A fundamental attribution error in CBT-e for eating disorders

There is something fundamentally wrong with Cognitive Behavioral Therapy (CBT-E) for anorexia nervosa and other eating disorders. I want to be clear that I think that some form of CBT is useful in treating eating disorders. After all, anorexia nervosa (AN) and bulimia nervosa (BN) are characterized by dysfunctional cognitions and behaviors that maintain the disorders and a CBT approach can profitably focus on changing these cognitions and behaviors.

My concern is with that Fairburn and his colleagues’ CBT-(enhanced) believe the core psychopathology of eating disorders is the “overevaluation of shape and weight”(1). (I think we agree on the behavioral part, people must normalize eating and stop purging.)

This conceptualization ignores history (2) and biology (3). Moreover, it can interfere with recovery (4) and contribute to sufferers’ and observers’ blaming victims for their plight.

History: Before the 1960s people with anorexia nervosa did not give fear of getting fat as their reason for eating little. They said they were full, their stomach was upset, they were not hungry. St Catherine of Siena thought she had difficulty eating earthly food because God had prepared a heavenly table where she would be able to eat. She died of malnutrition at 33 but not before shaking up the Catholic Church. Eleanor Roosevelt, Simone Weil and Virginia Woolf struggled with AN all their lives. They were not trying to be thin.

“Fear of being fat” is the culture-bound explanation that people have given since the 1960s for the chaotic and confusing signals welling up from the ancient, nonverbal hypothalamus. Signals that, in the case of AN, make it hard to eat and drive them to move, and in the case of BN drive them to bingeing and purging. People automatically search for an explanation for any mysterious feelings and readily agree to the currently popular psychological explanations.(5)

Biology: eating disorders are caused by powerful survival mechanisms that will not change as long as the body fears famine. Trad CBT for eating disorders does not reflect what biological researchers have learned about what causes eating disorder symptoms and how to heal them. It treated the brain as a black box, but it shouldn’t.

CBT was developed originally to treat depression. With depression, viewing the brain as a black box was necessary because little was known about the neurophysiology of depression and we have no natural animal model. This is not the case with eating disorders. Many other species spontaneously develop animal anorexias where they stop eating and start moving when starved. Binge eating can be induced in normal weight mice in the lab if they are dieted, refed, stressed and then given access to sugar water. The brain systems that are in charge of eating are highly conserved in mammals so there is no reason not to learn from animal neuroendocrine research.

Neuroendocrine researchers have teased out how ED behaviors are caused by changes in appetite and activity regulators and reward pathways in animals and people(6). This information provides the explanation that patients need for why certain behaviors lead to feeling out of control of binge eating, and why weight loss in people genetically vulnerable to AN can lead to restricting and exercising.

People need to know that only a tiny fraction of the brain is conscious and it cannot take care of deciding most of what we do. Nonconscious brain algorithms initiate most of our eating behavior, but, knowing nothing of that, we automatically assign agency to the conscious part we know.

CBT for eating disorders focuses on the wrong underlying cognitions. An underweight person with AN does not run and restrict because she is vain, but because of same neuroendocrine changes that cause rats, mice (7) and pigs (animals that like humans evolved as opportunistic omnivorous nomadic foragers) to run and restrict. These behaviors would remove them from a food-depleted environment. I believe that these behaviors, as well as binge eating, are adaptations that helped our ancestors from survive famines(3).

Recovery: And treatments based on the conventional CBT assumptions have a dismal success rate. In fact McIntosh and colleagues (4) found that patients in what was supposed to be the control group, given only nutritional counseling and emotional support, did better than those given CBT that assumed anorexia is caused by overvaluing body size.

Our updated CBT for eating disorders explains the powerful compulsions as adaptations to flee famine that borrowed from their body’s future to save their life today. Their body is trying to migrate. It feels so wrong to give up anorexia today because they were selected to persevere through pain and hunger.  I suspect that when they found better lands the band members helped them recover.

1.         C. G. Fairburn, Z. Cooper, R. Shafran, Cognitive behaviour therapy for eating disorders: A “transdiagnostic” theory and treatment. Behaviour research and therapy 41, 509-528 (2003).

2.         P. K. Keel, K. L. Klump, Are eating disorders culture-bound syndromes? Implications for conceptualizing their etiology. Psychological bulletin 129, 747-769 (2003); published online EpubSep (

3.         S. Guisinger, Adapted to flee famine: adding an evolutionary perspective on anorexia nervosa. Psychological review 110, 745-761 (2003); published online EpubOct (

4.         V. V. McIntosh, J. Jordan, F. A. Carter, S. E. Luty, J. M. McKenzie, C. M. Bulik, C. M. Frampton, P. R. Joyce, Three psychotherapies for anorexia nervosa: a randomized, controlled trial. Am J Psychiatry 162, 741-747 (2005); published online EpubApr (

5.         P. G. Zimbardo, in Advances in experimental social psychology, M. P. Zanna, Ed. (Academic Press, San Diego, CA 1999), vol. 31, pp. 345-486.

6.         J. Hebebrand, T. Muller, K. Holtkamp, B. Herpertz-Dahlmann, The role of leptin in anorexia nervosa: clinical implications. Molecular psychiatry 12, 23-35 (2007).

7.         W. F. Epling, W. B. Pierce, Activity based anorexia: A biobehavioral perspective. Int J Eat Disord 5, 475-485 (1988).