Biological explanation for anorexia nervosa (Holland et al, 1988): Evidence for a faulty gene.

FINDINGS:

• 9 out of 16 (56%) MZ twins were concordant for anorexia
• 1 out of 14 (7%) DZ twins were concordant for anorexia
• In 3 cases where the second twin did not develop anorexia, they did develop other psychiatric disorders and two suffered from minor eating disorders.

CONCLUSIONS:

• The findings show that there is a much higher concordance rate for anorexia in MZ twins compared to DZ twins
• This suggests provide good evidence for genetic factors causing some cases of anorexia, and possibly other psychiatric disorders.
• Since concordance rates for MZ twins were not 100%, faulty genes are unlikely to explain all of the cases. Alternatively, a faulty gene may still be present in the non-sufferers, leaving them vulnerable to developing the disorder, but it does not develop unless some other factor triggers it eg. a stressful life event.

CRITICISMS:

• Environmental factors have been ignored. MZ twins share very similar environments (eg. same home, schools, treated similarly by parents etc) and this could explain the higher concordance compared to DZ twins who may not be treated so similarly. Since 100% concordance was not found, a faulty gene may make the person vulnerable but some kind of environmental factor is needed for the disorder to develop eg. stress.
• Over-simplistic and reductionist (using a simpler biological explanation to explain a complex behaviour). The fact that it is difficult to separate nurture (environmental factors) from nature (genes) shows that the evidence is not strong enough for a purely genetic explanation
• Natural experiments lack experimenter control over variables eg. the IV (genetic relatedness) could not be isolated from other (possibly confounding) uncontrolled variables eg. environmental experiences, socio-economic backgrounds (class factors). This means that internal validity was low (factors other than the IV may have caused anorexia). At best, that data has shown an association between genetic relatedness and anorexia, but does not provide proof.

Psychological explanation for anorexia and bulimia (Behar et al, 2001):Effect of gender identity in causing eating disorders.

AIMS:

according to the behavioural approach, women are under more pressure to be thin because they are constantly presented with "thin ideal" role models in the media. The aim of the present study is to find out if acceptance of the female gender role (ie. the pressure to be thin) is higher in females suffering from eating disorders.

PROCEDURE:

• Natural experiment, as the IV (presence/absence of eating disorder) was not controlled by the investigators.
• 126 participants, of which 63 had been diagnosed with an eating disorder (anorexia or bulimia). The other 63 participants acted as healthy controls.
• Diagnosis was confirmed using standard criteria (DSM-IV) in a clinical interview. Gender identity was measured in a self-report survey (the Bem Sex Role Inventory)

FINDINGS:

• 43% of patients with an eating disorder were classified as feminine gender identity, compared to only 23.8% of controls. These differences were statistically significant.
• More controls (31.7%) than patients (19%) were classified as androgynous (showing a mixture of male and female traits).
• More controls (43%) than patients (27%) were classified as undifferentiated (not fitting into either masculine, feminine or androgynous categories)

CONCLUSIONS:

• The results support the behavioural hypothesis that a feminine gender identity is more frequent in patients with eating disorders, than in healthy controls ie. women face greater pressure than men to conform to the "thin ideal" body image and this places them at greater risk of developing eating disorders.

*A possible implication of the study is the application of androgyny as a defence against eating disorders.

CRITICISMS:

• Ignores the contribution from genetic factors (nature) causing eating disorders, for which there is good evidence.
• A behaviourist explanation, which considers environmental factors only (nurture) is too simplistic and reductionist, as there is equally good evidence from other approaches (eg. biological, cognitive, psychodynamic)
• Natural experiments lack complete control over variables (IV not manipulated), so the causes cannot be determined ie. cannot be sure that gender identities and eating disorders are causally related because the results might be due to uncontrolled variables
• Gender identity was assessed after eating disorders developed, which cannot test the hypothesis that gender identity causes an eating disorder to develop. At best, the results simply suggest an association between gender identity and eating disorders. Ideally gender identity should be measured first in healthy participants, then see if those classified as having feminine gender identity go on to develop eating disorders.