The Genetics of Thin-Ideal Internalization

The Tripartite Model of body image dissatisfaction postulates that three factors (peers, parents, and media) affect body image dissatisfaction and disordered eating through thin-ideal internalization and appearance comparison.

Thin-ideal internalization is the extent to which one accepts or “buys into” socioculturally defined beauty standards of thinness. The idea is that the more someone internalizes these standards, the more likely they are to engage in behaviours to achieve their “ideal”, and the more likely they are to develop an eating disorder.

A growing number of of studies have been done evaluating the validity of this model. Although I’m not well-read on the subject, it does seem like there is a growing number of studies showing an association between thin-ideal internalization and disordered eating practices.

But is the picture complete? Are peers, parents, and media the only or even the main factors that influence the extent of thin-ideal internalization?

One factor that’s curiously missing from the research is genetics. Can genetics play a role in explaining why some individuals are more prone to internalizing the thin-ideal than others?

This possibility is supported by the fact that despite almost ubiquitous exposure to thin-ideal images in Western countries, only some women ultimately internalize this ideal and go on to develop disordered eating behaviors. Individual difference factors, such as genetic risk, may predict which of these women are more likely to internalize the thin ideal and which are not. The presence of significant genetic influences on eating disorders, and disordered eating symptoms (e.g., dietary restraint, drive for thinness, weight preoccupation, body dissatisfaction, and shape/weight concerns) lend additional credence to the possibility that genetic factors may contribute to thin-ideal internalization.

To investigate whether genetic factors play a role in thin-ideal internalization, Jessica Suisman and colleagues compared the extent of thin-ideal internalization (assessed with questionnaires) in monozygotic and dizygotic twins. (Here’s a good explanation of twins studies.)


Suisman et al. found that genetic and unique environmental factors (factors unique to one siblings raised in the same family) but not shared environmental factors (factors that are common to both siblings raised in the same family). Indeed, their analyses showed that heritability for thin-ideal internalization is around 43%, meaning that genetic factors account for 43% of the variance we see in thin-ideal internalization.

These findings may help explain why only some individuals internalize the thin-ideal while others do not.

What could these genetic factors be?

Well, we certainly don’t have genes for “thin-ideal internalization.” So, what kind of genetic factors might account of this heritability?

One set of ‘‘genetic’’ predispositions that might explain the heritability of thin-ideal internalization is personality traits. Although research has not extensively investigated personality traits that are associated with thin-ideal internalization, initial results suggest some associations with variables
such as perfectionism, and researchers have hypothesized that personality may influence individual differences in internalization of the thin ideal.

Interestingly, such personality characteristics are known to be heritable and may contribute to the additive genetic variance on thin-ideal internalization. In other words, a large proportion of the heritability of thin-ideal internalization may overlap with genetic factors that influence personality.

I think in future studies, it would be interesting to untangle what mediates this heritability. Is it perfectionism? Are the people who are more likely to internalize the thin-ideal also more likely to internalize other social norms and ideals or is it specific to eating disorders? 

Moreover, the participants in this study were drawn from the Michigan State University Twin Registry and were not screened for Axis I or Axis II disorders (as far as I understand). So, we can assume the sample is pretty typical of a normal population, with the prevalence of mental health disorders that we see in the larger population. This made me wonder, if we were to take a sample of MZ and DZ twins that did not have any psychopathology, would the heritability go down?

My reasoning is as follows: if the heritability for thin-ideal internalization can be explained by other factors such as personality or temperament (which also seem to have similar levels of heritability, as I understand), and then we were to compare heritability among a more homogenous group of individuals (since we’d exclude those with mental disorders), environment would play a larger role in explaining the differences between thin-ideal internalization.

All of which brings me back to the question of what this heritability really means.


The nonsignificance of the shared environment in explaining the variance in thin-ideal internalization may seem surprising, particularly given the ‘‘shared’’ nature of exposure to media among women (i.e., almost all women are exposed to high levels of thin-ideal media) and the expectation that siblings growing up in the same home would be exposed to similar levels of such media, and thus, would likely internalize the thin ideal to a similar degree.

The present findings do not support these predictions, as siblings being raised in the same home did differ largely in the extent to which they internalize the thin ideal.


There’s one important point about gene and environment interactions. Gene x environment interactions inflate genetic effects in these types of twin model studies. So it could be that some of the value we are seeing for genetic effects is actually due to gene x environment interactions. It seems that we need bigger studies to resolve these questions.


In this study, the best model was one that reduced the effects of the shared environment (like exposure to media) to 0, suggesting that it plays no role in explaining the variability in thin-ideal internalization. It is important to consider, as the authors point out, that what we typically think of as “shared” environment might actually be operating at a level of the “nonshared” environment. One example would be where a parent reinforces dieting only in the slightly heavier child.


It will be important to see how the heritability measures differ in populations where thin-ideal is not emphasized. (Remember, heritability is not static, it depends on the population studied.)

Genetic effects would be expected to decrease, and environmental estimates would be expected to increase, in cultures in which there is more variance in exposure to the thin ideal across individuals. This is because, in cultures where thin-ideal exposure is rare, even individuals who have genetic predispositions for thin-ideal internalization may never internalize the thin ideal because they are never exposed to it. Thus, genetic predispositions cannot operate, driving down estimates of genetic effects.

It will also be interesting to investigate heritability differences (if there are any) in environments with high media exposure and those with low media exposure. Finally, it will be important to understand what is the “genetic architecture” that explains this heritability. In other words, do the genes contribute to personality and/or temperamental traits that make individuals more likely to internalize the thin-ideal or perhaps more likely to internalize societal ideals in general?

What are your thoughts?


Suisman, J., O’Connor, S., Sperry, S., Thompson, J., Keel, P., Burt, S., Neale, M., Boker, S., Sisk, C., & Klump, K. (2012). Genetic and environmental influences on thin-ideal internalization International Journal of Eating Disorders, 45 (8), 942-948 DOI: 10.1002/eat.22056

Liked it? Take a second to support Tetyana on Patreon!


Tetyana is the creator and manager of the blog. She has an Honours BSc in Neuroscience and an MSc in Medical Science. She can be reached at tetyana[at]scienceofeds[dot]org.


  1. I haven’t read the original paper, but what I would ask (and you could probably guess I would ask this…) is how important ‘thin-ideal-internalisation’ actually is in the aetiology of eating disorders – and whether this social influence plays a greater role in some EDs than other EDs.

    My understanding, from Becker and Stice’s research, is that ‘thin-ideal-internalisation’ is associated with disordered eating – and is more strongly linked to BN than to AN. Many people exhibit disordered eating, but they don’t necessarily have an ED.

    If I recall correctly, Carolyn Becker has stated that the cognitive dissonance model she has developed with Eric Stice, which aims to reduce internalisation of a thin ideal, is not effective in terms of prevention of all EDs, including most cases of AN. Of course, it is great if this model is effective in preventing some EDs, but the inference is that factors other than thin ideal internalisation play a role in the development of EDs in some people.

    I know (of) a number of people with AN who have said that they didn’t experience body image distortion or body dissatisfaction until they had lost weight and developed ED behaviours. This may suggest that weight loss itself, or other factors aside from cultural ones lead to alterations in the experience of the body (shape and size). In other words, the body image disturbances and dissatisfaction observed in some people with EDs may be a phenomenon of neurological origin rather than a societal phenomenon?

    • “My understanding, from Becker and Stice’s research, is that ‘thin-ideal-internalisation’ is associated with disordered eating – and is more strongly linked to BN than to AN. Many people exhibit disordered eating, but they don’t necessarily have an ED.”

      Yes, that’s my understanding too, but I’m not well-read on the literature.

      “I know (of) a number of people with AN who have said that they didn’t experience body image distortion or body dissatisfaction until they had lost weight and developed ED behaviours. This may suggest that weight loss itself, or other factors aside from cultural ones lead to alterations in the experience of the body (shape and size). In other words, the body image disturbances and dissatisfaction observed in some people with EDs may be a phenomenon of neurological origin rather than a societal phenomenon?”

      Yup. I’m one of those people. I don’t recall ANY issues with my body or really thoughts about my body, until I had lost some weight. I legitimately wanted to eat healthier. Others on the blog said similar things, as well as pointing out that the focus on body image in Tx might have actually made the situation worse for them. However, I wonder what role starting weight has in this. Maybe thin-internalization plays a bigger role for those who are further above the “ideal” weight than for those who aren’t? I’d hypothesize that for those who are overweight or obese, thin-ideal internalization might push them more to disordered eating and some of those to a full blown ED.

      I have more thoughts on the stronger links to BN than AN, but I’ll have to come back to that later tonight.

  2. You both raise some interesting points. If you haven’t had a chance to look at it, you may be interested in Jerome Kagan’s book “The Long Shadow of Temperament”. It addresses some of the questions that you had asked Tetyana. There’s also other research that addresses some of those questions which is already been published elsewhere.
    From a clinical perspective, differentiating between the maintenance role of the thin idealization versus the role in etiology is also important. Some good points here think about though.

  3. I would suspect that weight and perception of body size before the onset of an ED does influence thin ideal internalisation. I also suspect that age of onset of the ED (e.g. childhood, pre-puberty vs. adolescent post menarche) also influences the extent to which thin ideal internalisation plays any role in the development of an ED.

    I was always very slim as a child and had eating difficulties (linked to emetophobia, which was in turn a trigger for OCD). I never thought of myself as ‘fat’ and never feared ‘fatness’. My ED started at age 11. So it is unsurprising that my ED has been non-fat-phobic.

    At one time I was weight phobic but not fat phobic. I feared seeing the number go up on the scales because I associated this change with my life spiralling out of control. I think it would be helpful if clinicians could separate weight phobia from fat phobia where necessary, because they’re not always linked. When I was very underweight I didn’t perceive myself to be as thin, visually, as others did, but even so, my anorexic behaviours were fueled by compulsion and ritual rather than fat phobia.

    Anyway, this is an interesting post, as usual.

  4. Sorry if I am rambling and digressing… but I would really like to see clinicians treat all people with EDs as individuals and not to assume common aetiologies – even to explain current subcategories of ED.

    For example, two people with low weight, disordered eating and psychological distress may both meet the diagnostic criteria for AN, yet while one may be very fat phobic, with body dissatisfaction and disturbed body image, which fuel disordered eating, the other may be non fat phobic and have disordered eating that is fueled by compulsive, ritualistic behaviours (linked to OCD, ASD traits etc.). In my (too) long experience of treatment for AN, I have observed a tendency for professionals to focus on weight/thinness in AN, with the axiomatic assumption that thinness is THE goal. Sometimes it is, but sometimes it’s not.

    A patient feels more validated if they feel they are being listened to and shown empathy (if not sympathy) – and can feel completely invalidated if the clinician makes erroneous assumptions or, worse still, doesn’t believe them.

    I sound like I am ranting. Sorry 🙁

    • Not rambling! I will respond later today. A bit overwhelmed with things to do, but just wanted to say I’m not ignoring your comments, I just haven’t had time to get to responding properly :-).

Leave a Reply

Your email address will not be published.

11 + eight =