Can attachment theory help explain the relationship some people have with their “anorexia voice”?

A new paper in Psychology and Psychotherapy: Theory, Research and Practice argues that the relationship a person has with their eating disorder is shaped by that person's understanding of what meaningful relationships should look like – and, in turn, this can have important consequences for the severity of their disorder.

In particular, Emma Forsén Mantilla and her team from the Karolinska Institute in Sweden wanted to better understand eating disorders through "attachment theory". This is the idea that relationships with primary caregivers become scripts that we lean on to tell us how relationships "work". 

A parent perceived as being protecting will lead a child to feel trust, according to this theory, and to expect a protection-trust dynamic in future relationships. A more troubled caregiver-child relationship, in contrast, leads to a different form of attachment, and downstream consequences – including believing that it's normal that people who care about you attack you. In addition, attachment tells us how it's appropriate to treat ourselves – such as by attacking or judging ourselves. Once these dynamics have formed, we gravitate towards them especially when we're distressed (and being prevented from following these learned dynamics can trigger yet more upset).

Eating disorders are more common in people with less secure attachment styles, and on the surface there are certain parallels in how people experience eating disorders and the attachment dynamics described above – for instance, it is when distressed that people with eating disorders are most likely to seek consolation in comforting, but ultimately counterproductive, eating behaviours, and people with eating disorders can become even more agitated when they are prevented from performing these behaviours. 

However, there's a deeper connection between attachment theory and the experience of eating disorders in what some researchers and service users have previously labelled the "anorexic voice" – when symptoms are maintained through harsh and forceful judgments about oneself and instructions to the self that are attributed to the anorexia condition. Despite its negative, attacking nature, the voice is often still considered an ally rather than an enemy, again resembling certain insecurely attached relationships. So it seems a plausible to see eating disorders in these terms, especially as attachment patterns can be established in how people relate to non-humans – pets and even God.

Mantilla's team tried to establish a coherent pathway between eating disorder clients' attachment styles and their symptoms. They ran a correlational study involving 148 women with eating disorders aged 16-25 recruited from outpatient units in Sweden. The researchers measured eating disorder behaviours with questions like "over the past 28 days, on how many days have you made yourself sick (vomit) as a means of controlling your shape or weight?"and found, consistent with previous findings, that these behaviours were more pronounced in people with a less secure attachment style (assessed with questions like "I feel confident about relating to others.").  

Mantilla also asked participants to complete a survey normally used to look at specific attachment dynamics with a significant other, but in this case rephrased to focus on the eating disorder – e.g. "It punishes and tortures me, takes revenge" – as well as whether they turned these patterns inward. 

As the researchers predicted, less securely attached individuals in general tended to see their eating disorders as more controlling, and themselves as less autonomous. These individuals were also more likely to treat themselves in a blaming way, and this self-blaming and the sense of a controlling quality to their eating disorder entirely explained the association between overall attachment and their eating disorder symptoms and behaviours. 

In other words, a plausible story can be made from the data that insecure attachment produces a tendency to internalise blame and an expectation of a controlling relationship that then takes shape in an eating disorder, producing eating disordered symptoms accordingly. In a correlational design, of course, this is not proof but has a sense of plausibility to it.

In helping people to work past eating disorders, psychologists using narrative therapy already characterise the disorder as an external entity in a relationship. This research may be useful in honing in on the aspects of this relationship most strongly associated with actual eating disorder behaviours – the controlling nature of the disorder and how it may prop up and reinforce self-blame. It also suggests the likely obstacles faced by these clients in forming other relationships, which may prevent them developing emotional support and regulation that in itself can prevent relapse. And the results shed light on why people can be appear to be so wholly in the grasp of an eating disorder – because losing it can almost feel like losing someone who "gets" you. Anorexia and similar conditions can be like a tight embrace that goes both ways.

Further reading

—The unsafe haven: Eating disorders as attachment relationships

About the author

Alex Fradera (@alexfradera) is Staff Writer at BPS Research Digest