to reactive behavioural and learning mechanisms, or epigenetic processes. An example is maternal behaviour studies on Long Evans Rats (3). These show stressed mothers groom their pups for less time than unstressed mothers and this maternal behaviour is epigenetically inherited. The pups of low grooming mothers also enter sexual maturity sooner and begin sexual behaviours earlier. In an adverse ecology, the stress response leads to a faster mobilisation of reproductive resources to maximise fitness. This effect can be reversed by changing the ecology. Such work has implications for those working on teenage pregnancy (4).

So what about so-called 'conduct disorder'? This, according to DSM-IV (5), is 'a repetitive and persistent pattern of behaviour in which the basic rights of others, or major age-appropriate societal norms or rules, are violated'. It is over-represented in young males and in socioeconomically stressed populations (6). There is childhood onset persistent conduct disorder, adolescent conduct disorder and childhood limited conduct disorder. Some are aggressive and others simply break rules and conventions.

There are two possible hypotheses. First, some of the population exhibiting conduct disorder might be sociopaths. This is probably genetically heritable and held in the population by frequency-dependent selection (7). Sociopathy is a free-riding strategy in which social contracts are violated. A sociopath's typical life course involves moving from place to place parasitising populations until they are discovered and the strategy breaks down. Only a small percentage of the population can thrive as free-riders - if too many emerged the strategy would fail as insufficient co-operators would exist. So the genes that produce this strategy occur at a low frequency within the population.

Second, the aggressive behaviour we see in socioeconomically-stressed young men is precisely what we would expect from a primate facing adversity. When there is little to lose, competition becomes fierce for scarce resources, such as mating opportunities. Wilson and Daly (8), writing about male homicide deaths in Chicago note the close relationship between aggressive male populations and teenage pregnancy - all are on a fast life history and there are fitness pay-offs.

Those with aggressive conduct disorder are still a minority within any population so other factors must be at work. Evidence exists that maternal prenatal anxiety predicts increased cortisol activity and associated stress responses in children (9). So, like the Long Evans Rats, this suggests a mother's response to anxiety-invoking environments may alter the psychological behaviour of her offspring so they are prepared for the same adverse ecology.

Outen argued for culture, and we are arguing for ecology, which incorporates social effects. The models emerging from behavioural biology are increasingly sophisticated. They do not undermine the quest for candidate genes, but augment our understanding of why those genes might persist and be differently expressed in populations.