Binge eating is one of the major diagnostic criteria for bulimia nervosa and binge eating disorder (DSM-IV, American Psychiatric Association, 1994). During binge attacks, large amounts of food are consumed in a short period of time, while the person affected experiences an acute sense of loss of control over eating. Several models have been put forward to explain the occurrence of binge eating. There is an evidence that psychological stress may have an impact on the maintenance of binge eating in eating disordered individuals (Polivy & Herman, 1993). Another approach follows a conditioning model of binge eating (CBE; Jansen, 1990 and Jansen, 1994). The CBE postulates that through classical conditioning, certain cues such as the sight and smell of food elicit physiological responses that are experienced as craving. If a binge occurs, the link between cues and binge behavior is strengthened and the probability that the next encounter with a set of binge cues triggers another binge attack is increased (Overduin & Jansen, 1995).
The CBE model assumes a central role of the physiological responses (i.e. cephalic phase responses, CPRs) to food cue exposure. CPRs prepare the organism for the ingestion of food through secretion of saliva, insulin, gastric juices, etc. These responses are believed to optimize digestion, absorption and use of ingested nutrients and control food intake (Mattes, 1997 and Rodin, 1985). CPRs can be the direct result of sensory stimulation, but conditioned processes as postulated by the CBE model are also thought to play a major role particularly in the case of eating disordered individuals (Nederkoorn et al., 2000).
Most of the studies investigating CPRs are experimental in nature, where participants are exposed to their favorite binge food while monitoring of physiological and verbal responses ensues. The results of these studies are, however, difficult to interpret as they differ in populations investigated, and measures and monitoring techniques employed. Investigations of non-eating disordered participants tend to show increases in salivation in response to food exposure (Epstein et al., 1996, Franchina and Slank, 1988 and Nederkoorn et al., 2000). Results on peripheral measures of activation, e.g. heart rate, blood pressure and electrodermal activity, are, however, equivocal with one study reporting increases in these parameters (Vögele & Florin, 1997) while others fail to find such effects (Andersen et al., 1992, Overduin and Jansen, 1996 and Sjövall et al., 1990). Reports on CPRs in eating disordered individuals cover almost the whole spectrum of eating disorders including anorexia nervosa, bulimia nervosa, binge eating disorder and obesity, and this heterogenity in populations investigated may in part explain the inconsistent results. No effects of food exposure on serum insulin, free fatty acids and plasma glucose were found in persons with eating binges (Karhunen, Lappalainen, Tammela, Turpeinen, & Uusitupa, 1997). However, increases in salivation in response to food exposure are reported by Klajner et al., 1981, LeGoff and Leichner, 1988 and Tuomisto et al., 1999 while others found no effect or even a decrease in salivation (Bulik et al., 1996 and Karhunen et al., 1997). Results on peripheral psychophysiological measures (i.e. blood pressure, electrodermal activity) are similarly contradictory.
Overall, current findings on CPRs in eating disordered individuals are mixed. Although we previously found evidence for higher arousal in binge eaters when exposed to food (Vögele & Florin, 1997), we did not include measures that could be interpreted more directly in terms of CPRs. The current study was designed in order to replicate and extend our previous findings in a clinical sample of eating disordered individuals. We tested the assumption that CPRs to food cue exposure are more marked in eating disordered individuals than non-eating disordered controls, because emprirical findings indicating decreases in salivary secretion or unspecific changes can be critized on methological grounds.
Specifically, we predicted that women diagnosed with bulimia nervosa would show increased salivation compared to non-eating disordered controls. It was also hypothesized that measures of salivary cortisol would reflect ratings of distress and be higher in the bulimia nervosa group.
