Eating disorders comprise phenotypically diverse manifestations of problems with the regulation of food intake and body weight.
AN and BN are associated with a strong desire to gain control over pressures to fulfil socio-biological role models.
From an evolutionary point of view, the prevalence of eating disorders poses a paradox in two respects.
First, there is an inverse relationship of food availability and prevalence of eating disorders across societies.
Thus, ironically, food abundance in Western societies is associated with high prevalence rates of eating disorders, especially when subsyndromal manifestations are being considered, whereas eating disorders are almost absent in less developed countries.
Second, irregularities of the menstrual cycle or amenorrhoea, which frequently occur in eating disorders, counteract any reproductive effort.
How can this be consistent with an evolutionary explanation of behaviour?
A couple of hypotheses exist that take into account female preponderance in eating disorders.
For example, eating disorders manifest around the time at which female fertility peaks.
Moreover, some findings point to the fact that peculiarities of the family environment may increase the risk for young females to engage in food restriction and control of body weight.
These hypotheses based on evolutionary theory share the assumption that the suppression of reproduction (i.e.amenorrhoea) is the main 'function' of eating disorders, rather than being a by-product of the symptomatology.
It needs to be emphasized, however, that eating disorders do not convey any adaptive value in the strict sense.
Like other psychopathological signs, symptoms and syndromes they represent extremes of variation that reduce an individual's fitness.
At the proximate level, the avoidance of food, and efforts to reduce body weight are strong signals sent by the affected individual, indicating, at the very least, conflict over resource allocation, where the addressee of these behaviours is most likely the nuclear family.
Studies suggest that mothers of adolescent girls with AN tend to be
overprotective, dominating and monopolizing.
Patients with AN, on the other hand, are often overcompliant, and worry excessively about the wellbeing of the family.
Consistent with this is the observation that anorexic females often dismissively negate family problems.
AN is also associated with low self-esteem, introversion, pronounced harm avoidance and reduced novelty seeking, and the co-occurrence with anxiety disorders and depression suggests that patients with eating disorders unconsciously assume the behavioural strategy of a subordinate individual to which genes may contribute heritable susceptibility.
Reproductive inhibition in subordinates is common in non-human primates, and this is often the result of active manipulation by higher-ranking females.
Elevated stress levels in subordinates, for example, are known to contribute to the suppression of ovulation, and such a scenario is similarly conceivable in humans.
In addition, women may also suppress the reproductive potential of other women by cultural means, for instance, by unconsciously promoting that the 'ideal ' female figure is one of below average weight.
This is confounded by the tendency to perceive thinness as indicator of youth.
As youth signals attractiveness, there is intrasexual competition in women for looking thin, which may, to some extent, enhance the preference for thinness not only by men but also women.
In light of the close genetic relatedness of daughters with mothers and siblings, it might in some cases even be possible that suppressed reproduction is not due to the interference of an unrelated higher-ranking female, but that the mother of the affected adolescent girl is the source of manipulation.
From a genetic point of view, the logic behind this hypothesis could be that the mother suppresses reproduction in her own daughter such that the daughter may serve as a 'helper at the nest'.
In other words, altruistic behaviour of adolescent girls may be imposed upon the daughter by her mother, particularly in the presence of (male) siblings.
Theoretically, the reproductive success of a male descendant exceeds that of a female, at least under favourable environmental conditions (compare Chapter1), which could explain why AN is especially prevalent in upper middle and upper class girls in socially stratified societies.
Since the affected individual would indirectly benefit from supporting her mother at the expense of own reproduction, this sort of altruism in female individuals with AN would be typical of kin-selected behaviour.
There may however, be, another reason for mothers (and fathers)to suppress fertility in adolescent daughters.
In western societies, the abundance of nutrient-rich food has accelerated sexual maturation over the last decades.
Suppression of premature sexual activity in adolescent girls may well be in the interest of parents to increase their daughters' reproductive success in the long run.
This possibility may be accentuated in families in which the father is absent.
Unlike other primates, father absence in humans is associated with earlier maturation and sexual activity.
In other words, the presence of fathers (and high paternal investment) postpones sexual maturation in girls, perhaps as a reflection of a K-selected reproductive strategy (compare Chapter 3).Accordingly, in families with fathers being absent, manipulation of girls to restrict food intake may indirectly serve the function to turn an early maturer into a late maturer.
Similar interest in delaying reproduction could also be on the side of the adolescent girl.
The onset of eating disorders is often preceded by (unwanted)first sexual experiences.
Thus AN may be seen as a counter-strategy to avoid premature pregnancy.
Moreover, the increase of eating disorders in western societies and threshold countries may reflect changes in social structure associated with increased sexual competition between females.
Therefore, under adverse environmental conditions, which, above all, may include the absence of emotional availability of significant others and lack of social support, long-lived iteroparous species like humans may be better off by postponing reproduction until the conditions are more favourable.
At the proximate level, this is maintained by reducing the amount of body fat.
In contrast to male fertility, ovulation ceases if the amount of body fat undercuts 15 per cent of body weight.
This makes sense from an evolutionary perspective, because in times of food scarcity the likelihood of successfully raising offspring declines sharply, such that the hypothalamus stops producing gonadotropin-releasing hormone.
Moreover, the fall of oestrogen and extreme loss of body weight leads to a reduction of secondary sexual characteristics in women, such that in extreme cases women become sexually less attractive for males.Consistent with this, sexual interest often declines in young women with AN, which further reduces the risk of unwanted pregnancy.
However, such a strategy pays off reproductively only if the probability of improved conditions in the future is predictably high.
This is the case in western societies in which, on average, adult mortality is low (accordingly, within western societies the prevalence of eating disorders is predicted to be low in neighbourhoods with high criminality and homicide rates).
This notion is again supported by the evidence that eating disorders occur more frequently in socially well-situated families.
By contrast, if the likelihood of future environmental condition
s to improve is low, it does not make sense to postpone reproduction, which may explain low prevalence of AN and BN in developing countries and in socially disadvantaged people.
s to improve is low, it does not make sense to postpone reproduction, which may explain low prevalence of AN and BN in developing countries and in socially disadvantaged people.
Furthermore, the hypothesis of reproductive delay is also plausible to explain why eating disorders manifest in adolescence or early adulthood; for older women, postponing reproduction would be too risky a strategy, because it could preclude reproduction at all.
This hypothesis also explains why for males such a Strategy would not pay off.
According to parental investment theory, male reproduction is theoretically constrained by the number of sex partners, whereas females invest more in individual offspring; hence their reproductive success is limited by a much lower theoretical maximum number of offspring that corresponds with the number of fertile cycles, social support and inter-birth intervals.
Accordingly, for young women, optimal environmental conditions, including the presence of a male partner who is willing to invest in offspring, are much more important than for young men.
Conversely, in light of the lower paternal investment and the more intense male-male competition, it does not make sense for males to postpone reproduction.
As already pointed out, these hypothetical models of the evolutionary background of eating disorders hinge upon the proposition that amenorrhoea is the (unconscious) goal or function of eating disorders, which is achieved by restricted food intake, weight control and excessive exercising or other compulsive behaviours.
However, a substantial number of patients with eating disorders, particularly BN, do not stop ovulating and show promiscuous behaviour.
Moreover, women with a history of eating disorder have a greatly elevated risk of hyperemesis during pregnancy, more often deliver children with lower birth weight and smaller head circumference, and have significantly more miscarriages and caesarean deliveries compared to women without a history of eating disorder.
The elevated number of birth complications is not compensated for by a greater number of children or number of pregnancies.
There is, therefore, little evidence that postponing reproduction pays off for women with a history of eating disorder.
Moreover, research into perinatal risk factors for the development of AN or BN has revealed that maternal anaemia, diabetes mellitus, preeclampsia, placental infarction, neonatal cardiac problems, and hyporeactivity independently predict the development of AN later in life, with similar, though less pronounced risks for BN.
Also, inter generational studies of attachment in patients with AN and their mothers have shown that insecure attachment, dismissive attachment styles, and poor reflexive functioning ('mentalizing') may be 'transferred' from mother to daughter.
In addition, the risk of childhood trauma is increased in eating disorder.
Drawing on attachment theory and life history theory these findings suggest, above all, poor parental investment that is conserved over successive generations.
The consequences of AN (and less so of BN), namely a visible waste of resources and profoundly health-threatening behaviour, can therefore be interpreted as a correlate of parent-offspring conflict.
In parent-offspring conflict, the child has only a limited set of behavioural strategies at hand to increase parental investment.
One such strategy is the expression of temper tantrums, which typically occur during the weaning period.
The 'logic' behind such behaviour is that it not only threatens the health of the offspring, e.g.by attracting predators, but also that of the parents, because the latter would lose all resources they have already invested.
Confronting a parent or both parents with the risk of losing a child through deliberate starvation is therefore one of the strongest conceivable signals of the child to mobilize parental investment.
The timing or onset of the behaviour peaks around or precedes the time of maximum fecundity of adolescent girls and young women, and thus constitutes the maximum threat to the inclusive fitness of both parents.
If the parent is unable or unwilling to provide additional investment, however, the situation is self-perpetuating, and a vicious circle may result that can only be resolved through therapeutic interventions fostering greater independence of parental investment in the child.
Eating disorders are so heterogeneous in aetiology that none of the hypothetical models can embrace all clinical and neurobiological facets.
In some patients with eating disorders, it is plausible to assume that the affected individual carries the burden of maladaptive consequence of suppressed fertility, which benefits others, perhaps including the patient's own mother.
Other patients with AN may be forced to postpone reproduction if environmental conditions are unfavourable, with the expectation that condition will predictably improve in the future.
Finally, in some patients with eating disorders, the symptomatology may be an exaggeration of behavioural consequences related to parent-offspring conflict to increase parental investment.
None of these behavioural 'strategies' involve conscious awareness of the underlying conflict.
Nor do they convey a reproductive advantage; on the contrary, eating disorders can be seen as emergency strategies that emerge in young women who struggle with the achievement of biosocial goals, among which the search for social security is imperative.