Psychological Factors in the Development of Eating Disorders. The Contributions of Personality and Cognitive-Affective Processes
Chapters 5 and 6 covered two spheres of social influence on the development of eating disorders: society and family/peers. Although other spheres of social influence (e.g., participation in specific sports) may also have an effect, this chapter shifts the focus to psychological factors that contribute to eating disorder etiology. Psychological factors occur within the individual and explain individual differences in susceptibility to developing an eating disorder in response to social influences. Chapter 4 introduced some psychological factors in the context of risk factor research, and Chapter 6 considered psychological factors in the context of psychoanalytic, psychodynamic, and social learning models. This chapter focuses on the roles of personality and cognitive-affective processes in risk for developing and maintaining eating disorders.
Personality
Personality has been defined as a stable way in which individuals perceive, react to, and interact with their environments that is influenced by both biology and experience (Phares, 1988). Temperament has been defined as a biologically based predisposition to experience certain emotional and behavioral responses (Cloninger, Svrakic, & Przybeck, 1993). Temperament is the building block upon which personality develops. While acknowledging that considerable debate surrounds the definitions of temperament and personality (Craik, Hogan, & Wolfe, 1993), this section will review models that have been applied to understanding the development of eating disorders.
Cloninger described four dimensions of temperament: novelty seeking, harm avoidance, reward dependence, and persistence (Cloninger, 1987; Cloninger et al., 1993) that have been examined in relation to eating disorders. Novelty seeking is a tendency to pursue rewards. Harm avoidance is a tendency to avoid punishment by inhibiting behavior. Reward dependence is a tendency to continue rewarded behavior. Persistence is a tendency to continue behavior that is not immediately rewarded and has been linked to ambition, obstinacy, and obsessive-compulsive features.
On self-report measures, AN has been associated with low novelty seeking, high harm avoidance, and high persistence (Amianto, Abbate-Daga, Morando, Sobrero, & Fassino, 2011; Bulik, Sullivan, Weltzin, & Kaye, 1995; Fassino, Abbate-Daga, et al., 2002; Fassino, Amianto, & Abbate-Daga, 2009; Klump, Bulik, et al., 2000), even when individuals with AN are compared with their healthy siblings (Amianto et al., 2011). Bulimia nervosa has been associated with high harm avoidance but high novelty seeking (Bulik et al., 1995; Fassino, Abbate-Daga, et al., 2002; Fassino et al., 2009). In a review of studies examining self-reported differences between patients with eating disorders and healthy controls, Harrison, O’Brien, Lopez, and Treasure (2010) found that eating disorder patients, regardless of diagnosis, were more sensitive than controls to punishment. However, the restricting subtype of AN was characterized by less sensitivity to reward compared to controls, whereas both the binge-eating/purging subtype of AN and BN were characterized by greater sensitivity to reward compared to controls.
These findings map onto the behaviors typical of these disorders. In ANR, low sensitivity to reward could contribute to decreased eating, and high harm avoidance could contribute to behaviors intended to prevent weight gain, such as excess activity. High persistence would enable women with AN to reach and maintain extremely low weights. For ANBP and BN, the combination of high sensitivity to both reward and punishment captures the conflict between engaging in rewarding behavior (increased eating during binges) and trying to avoid punishing weight gain (engaging in inappropriate compensatory behaviors).
Data regarding temperamental differences that may contribute to BED have been less clear-cut. Compared with healthy controls, individuals with BED have higher novelty seeking and higher harm avoidance (Fassino, Leombruni, et al., 2002), consistent with patterns observed in BN. No differences emerged, however, when BED patients were compared with obese controls who did not have BED (Dalle Grave, Calugi, Marchesini, et al., 2013; Fassino, Leombruni, et al., 2002).
Tellegen (1982) described three personality dimensions: positive emotionality, negative emotionality, and constraint that have also been examined in relation to eating disorders. Positive emotionality is the tendency to enjoy and actively engage in work and social interactions. Low positive emotionality is characterized by high levels of introversion or a tendency to keep to oneself. Negative emotionality is the tendency to experience negative mood states (e.g., sadness, anxiety, and anger). Low negative emotionality is characterized by a tendency to be calm and unflappable. Constraint is the tendency to inhibit impulses and show caution, restraint, and conventionalism. Low constraint is characterized by a tendency to act impulsively.
Consistent with the results regarding temperament, AN has been associated with high levels of constraint, low levels of positive emotionality, and high levels of negative emotionality compared with controls (Casper, Hedeker, & McClough, 1992; Pryor & Wiederman, 1996). Women with BN also have low levels of positive emotionality and high levels of negative emotionality (Casper et al., 1992; Peterson et al., 2010; Pryor & Wiederman, 1996), but BN is associated with lower constraint than is seen in AN (Casper et al., 1992; Pryor & Wiederman, 1996). In contrast, no differences have emerged on these basic personality dimensions between women with BED and either normal-weight or obese control participants without BED (Peterson et al., 2010).
Research findings on temperament and personality in patients with eating disorders map onto early clinical descriptions ofpatients suffering from eating disorders. Bruch (1978) characterized patients with AN as having a high level of perfectionism. Her patients tended to be straight-A students with many accomplishments. In addition, they were less likely than their peers to drink alcohol, use illicit substances, or be sexually active—suggesting higher levels of constraint. Prior to the onset of their illness, they tended to hide negative feelings. In contrast to their happy external appearance, Bruch’s patients experienced significant anxiety, sadness, and concern about disappointing others (high negative emotionality). Finally, they tended to isolate themselves from others and avoid social engagement (low positive emotionality). Many of these features can be seen in Emily’s case study.
